Unite 2O2ONE—Respect, Inspire, Serve

The 2021 Vizient Connections Summit Report

2021 Summit Snapshot

Total attendees: 3407

Member attendees: 1843

Power Huddles: 89

Posters: 57

Overall member satisfaction: 91%

(In-person and virtual attendees at live sessions)

The 2021 Vizient® Connections Summit brought together more than 3400 members, suppliers, and others to celebrate their work in caring for and supporting patients and communities during the pandemic, provide insights into how to sustain top performance in quality and accountability, and offer perspectives on potential systemic changes and how organizations should prepare for them.

Held in November 2021, the summit debuted the hybrid format, with both in-person and virtual options. Members pivoted with Vizient to make this format a huge success. Offering 3 ways to participate, the meeting allowed attendees and speakers to join the summit in person in Las Vegas, attend live virtually, or watch asynchronously through on-demand content—with all formats offering continuing education credits. All participants had access to 89 educational sessions and 57 poster presentations. This innovative format proved quite successful, with attendance almost evenly split between in-person and online. Attendees expressed strong approval for content (91% satisfaction), presentation effectiveness (95% satisfaction), and session format (95% satisfaction).

The material represented the best of the best from Vizient members, who submitted 372 abstracts—an impressive number given many competing challenges, priorities, and a national pandemic. This response demonstrates a continued high level of passion and dedication for sharing stories and helping others during these unprecedented times. With attendees gathering in person for the first time in 2 years, Vizient members took time to connect, engage with, and learn from one another, sharing successes and proven strategies for improving patient care—even in the midst of a nonstop pandemic. Hardly a “sit-and-get” gathering, the summit was highly interactive, with ample opportunities for in-person and virtual attendees to engage in dialogue with speakers and each other.

The power huddles and poster sessions spanned 7 focus areas: reliability, systems of care, growth and sustainability, workforce, supply chain and clinical supply integration, strategies for a high-value pharmacy enterprise, and service line optimization. Not surprisingly, many presenters described their organization’s response to the challenges presented by COVID-19 and the innovative ideas and strategies that helped get them through the worst of the pandemic. Much of this work focuses on use of predictive modeling related to capacity, workforce, supply chain, and other pressing issues, often using Vizient databases and analytical tools. The goal is to become more thoughtful and rigorous in anticipating and preparing for the future.

Overarching Theme—Unite 2O2ONE: Respect, Inspire, Serve

The 2021 theme, “Unite 2O2ONE: Respect, Inspire, Serve,” could not have been more appropriate for framing the sessions and discussions.

Unity

Byron Jobe, president and CEO of Vizient, noted during the opening panel that unity acts as a powerful force—one that has been readily apparent over the past 2 years. Health care workers have faced challenges that most never expected to see in their lifetimes. They continue to have a front-row seat to what many people have experienced only through the news. Separated from their families for weeks and months to protect them from COVID-19, health care’s best came together as one to serve patients facing life-altering conditions. They have witnessed heartbreaking loss firsthand, even among their own ranks. By joining forces in the spirit of unity, they have prevented even greater loss and heartache.

Consistent with the unity theme, 2021 marked the first time that the spring Vizient Connections Summit (supply chain) became part of the fall Vizient Connections Summit, providing opportunities for providers and suppliers to work collaboratively. In fact, team-based collaboration served as a consistent theme across Power Huddles and Poster Presentations. Attendees were eager to hear about how best to work together in what has clearly become an interprofessional world.

Panelists and presenters repeatedly discussed how COVID-19 became an unanticipated catalyst for unity. While the demands of the pandemic kept organizational leaders busy, it also provided an unexpected opportunity to reconnect with loved ones—even while juggling workplace demands. During the opening session, Jobe and all 3 panelists (Sanjay Gupta, MD, neurologist and chief medical correspondent for CNN; Scott Gottlieb, MD, former FDA commissioner; and Marc Harrison, MD, president and CEO of Intermountain Healthcare) discussed how they used their time at home to forge tighter bonds with their families. Gottlieb developed some close, trusting friendships completely through virtual interactions. Perhaps fittingly, for Gupta, the summit served as bookends for his public appearances, with the 2019 summit being his last in-person engagement before COVID-19 and the 2021 gathering being his first since the pandemic began.

Respect

The theme of respect served as an underpinning throughout the conference—respect among and between administrative leaders, clinical leaders, physicians, nurses, and frontline staff; respect for patient and families; and respect for the greater community at large. The opening panel touched repeatedly on this theme. Panelists talked about access to health care services being a right, not a privilege. The discussion touched on examples of cooperation between tertiary and quaternary hospitals and their community partners involving telehealth, digital platforms, as well as other support to rural providers and communities with a goal of keeping care in local settings. Harrison noted that Intermountain has made this a priority with critical access hospitals for years, helping make Utah one of only a small number of states that has not closed a rural hospital in the last decade.

Inspire

Inspiration served as another cross-cutting theme. The most inspiring stories came from the many members who discussed how frontline staff continue to come to work every day during a seemingly never-ending pandemic. These heroes have managed against all odds during the most difficult of circumstances. As multiple speakers noted, some sacrificed their lives in doing so—a sacrifice recognized and honored by Jobe and other presenters and attendees.

Serve

Many sessions touched on the theme of service, often focusing on the need to do a better job serving patients and communities by finally addressing longstanding challenges facing the industry. While these and other challenges existed before COVID-19, the pandemic brought about a sense of urgency in addressing them. This list includes enhancing access to mental and behavioral health services; reducing racial, ethnic, and socioeconomic health disparities; addressing social determinants of health; and scaling appropriate use of telemedicine, particularly in rural communities.

The summit covered various ideas and initiatives that begin to address these seemingly intractable issues. Overcoming some challenges will be easier than others. Perhaps none garnered greater attention than health equity, with multiple panelists and presenters discussing the importance of destigmatizing the issue and calling on organizations to proactively identify, measure, and address inequities and social determinants of health. Several of the most popular Power Huddles and Poster Presentations focused on the issue of health equity.

Awards and Recognitions

The summit featured multiple awards and recognitions:

• Sustainer organizations: A panel discussed the Vizient “Sustainers Study,” highlighting 5 organizations that have consistently been high performers on the Vizient Quality and Sustainability Study: Mayo Clinic, Rush University Medical Center, University of Utah Health, The University of Kansas Health System, and NYU Langone Health. More details are available in the following article titled “Sustaining Success During a Once-in-a-Lifetime Pandemic.”
• Supplier awards: Vizient recognized suppliers and distributors in 14 categories of service excellence. These organizations put forth an incredible effort throughout the pandemic, and this recognition honors their commitment under the most trying circumstances.
• Volunteer activity: Vizient brought member volunteers together to prepare 2000 care packages, including inspirational messages and nutritious snacks, that were then donated to Three Square Food Bank, which supports the Las Vegas community.
• Norman Borlaug Humanitarian Award: Vizient honored Ignite, a Chicago-based nonprofit that provides support to youth ages 14–26 who need shelter, primary and mental health care, employment opportunities, and skills to live a healthy and productive life. This award, which includes a $50 000 grant, is named for Norman Borlaug, PhD, an agronomist known for his revolutionary approach to plant breeding.

Finally, Vizient recognized the many presenters who developed catchy, creative, and often humorous titles for their work that played an important role in piquing the interest of participants. This led to high attendance, and hence, more exposure to the innovative ideas and the positive impact these projects had on patient outcomes (see pull-out box for a representative sample).

Creative Title Examples:

• Is There a (Supply Chain) Doctor in the House?
• Thinking Outside the Mask (N95 Alternative)
• More Bang, Less Bother: Efficiently Maximize Quality by Decreasing LOSI
• How Not to Strike a Nerve: Engaging Spine Surgeons in Vendor Reduction
• Getting Better at Getting Better: Intermountain Healthcare Drives Rapid Improvement
• We Deliver Babies, Not Infections
• The Game of Life: Care Across the Continuum
• Three People Walk Into a Clinic: The Patient, the Provider, and the Scheduler

Sustaining Success During a Once-in-a-Lifetime Pandemic

Julie Cerese, PhD, RN, MSM

Group Senior Vice President, Performance Management and National Networks, Vizient

David Levine, MD, FACEP

Group Senior Vice President, Advanced Analytics and Product Management, Vizient

Launched in 2005, the Vizient Quality and Accountability Study identifies top performers quantitatively in terms of quality, safety, efficiency, effectiveness, patient centeredness, and equity. Over the years, Vizient has studied organizational and cultural factors and found that being the best requires 5 key characteristics: a shared sense of purpose; a cooperative leadership style and atmosphere; a results-driven environment that always strives for top performance; centralized and decentralized accountability structures; and a culture of collaboration and respect, with deference to expertise.

In 2019, Vizient launched its “Sustainers Study,” with the goal of understanding the secret sauce of organizations that have achieved top Quality and Accountability performance over a sustained period. Vizient identified 5 organizations that received top-performer designation consistently since 2005 (including at least the past 2 y): Mayo Clinic (12 times), Rush University Medical Center (11), University of Utah Health (10), The University of Kansas Health System (8), and NYU Langone Health (8). Vizient conducted an in-depth evaluation of 3 of these organizations (Rush, Utah, and Kansas) and assessed a comparison group of 3 similar organizations that had not achieved top-performer status in the last 2 years and had appeared on the list only sporadically (but at least once) since 2005.

The Secret to Sustained Success

Providing the best patient experience requires a palpable intensity related to the 5 previously mentioned key areas. Nonsustainer organizations also focus on these areas, but not to the degree or consistency that sustainers do.

People

At top-performing organizations, everyone—leaders, physicians, and staff—has a deep understanding of organizational goals and a high level of intensity and energy around achieving them. Staff members understand they have a critical role in achieving top performance and feel accountable for doing so. Leaders routinely encourage staff to be their best self for patients. The University of Kansas Health System encourages employees to take care of themselves and rely on each other when they have a bad day. The medical staff at sustainer organizations has a remarkably high regard for nursing, with physicians and nurses respecting each other as leaders, colleagues, and team members.

At nonsustaining organizations, these same elements exist in certain pockets, but not with the same intensity or consistency. Accountability for results varies across individuals and teams, as does the level of physician engagement, with many physicians lacking confidence in and alignment with executive leadership. Finally, the degree of partnership between physicians and nurses tends to be inconsistent, with hierarchy and legacy cultures and behaviors remaining in place, thus inhibiting collaboration and joint decision-making.

Board Support

The board has an infectious energy at sustainer organizations. Without apology, board members get highly involved in guiding the organization’s success. They are knowledgeable about quality and embrace their role in creating accountability for strong performance. Sustainers are committed to full and complete transparency with the board, sharing the good, the bad, and the ugly. The board routinely evaluates performance versus established metrics. They understand the actions that affect outcomes and expect closure from leaders whenever a performance issue arises. The board spends as much time on quality as they do on finance-related issues.

At nonsustainer organizations, boards are often kept at bay, with little transparency about performance. Boards are often distracted with other activities such as credentialing. Wide variation exists as to levels of knowledge and how board members see their role with respect to the quality agenda.

Processes

At sustainers, organizational priorities are cascaded to units and departments, creating team and individual accountability. Performance is monitored daily through various management systems and processes, including rounding, huddles, just-in-time coaching, and transparent data. An army of problem solvers exists at the unit and department levels that addresses issues in real time.

Nonsustainers apply these management systems only sporadically. While there may be a handful of problem solvers and individual pockets of high performance, processes, systems, and expectations remain highly variable and inconsistent. There is heavy reliance on the central team with more sporadic team accountability and innovation.

Decision-Making

At sustainer organizations, leaders respect and support organizational decision-making processes, even when individual decisions do not go their way. This approach facilitates buy-in to any decisions. Nurse and physician leaders collaborate in an environment of bidirectional mutual respect when prioritizing opportunities and implementing action plans. For their part, staff members are empowered to make independent decisions in the best interests of the patient. This autonomy often leads to rapid improvements.

At nonsustainer organizations, relationships between physician and nurse leaders—and the accompanying degree of collaboration—remain inconsistent and highly variable. Titles and hierarchy stifle discussion and influence the quality of decision-making related to priorities and patient care.

Consistent Goals and Reliable Metrics

At sustainers, the organization establishes a clear north star guiding principle and remains focused and aligned on achieving and sustaining top performance. Embracing data-driven transparency, leaders and staff are never fully satisfied and always looking to reflect and improve. Performance versus established metrics is reviewed daily, weekly, and monthly.

At nonsustainer organizations, goals are set to avoid penalties rather than to establish a north star. If a north star is chosen, it often changes every 2 or 3 years. Variation exists across physicians and staff in the degree to which they understand and buy in to organizational goals. Often too many disparate goals and metrics exist, with no clear consistency or coherence connecting them. In many cases, benchmarking and risk-adjustment processes become a source of debate and distraction.

Sustainer Case Studies

The 2021 Vizient Connections Summit included sessions that featured leaders from the 3 high-performing organizations evaluated in the “Sustainers Study.”

The University of Kansas Health System (Kansas City, Missouri)

In 1998, The University of Kansas Health System was in dire straits. Part of the state university system, the organization had some of the worst patient satisfaction scores in the country, high staff turnover (33%), declining patient volumes, dwindling cash reserves, and projected losses of $20 million a year. A consultant produced 3 options for the hospital (the only academic medical center in Kansas): close, be sold, or become a public authority. Organizational leaders convinced the state legislature to pass legislation making the system a public authority, severing formal ties with the university. The new organization received $23 million in cash (enough to last roughly 30 d), along with aging, undercapitalized buildings and equipment. Approximately 2200 former employees joined the new organization. Leaders partnered with them to embark on a remarkable turnaround story that has led to sustained top-notch performance, achieving the Vizient top-performer list 8 times since 2005. Success was driven by an intense focus on the 5 critical ingredients identified in the “Sustainers Study.”

• People: The health system unites around the vision of becoming the best in the country in caring, healing, teaching, and discovering. Everyone on campus understands this vision and buys into it. The premise is simple—if every patient has a great experience, the organization will grow and do well financially. Senior leaders do not expect anyone to do anything they are not willing to do. They regularly participate in rounding to understand what patients and staff experience on a firsthand basis. They personally lead major initiatives and meet with area leaders who do not meet targets. Leadership established 7 simple expectations for everyone and created a formula to guide the organization (see accompanying side bar). To strengthen accountability, all 30 000+ employees go through an expert-led customer service training program.
• Board support: The board sees all data on a regular basis, regardless of performance level. The board regularly has in-depth discussions with leadership about performance, with an eye toward celebrating successes and stimulating improvements.
• Processes: The organization is “hard on the process, not the people,” according to Bob Page, President & CEO. Problems are seldom, if ever, about 1 person, but rather about a process that has failed multiple times. The system uses various performance improvement methodologies, as no single approach works best. The goal is to get things done in as few steps as possible (ideally 1).
• Decision-making: The organization uses a dyad model that features physician-administrator partnerships. The system is clinically integrated and governed by a single, common board of directors, with no separate departments or organizations. Employing a true partnership model, physicians serve on both the board and executive team.
• Consistent goals and reliable metrics: Leadership has established a few consistent metrics in a handful of areas—including service, quality, growth, people, and sustainability. While the goal is to be in the top decile in performance, reasonable interim goals are set. Leaders share and discuss performance information at biweekly meetings of 400–500 key individuals. These meetings have used the same agenda and a consistent scorecard for many years, celebrating successes and discussing ways to improve areas of underperformance.

Commit to full transparency—both the good and the bad. Be tough on systems and processes, not people. Learn to celebrate successes, even if you’re just moving from the 5th to the 10th percentile. After celebrating, move to the next task of improving further.

—Bob Page, President & CEO, The University of Kansas Health System

Staff take their cues from senior leaders. If it’s important to us, it’s important to them.

—Tammy Peterman, MS, RN, FAAN, NEA-BC, President of the Kansas City Division, The University of Kansas Health System

As noted, system leaders created a formula to guide the organization, along with 7 simple expectations of all employees. Together, they have been the foundation of the organization’s success.

Seven Simple Expectations

• We are all here for 1 reason: patient care—providing the best care and support to patients and families.
• There can be no bad days in front of patients or families. Staff can have a bad day, but they must do so behind the scenes, where they can rely on fellow team members for support.
• First impressions matter. They always have and always will.
• Everything we do, we do as a team. Team members must be respectful of each other across disciplines and provider types.
• We must become the safest, most efficient organization in the country; safety always comes first. New employees are encouraged to speak up if they see something that could compromise safety.
• Be proud of where we are, but never satisfied with it. Strive to be better teammates and communicators.
• Own everything that’s yours and a little bit more. Make sure that nothing falls through the cracks and make it a habit to do something not in your job description.

Rush University Medical Center (Chicago, Illinois)

A top performer 11 times since 2005, Rush University Medical Center succeeds because of its strong commitment to setting aggressive goals, consistently measuring performance against those goals, and addressing problems as they surface. Once again, the secret sauce lies in the 5 critical ingredients:

• People: Rush’s leadership has established a clear mission—to improve the health of the community it serves. Without a true north to serve as a guiding principle, organizations falter during difficult financial times or due to social pressures. Leaders also establish aggressive, mission-related goals, strategies for achieving those goals, and clear expectations for everyone associated with the organization. These steps make it easy to attract, hire, and retain the right people, because everyone buys into what the organization is trying to achieve.
• Board support: The board fully embraces the mission and focus on quality and outcomes. It regularly receives and reviews fully transparent performance data. The board supports Rush’s efforts to do what is right for the community, even when such activities may come at the expense of other priorities, including financial considerations.
• Processes and decision-making: Rush operates with the premise that everyone is trying to do well. The problem rarely lies with a person or group, but rather with underlying processes. Data shines a light on the problem, and key stakeholders work together to address it. A close collaborative relationship exists between clinical and administrative teams. Neither side leads the other; the 2 groups make decisions as true partners.
• Consistent goals and reliable metrics: Rush establishes clear, aggressive goals on established metrics and consistently measures performance against those goals. The organization regularly celebrates successes while simultaneously addressing identified problems.

Health care is full of good people. But they are being pulled in too many directions. Leaders must provide clarity about what we’re trying to achieve . . . an organization has an entirely different feel when clinical leaders are not aligned with administration. You can tell in about 5 minutes.

—Omar Lateef, President and CEO, Rush University Medical Center

University of Utah Health Hospitals and Clinics (Salt Lake City, Utah)

A top performer 10 times since 2005, the University of Utah Health Hospitals and Clinics similarly focuses on the critical ingredients to sustained success:

• People: The organization demonstrates a clear commitment to employees through compensation and by promoting an appropriate work-life balance to ensure that staff remain healthy and engaged. Leaders communicate and reinforce this commitment, acknowledging the critical role that everyone plays. Leaders emphasize the core values of mutual respect, trust, and teamwork. These values resonate with staff members, who routinely cite their relationships with teammates and coworkers as the reason they come to work each day.
• Board support: Board members are extremely engaged and focused on what is important to the organization. Individual board members have expertise in quality improvement methodologies and regularly query leaders about what they have learned during rounding from frontline staff and patients.
• Processes and decision-making: The organization’s emphasis on people translates into collaborative, mutually respectful, nonhierarchical decision-making. Quality improvement processes such as Lean focus on fixing systemic problems rather than calling out individuals.
• Consistent goals and reliable metrics: The organization’s improvement journey began 14 years ago with the establishment of specific goals related to providing an exceptional patient experience and high-quality care while maintaining financial stability. Clear metrics cover each of these areas, with regular measurement, monitoring, and feedback against these goals.

Never give up a long-term relationship for a short-term gain. Acknowledge what your people do, support them, and lift them up every day.

—Dan Lundergan, MHA, CEO, University of Utah Health Hospitals and Clinics

Sustaining Excellence During a Pandemic

At no time have the 5 critical ingredients been more important than during the pandemic. Sustainer organizations have enjoyed even greater relative outperformance since 2019 than they did before the pandemic. The reason is simple: the cultures and systems in place at these organizations allow them to make decisions and take actions that many others simply cannot, as illustrated by the examples below.

The University of Kansas Health System

When COVID-19 hit, The University of Kansas Health System doubled down on what made it successful. As a result, performance improved in 2020 compared to the year before. The secret was staying true to the formula and the people who make up the organization. In April 2020, senior leaders announced there would be no staff layoffs or furloughs—even though the system was generating only 40% of prior-year revenues. The reasoning was simple—engaged staff are critical to the previously described guiding formula, and a global pandemic was no time to deviate from that formula.

At the end of the day, it’s about having a sustained, consistent culture—it’s what we have built and what has led to our success over the last 20 years.

—Tammy Peterman, MS, RN, FAAN, NEA-BC, President of the Kansas City Division, The University of Kansas Health System

Rush University Medical Center

Because COVID-19 did not hit Chicago immediately, Rush had the opportunity to learn from others. Supported by a culture that discourages egos and values humility, Rush leaders began gathering information from those in early hot spots such as Seattle and New York. Rush created a dedicated triage area and put extra intensive care unit beds and equipment in place before patient zero arrived. Because no one knew what to expect, Rush’s leaders emphasized to employees the need to be open and honest about what they were learning every day. To keep staff engaged, leaders remained highly visible. The board also played a critical role, making it clear that Rush would be led by its values and continue putting patient outcomes first. Facing $10 million in losses a week, the board chair approved the idea of Rush taking safety-net transfer COVID-19 patients from everywhere, including the uninsured and those without documentation. Rush expanded capacity to handle these patients, including installing equipment in areas not traditionally used for patient care.

We didn’t have to do a lot of the things we did when COVID hit. But it’s a part of who we are. We all came together to deliver for the community.

—Omar Lateef, President and CEO, Rush University Medical Center

As the “Sustainers Study” and these case histories illustrate, achieving top performance over a sustained period requires a palpable intensity related to 5 areas: people, board support, processes, decision-making, and consistent goals and reliable metrics. Paying attention to these areas drives sustained excellence under both normal circumstances and during unprecedented challenges like the COVID-19 pandemic.

A New Growth Formula: Relevance, Diversification, and Scalability

Joan Moss, RN, MSN

Managing Principal and Chief Nursing Officer, Sg2, a Vizient company

John Becker

Group Senior Vice President, Strategic Growth Solutions, Vizient

As the United States and the world slowly transition to an endemic phase of COVID-19, hospital and health system leaders finally have time to breathe again, to turn at least some attention away from day-to-day crises and think again about long-term imperatives like growth and capital planning. Senior executives are having strategic conversations again, including discussions about how to restart growth. Surveys of Sg2® members conducted during the November 2021 Sg2 Executive Summit signaled unequivocally that growth was at the top of the agenda for 2022, an unsurprising finding after 2 years of pandemic-induced volume disruptions.

The Realities Uncovered by COVID-19

These executive conversations are not about returning to the pre-COVID-19 way of doing business. The pandemic accelerated underlying trends that can no longer be ignored. First and foremost is the establishment of virtual care as a permanent feature of our health care system, as people worldwide discovered that they do not have to leave home to receive high-quality care. A second related trend concerns the pandemic’s impact on ongoing shifts in sites of care. COVID-19 accelerated the movement of care outside the hospital, as patients are increasingly reluctant to travel long distances to an inpatient facility full of sick patients when they can more easily go to a neighborhood doctor, clinic, or surgery center, or, in some cases, receive care in the comfort of their own home. Third, the pandemic unmasked the incredible burden of behavioral and mental health issues in the United States and our current system’s utter failure to address them effectively. Many who previously felt untouched by such challenges now have an appreciation for the importance of good mental health and the need to easily access services without shame or stigma. Finally, the pandemic has shined a bright light on health equity, as it has intensified long-existing health disparities for racial and ethnic minorities and low-income individuals. One only needs to consider COVID-19 infection, hospitalization, and death rates among minority and low-income populations to see that today’s health care system does not serve all populations equally.

A New Growth Formula

Americans continue to need more health care services, a trend that will accelerate as the population ages. This will provide ample opportunity for organizations to grow. The key question is: Where will future care be delivered? As noted, less will occur at the hospital, including inpatient facilities and emergency departments (EDs). Overall, Sg2 forecasts that adult inpatient volumes in 2029 will still be 1% below pre-pandemic levels of 2019. Similarly, ED patient volumes in 2029 will be 5% below 2019 levels. That is not to say that inpatient facilities and EDs in 2029 will look the same as they did in 2019. Those who need in-hospital or ED care will be much sicker, typically suffering from 1 or more chronic physical and/or mental health conditions. Inpatient case mix indices and length of stay will increase, while many care sites outside the hospital will enjoy relatively robust growth, including home health and hospital-at-home services.

Resources—Available to Vizient/Sg2 members

For more information and details about Sg2 service line-specific utilization forecasts, visit:

1) Sg2 2021 Impact of Change® Forecast: Post-Pandemic Recovery, Rising Acuity and Ambulatory Shifts
https://www.sg2.com/health-care-intelligence-blog/2021/06/sg2-2021-impact-of-change-forecast/ 
2) Sg2 Impact of Change Forecast Predicts Enormous Disruption in Health Care Provider Landscape by 2029
https://newsroom.vizientinc.com/sg2-impact-change-forecast-predicts-enormous-disruption-in-health-care-provider-landscape-by-2029.htm?pressrelease

Growing an organization over the next decade will require hospital and health system leaders to adopt a new formula. As depicted in the following graph, growth potential will depend on an organization’s ability to:

• Establish greater relevance with both consumers and payers.
• Diversify into new, nontraditional areas.
• Scale these offerings to previously unserved consumer segments and geographic markets.

Source: Sg2 concept.

Relevance

Hospitals and health systems have not historically been proactive in identifying and engaging consumers. Some use general, untargeted media campaigns (eg, television, print), while most depend on others to direct patients to them, such as payers (through network provider lists) and physicians (via referrals). Consumers often rely on recommendations from families and friends when choosing providers, although increasingly they also conduct online searches. To grow in today’s environment, hospitals and health systems need to be more aggressive in targeting and establishing greater relevance with both consumers and payers.

Relevance With Consumers

Consumers often need help accessing services. While conventional wisdom suggests that finding primary care is a barrier, many consumers express frustration with specialty care. A recent Sg2 survey of 30 000 consumers found that 43% of those with 1 or more chronic conditions had trouble accessing specialty services. Hospitals need to tap into these and other unmet consumer needs by proactively identifying and supporting consumers who need care. This direct-to-consumer channel can pay huge dividends, but it demands a new, less familiar approach to marketing. To succeed, hospitals must understand how patients look for and choose providers, including how they search the Internet. Supported by Sg2 tools, Vizient member organizations have had success using data mining and online, precision-targeted ads to establish relevance with potential patients and direct them to providers who can meet their needs. The typical yield from such activities can be quite high—for every 1100 targeted patients, 70 become qualified leads and 30 schedule an appointment, according to Sg2 research. Acquiring these patients may require additional investments in easy-to-use online scheduling systems and potentially care navigators who assist patients in finding the right provider and scheduling an appointment.

It’s all about finding, meeting, and activating potential patients through digital means.

Source: Sg2 research, 2022: The Year Ahead: If Everyone Is Heads-Down, Who Is Looking Ahead? (webinar)

Relevance with Payers

Hospital and health system leaders must decide the extent to which they want to enhance their position within payer networks. The right strategy will generally be market specific and depend to some extent on the degree to which payers are using value-based strategies to steer patients to providers. Some organizations remain extremely cautious about value-based contracts and continue to rely on fee-for-service payments. Others are more aggressively embracing risk-based contracting. The right answer will be different for every organization and depend in large part on how well-positioned the organization is to take on varying degrees of risk.

Diversification

Academic medical centers and other large hospitals will compete fiercely to provide tertiary and quaternary inpatient care to the most complex patients. A handful of organizations will likely be successful playing only in this segment, with that success being driven by the ability to differentiate services from the competition.

Most organizations will need to do more than just differentiate themselves in the inpatient arena. They must diversify into new service offerings, primarily lower-acuity care that, as noted, will occur outside the hospital. Success will require new investments in the physical assets of local communities and digital assets that reach into those communities and beyond. The good news is that hospitals and health systems have many opportunities to diversify into “adjacent” services, such as consumer wellness, primary care, direct contracts with employers, digital/virtual health, and commercial enterprises.

Source: Sg2 concept.

For larger organizations, opportunities may exist to generate nonoperating revenues by selling consulting and other services to outsiders and through new business partnerships, such as with venture capital. A few organizations have been so successful with these kinds of activities that nonoperating revenues exceed those from traditional patient care.

Hospital-at-home services can be an especially good diversification opportunity. The home is rapidly replacing the hospital as the “hub” for health care services. To succeed, hospitals and health systems must invest in and deploy technologies that allow patients to be monitored at home. The same approach can bring more complex care to rural areas. For example, Intermountain Healthcare in Utah deploys various digital technologies to bring services to critical access hospitals, keeping jobs in local communities and eliminating the need for patients to travel long distances.

Scalability

The most successful growth stories will continue to be organizations that take their enhanced relevance and diversification efforts to scale, extending them to new customer segments and geographic markets. Building a robust digital platform may offer the best opportunities for achieving meaningful scale, as services delivered digitally need not be bound by traditional geographic markets or existing care sites. Those wishing to scale digital offerings must organize them appropriately. While early-stage efforts can operate as disparate pilot programs, evolving programs will need to be governed and managed at a department level, and fully scaled initiatives will require a chief digital officer with centralized decision-making authority, supported by robust marketing, clinical, operational, and financial functions.

Using the Growth Formula to Address Health Equity

Hospitals and health systems are using data mining and other strategies to establish relevance—not just with high-income, high-margin consumers, but also to better serve patients and communities that have an increased risk of unmet social needs. Members are using the new Vizient Vulnerability Index to identify specific zip codes—and the context around the obstacles that patients in these zip codes face in accessing health care. The Index can also be used to help quantify the direct relationship between those obstacles and patient outcomes personalized to their communities. Often in partnership with community-based organizations, these health care providers are bringing nutritious food, safe housing, transportation, mental and behavioral health care, and other services to high risk, underserved populations, with the goal of reducing health disparities. Examples of these kinds of efforts can be found in the next article Striving to Achieve Health Equity.

Emerging from the pandemic, success will require adoption of a new growth formula focused on establishing greater relevance with consumers and payers, diversifying into nontraditional areas, and scaling new offerings to capture previously unserved consumer segments and geographic markets.

Striving to Achieve Health Equity

Mike Humphrey

National Vice President, AMCs, Sg2; Executive Officer, AMCs, Member Connections, Vizient

Shaifali Ray, MHA

Senior Member Networks Director, Member Connections, Vizient

While the pandemic has devastated many communities, nowhere have the health and economic effects been more pronounced than communities of color. COVID-19 infections, hospitalizations, and deaths disproportionately affected racial and ethnic minorities, who are also more likely to have endured pandemic-related job losses and delays in accessing care for non-COVID-19-related conditions.

These inequities existed long before COVID-19. Significant racial, ethnic, and socioeconomic disparities have persisted for decades, if not centuries—in terms of both health status and access to health care services. The events of the past few years, including not only the pandemic but also the nationwide social unrest, have heightened awareness of these inequities and the systemic practices that underlie them.

A Renewed Commitment to Address Inequities

The good news is that key stakeholders, including health systems, community-based organizations, and federal and state governments around the country have taken notice. Awakened and reinvigorated, they made a renewed commitment to promote health equity. Much of this work focuses on addressing social determinants of health, such as better access to affordable housing, nutritious food, safe places to exercise, and convenient transportation. Hospitals and health systems are not always well-equipped to deal with such disparities, which tend to be driven by societal factors outside their control. Nonetheless, the pandemic created a burning platform to try to make a difference, often through partnerships with government agencies, local nonprofits, and other community-based organizations.

Focusing on health equity and social determinants of health is not new for most hospitals and health systems, which have long had such programs in place—often through specific grant-funded initiatives or at individual sites operating in communities that face challenges. What is new, however, is the movement away from isolated, one-off initiatives to more concerted, coordinated, and comprehensive approaches. Health equity is increasingly seen as a business imperative as health care organizations look to improve quality of care to address disparities in outcomes, remain competitive, and anticipate increased regulatory accountability and use of value-based payment models. In addition, organizations are building a renewed commitment to employees by integrating health equity into larger diversity, equity, inclusion initiatives. For their part, leaders see health equity as central to their organization’s mission, vision, and values, and, reflecting this view, equity is now appearing as a formal pillar in health systems’ strategic plans.

This commitment to advance health equity was a focal point at the 2021 Vizient Connections Summit, as well as several related Vizient events. The summit featured 8 Power Huddles and 4 Poster Presentations focused on the topic, along with others where the topic of equity surfaced in the context of presentations and discussions. During the same week, Vizient hosted 2 additional meetings on health equity (described in the next section). Consistent with the overall summit theme of unity, these meetings brought together diverse stakeholders with a common interest and passion in addressing health equity.

Health Equity Network

A 4-hour meeting of the Vizient Health Equity Network provided an opportunity for leaders who oversee their organization’s health equity strategy to share ideas in 3 distinct areas.

• Data and analytics: Presenters and attendees discussed how to use data and analytics, including the newly developed Vizient Vulnerability Index, to identify community needs for at-risk, underserved patients. Network members, including Cottage Health in southern California, shared how they use data and analytics as a foundation for improving health and wellness for at-risk patients and staff in their hospitals and communities.
• Structure and organization: Discussions focused on organizing and structuring equity initiatives within an organization, including how to create a culture of improvement, measure success, and integrate initiatives with existing diversity, equity, inclusion resources. For example, representatives from Yale New Haven Health System discussed how their past work in health equity informed the structure and design of a new, centralized Office of Health Equity.
• Community partnerships: Presenters from Lifespan Health in Rhode Island highlighted examples of 2 partnerships with community-based organizations that serve specific patient cohorts, including a medical respite program for those experiencing homelessness and a transitions clinic to support individuals recently released from incarceration.

Health Equity Symposium Highlights the Importance of Collaboration

A virtual, 4-hour symposium brought together Vizient Health Equity Network representatives with clinical executives (chief nursing officers, chief medical officers, chief quality officers) and other leaders (eg, advance practice leaders, pharmacy executives) to focus on the national landscape surrounding health equity. Over 80 attendees heard a keynote presentation from Dayna Bowen Matthew, JD, PhD, Dean and Harold H. Greene Professor of Law at George Washington Law School. A nationally recognized leader in public health and civil rights law focusing on racial disparities in health care, Matthew explored the need to work collectively to eliminate health inequities by calling out and addressing systemic practices that contribute to poor health among minority groups. She reviewed structural determinants of health and the role that policy can play in promoting equity. Noting that the country has been down this road before, she highlighted the important role that both federal policies and health care professionals played in ending segregation in the provision of health care services in the 1960s, and called on policymakers, organizations, and individuals to take action to make an impact.

Three Levels of Commitment

Health care leaders must develop a common understanding of health inequities and how to address them through an integrated suite of programs. The depth of commitment will vary by organization and depend on both internal capabilities and local circumstances, including the availability of community-based partners and funding. Vizient often works with members to apply the Sg2 equity readiness framework to determine the appropriate approach.

Level 1: Addressing a Specific Clinical Issue

The first level fits squarely into the traditional role of hospitals and health systems. These initiatives use data and analytics to identify and address specific clinical manifestations of social determinants (such as readmissions or missed appointments), with the solution usually involving increased patient engagement and partnerships with at-risk populations. Programs are often led by the health system and established in collaboration with community-based organizations, such as employers, houses of worship, or local nonprofits. For example, one Health Equity Network participant reviewed its inpatient, emergency department (ED), and prescription data and identified individuals with diabetes or prediabetes who frequently presented to the ED. Supported by an outside grant, the organization partnered with a local YMCA to provide group coaching, nutrition counseling, and exercise classes to these 2 cohorts. The grant allowed for the provision of incentives (eg, gift cards) to encourage engagement, with most participants attending over three-quarters of classes. Overall, the program resulted in reductions in weight, hemoglobin A1c levels, incidence of depression, and ED visits.

A second example comes from UC Davis Health, which worked with a partner to expand access to methadone treatment for opioid overdose patients who present in the ED. UC Davis worked to identify and hire former addicts who now serve as counselors to educate and consult with patients. ED physicians received group training (known as X-waiver training), allowing them to prescribe buprenorphine to patients. The program significantly enhanced access to treatment, with the number of patients receiving buprenorphine prescriptions increasing more than 6-fold (from 123 to 787) over a 3-year period.

Level 2: Aligning to Address an Upstream Social Need

The second level requires hospitals and health systems to further expand outside their wheelhouse by connecting with community partners to address an upstream social need. Unlike level-1 interventions that target specific clinical issues, this approach involves identifying patients with specific needs using widespread screening, sharing data across partners, and relying on significant engagement with community stakeholders. For example, UK HealthCare in Kentucky works as part of a community coalition with Lexington community paramedicine, care navigators, behavioral health and addiction services, and a local, federally qualified, health center to support homeless individuals who frequently utilize the ED for nonacute needs. The program locates housing and shelter options for these patients and provides access to follow-up care. Significant progress had been made 1 year after implementation, including an 8% decline in ambulance runs and a reduction in nonemergent visits. The collaboration has also proven essential during the pandemic, as it has helped manage the spread of COVID-19 among the homeless population in the Lexington community.

Level 3: Becoming an Anchor Institution

At the third level, health system leaders are taking a step back and asking what their organizations can do to improve health equity on a communitywide level by addressing economic, environmental, and other structural issues that drive disparities. Working with public and private partners, these organizations are declaring themselves to be “anchor institutions” committed to promoting equity and economic growth.

One example of this approach comes from Rush University Medical Center. As described in a summit Power Huddle, Rush leaders found that life expectancy in 2 Chicago communities varied by 16 years—even though the communities are located just 2 miles apart. While residents of the city’s Gold Coast generally enjoy long, healthy lives, those living just 2 miles west endure high levels of infant mortality, poverty, crime, and other social ills that too often lead to premature death. Rush leaders decided to be part of the solution by partnering with other hospitals, businesses, universities, governments, and faith-based institutions. These organizational leaders set the audacious goal of decreasing the gap in life expectancy between the 2 neighborhoods in half—to 8 years—by 2030. A second example comes from UMass Memorial Health in Worcester, Massachusetts. During the Vizient Health Equity Network meeting, UMass leaders provided an overview of its program “Anchored in Our Community,” designed to invest in and transform community health in partnership with local organizations.

The COVID-19 pandemic and social unrest have brought renewed attention to health inequities, with key stakeholders around the country making a commitment to address them. Collaborating with others, hospitals and health systems can make a meaningful difference in promoting equity, be it through targeted programs to tackle specific disparities or communitywide initiatives to address broader ones.

Data and Analytics Drive Improvement: Moving Beyond Low-Hanging Fruit

David Levine, MD, FACEP

Group Senior Vice President, Advanced Analytics and Product Management, Vizient

Steve Meurer, PhD, MBA, MHS

Executive Principal, Data Science and Member Insights, Vizient

The global pandemic has touched every aspect of health care, including how hospitals and health systems use data and analytics to drive improvement. Unfortunately, some organizations have been forced to dial back their efforts, with quality improvement (QI) resources being temporarily repurposed to manage pandemic-related challenges. Others continue to move ahead with QI efforts, with several trends and themes emerging in the use of data and analytics.

Continuing Trend: Centralized Approach Still Produces Superior QI Results

The Vizient Quality and Accountability Study highlights the continued importance of taking a centralized approach to data and analytics, in which a system-level department produces standard performance reports, with drill-down capabilities to inform local units and departments. Significant accountability remains at the hospital, department, and unit levels to act on the data through innovations tailored to the local environment. As reported at last year’s summit, 5 health systems accounted for 45% of Vizient top-performing hospitals. All 5 employ this type of approach, while others that operated in a siloed manner saw more uneven performance. Fast forward to 2021 and this trend remains solidly in place, with the latest Vizient Quality and Accountability Study showing that the top 5 systems (which again, all employ a centralized approach) account for 48% of top performers.

The summit included a lunch panel of representatives from several systems that have consistently achieved top performance. The “Sustainers Study” found that these organizations intently focus on 5 areas: people, processes, decision-making, board support, and consistent goals and reliable metrics. Success in each of these areas, in turn, depends on having a strong, centralized data and analytics infrastructure. The centralized department regularly provides timely, detailed, actionable performance information to key stakeholders within the organization. Progress is monitored and measured on a regular basis to create and sustain alignment and accountability for improvement. More information on the “Sustainers Study” can be found in the article “Sustaining Success During a Once-in-a-Lifetime Pandemic.”

Organizations that routinely provide data broadly, including nursing, physicians, executive leaders, but also clinical and non-clinical staff, are much more likely to sustain advancements in improvement.

—2021 Vizient Sustainers Study

New Trend: Successful Organizations Focus on the “O” (Observed) and the “E” (Expected)

Sustainer organizations focus on both parts of the quality measurement equation—the observed or actual number of cases (the numerator) and the expected number of cases (the denominator).

Tackling the “E”

Many organizations understandably focus their initial efforts on the low-hanging fruit—that is, the relatively easy opportunities to boost performance. For quality measurement, these opportunities often lie in better documentation and coding of a patient’s underlying medical conditions at admission. Correct and complete documentation improves the accuracy of an organization’s case mix index, which in turn determines risk-adjusted (or expected) rates for key performance indicators.

Tackling the “E” remains critical for many organizations. Hospitals and health systems new to the Vizient Clinical Data Base often find that their expected mortality and morbidity rates are significantly lower than they should be, given the true clinical condition of their patients (a lower expected rate means a lower denominator and hence, poorer risk-adjusted performance). These organizations should focus on improving documentation and coding integrity to ensure that measurement accurately captures true performance. The 2021 Vizient Connections Summit included multiple examples of organizations that have successfully tackled documentation and coding. For example, the University of Rochester Medical Center used education and technology to improve clinician documentation for stroke patients, resulting in a 14% increase in case mix index. Better documentation and coding also help clinicians identify and address underlying risk factors. For example, MD Anderson Cancer Center in Houston improved screening and documentation of malnutrition in surgical patients, thus allowing proactive intervention before and after procedures.

Effectively tackling the “E” depends on strong, centralized data and analytics capabilities that allow organizations to see how they stack up not only against similar organizations, but also their internal peers. Individual physicians often believe that they treat the sickest patients; without robust data and analytics, for example from the Clinical Data Base, there is no way to determine if that is the case—and no way for physicians to improve.

Tackling the “O” (Observed)

Better documentation and coding are necessary first steps, but that can only take an organization so far. To continue improving and ultimately have a significant impact on patient outcomes, top performers tackle the “O.” The goal is to reduce actual cases of undesirable outcomes, including in-hospital deaths, readmissions, complications, and lengthy stays.

Reducing observed cases requires a multifaceted, systematic approach—again driven by centralized data and analytics. Success entails a willingness to tackle tough, historically intractable problems that have long been on the periphery of health care, including harm reduction, end-of-life care, unnecessary clinical variation, and social determinants of health (SDOH). The pandemic, however, has created a burning platform to begin addressing these issues.

Early Intervention to Reduce Harm

Top performers proactively look for and address issues that lead to patient harm, such as safety errors and hospital-acquired infections. The summit featured multiple Power Huddles and Poster Presentations led by organizations that have had success in intervening early to reduce harm. For example, UC Davis Medical Center created a new nurse practitioner role dedicated to the early identification and treatment of sepsis in adult patients. Using a data-driven predictive model based on vital signs and laboratory values, nurse practitioners facilitated the provision of timely, evidenced-based treatment, leading to reduced mortality. Not surprisingly, Vizient has found that organizations with the lowest hospital-acquired sepsis mortality tend to be top performers.

Improving End-of-Life Care

Hospitals across the country face an unfortunate reality—individuals who are near death often come to the emergency department, many times because family members are hoping for a different outcome. These patients frequently are admitted, in some cases, because no one has explained the reality of the situation or offered other options for care. Some top-performing organizations have begun making a concerted effort to identify patients at the end of life, communicate with and talk to these patients and family members about their options, and, when appropriate, proactively facilitate a transition to caring, patient-centered palliative care. While this approach cannot prevent an unavoidable death, it puts the patient in the right setting and makes the end-of-life process as comfortable and respectful as possible.

The summit included a Power Huddle in which representatives from Froedtert & The Medical College of Wisconsin (Milwaukee) described their efforts to bring expert end-of-life care to the bedside and expand access to palliative and hospice services to all patients hospitalized with serious illness. Using hospice pathways, engaged and empowered clinicians offer this benefit to eligible patients and families. The program led to a 39% increase in use of the hospice benefit in 1 year, resulting in more than 300 additional patients and their families accessing end-of-life services outside the hospital.

Another example comes from Intermountain Healthcare, Salt Lake City, Utah, which used data from the Vizient Clinical Data Base to identify opportunities to reduce observed mortality by developing a robust, end-of-life care process. In a review of 136 patients who died soon after admission, Intermountain found that a significant percentage had been admitted despite being moribund, and typically with no input from hospice care. For almost half the patients, the intent from the beginning was to provide comfort care. Intermountain trained emergency department case managers to identify these patients and work with on-call physician advisors to determine the best course of action, in consultation with the patient and family. Intermountain also charged inpatient care managers with identifying moribund patients admitted at night when the on-call advisor was unavailable. The program has significantly improved the ability to secure compassionate end-of-life care for these patients.

Eliminating Unnecessary Clinical Variation/Enhancing Value

Top performers have increased their focus on enhancing the value of health care, a trend that has become especially important as organizations grapple with workforce and supply shortages. Vizient members increasingly show interest in combining cost and quality information, using advanced analytics to optimize workflows and manage supply costs and clinical outcomes. Chief operating officers and other operational leaders who previously focused only on the Vizient Operational Data Base now want to merge department-level staff, productivity, and cost data from the Vizient Operational Data Base with clinical data from the Vizient Clinical Data Base. The goal is to understand and compare costs, quality, and utilization at a granular level, by patient, physician, disease, and condition. With traditional strategies for margin improvement (such as headcount reductions) no longer available, these organizations are using data and analytics to tackle the difficult task of reducing unnecessary variation in practice patterns—be it streamlining use of expensive tests, drugs, and supplies or eliminating duplicate standing orders. For example, Mayo Clinic in Rochester, Minnesota, found that congestive heart failure patients routinely received 10 tests daily. Using the Vizient Clinical Data Base, Mayo discovered that 80% of peer organizations perform these tests only every other day with similar outcomes. Mayo physicians clarified that they were unaware of the frequency of testing and they quickly agreed to reduce it. Another example comes from MU Health Care in Columbia, Missouri. During a Power Huddle, clinical leaders described how they worked with physicians to reduce unwarranted variations in the treatment of acute myocardial infarction, leading to significant declines in risk-adjusted mortality (57%), readmissions (63%), and length of stay (19%).

Similar analyses have been incorporated to determine the best use of virtual care. Telehealth use exploded early during the pandemic, and then dropped as organizations returned to in-person visits. Some organizations are now trying to figure out the best way to use it in a “normal” world. During the summit’s opening session, Sanjay Gupta, MD, discussed how he continues to use telemedicine for neurology patients with movement disorders, as he finds it helpful to observe patients walking around their homes. Virtual care may also prove more effective for other types of visits where no physical exam is necessary (such as behavioral health) and for medical conditions where care is primarily guided by monitoring of lab values and/or by information in a patient’s medical history. By leveraging advanced data and analytics, pioneering organizations will learn with greater precision when telemedicine may be equal or better, thus enhancing care while simultaneously freeing up scarce capacity to serve those who need in-person care.

Addressing Health Inequities and SDOH

As discussed in greater detail in the companion article on health equity (“Striving to Achieve Health Equity”), significant racial, ethnic, and socioeconomic disparities have long existed in the United States, both in terms of health status and access to health care services. The pandemic brought these disparities to the forefront, and for some, served as a call to action to begin addressing them. These organizations are looking to data and analytics to help, specifically for common metrics that offer a lens through which to view the issue of equity. To that end, Vizient and its members developed the Vizient Vulnerability Index. Released in 2022, this index provides hospitals and health systems with highly targeted data, allowing them to develop customized support to address specific SDOH-related issues affecting individual patients, patient cohorts, and communities.

The summit featured multiple organizations using data and analytics to support SDOH efforts. One such example comes from Stanford Health Care (California), which created a health equity dashboard stratified by demographic characteristics, medical conditions, and SDOH. The dashboard was developed with input from multidisciplinary stakeholders, including patients, caregivers, and clinicians from underrepresented and at-risk populations. Stanford is using the dashboard to design interventions that address disparities. Additional examples can be found in the companion article.

In summary, sustainer organizations focus their data and analytics efforts not only on low-hanging fruit (primarily better coding and documentation), but also on more difficult, intractable challenges facing the industry. These activities seek to reduce observed (or actual) negative outcomes through early interventions to prevent harm, improve end-of-life care, and implement programs to address unwarranted clinical variation and health inequities.

Defying Gravity: What if Not Everything Returns to Normal?

Tom Robertson

Executive Director, Research Institute, Vizient

On January 30, 2020, the World Health Organization declared a “public health emergency of international concern” for only the sixth time in history. As the pandemic turned the health care industry and the entire world upside down, it appeared as if things might never be the same again. With no prior frame of reference for a global health crisis of such speed and magnitude, the Vizient Research Institute undertook a study that posed the question: What if not everything returns to normal?

Our study began with careful observation and extensive interviews with CEOs, clinical leaders, and senior strategists from member health systems, as well as national stakeholders ranging from political policy advisors to biodefense experts. We compiled a list of what we termed big audacious changes (BACs) based on 3 things: (1) changes already underway as a result of the pandemic; (2) a national survey of commercially insured patients that assessed their confidence and preferences; and (3) a series of hypothetical “what if” scenarios.

These 11 BACs were not predictions, they were postulates intended to stimulate critical thinking:

• Patient reticence to engage with the health care system
• Demand elasticity for elective procedures
• Reduction in number of ancillary services per visit with the shift to virtual care
• Seasonal and pandemic-related fluctuations in surgical volume
• Narrowing of the gap between surgical and medical prices
• Equalization of public and private insurance rates through regulation
• Emergence of a federal health plan
• Prospective determination of providers’ annual revenue based on the prior year’s fee-for-service experience
• Intrasystem clinical program consolidation
• Intensification of primary care practice acuity
• Disruption of traditional leadership succession planning

Abbreviation: BACs, big audacious changes.

Survey Findings

Survey respondents ranked the 4 supply and demand BACs as by far the most likely to occur, giving them 4 of the top 5 spots and a 41% higher likelihood rating on average than the 4 systemic BACs. Reflecting the strong sentiment expressed by senior executives during our spring interviews, clinical consolidation was the second-most likely BAC. The 4 systemic BACs were rated least likely by the members, with all 4 falling below the neutral midpoint of the scale. Although they were judged less likely to occur, they were seen by members as being far more disruptive if they were to occur, with disruptiveness scores averaging 33% higher than those for supply and demand BACs. The relationship between expected likelihood and relative disruptiveness is shown below in Figure 2.

Preparedness to adapt to various potential changes was assessed across 7 organizational readiness domains:

• Financial (operating margin, liquidity, access to capital)
• Organization (operational decision-making structure)
• Workforce (head count, skills/experience, development capability)
• Culture (behaviors, relationships, teamwork)
• Outside relationships (community organizations, other providers)
• Physical assets (locations, size and condition of buildings, equipment)
• Technology (existing capability, speed to adapt)

Survey responses show that Vizient members feel unprepared, especially financially, for the 4 supply and demand BACs. Financial preparedness was rated below every other readiness domain for all 4; most troublesome were demand elasticity and patient reticence. The level of concern that there would be fewer ancillary services per ambulatory encounter was almost identical to concern about cyclical surgical volume; both BACs were well below the neutral midpoint of the preparedness scale. After financial preparedness, workforce alignment was seen as the biggest organizational challenge across the supply and demand BACs.

Members feel even less prepared for the systemic BACs, for which anticipated disruption is considerably higher than for supply and demand changes. Financial preparedness again trails all other readiness domains across all systemic BACs and in fact, is even lower than for the supply and demand BACs. Financial readiness was lowest for global spending budgets and all-payer rate regulation, followed by price compression and a federal health plan. Across systemic BACs, organizational decision-making, workforce, and culture show the largest readiness gaps after financial preparedness.

Figure 3 shows that health systems are more prepared for what is immediately ahead than for what may be farther down the road. The bars in each graph represent the preparedness scores for each readiness domain across supply and demand BACs (left side) and systemic BACs (right side). The lines across the top of each graph display member scores for disruptiveness. The “disruptiveness gap—the space between disruptiveness and preparedness scores—is considerably larger for systemic BACs than for supply and demand BACs.

Clinical Consolidation

With provider attention riveted on crisis management at the height of the pandemic, there was no time for opportunism. The emergence of multiple vaccines, developed at a breathtaking pace, may now allow health systems to focus on some of the strategic opportunities that drew attention early in the crisis. Delivering on the promise of intrasystem clinical consolidation, with its potential to simultaneously improve both quality and efficiency, is at the top of that list.

The most compelling reason for a health system to engage in clinical program consolidation is proficiency—volume matters. Minimum volume thresholds for clinical proficiency have been published in peer-reviewed journals. Despite the evidence linking volume to outcomes, however, our study found a significant percentage of surgical cases in large metropolitan markets occurring in programs with volumes below published proficiency thresholds. We excluded rural markets where an argument can be made that low-volume programs are better than no surgical availability at all. Our study found between 30% and 40% of cancer-related lung resections occurring in programs with volumes below proficiency thresholds in 12 large urban markets, where high-volume alternatives are readily available. For mitral valve replacement, nearly half of the health systems studied had surgical programs operating below published proficiency thresholds, and in every case, there was an alternative within the same system where volume was above the threshold. Only 8 of 35 health systems studied had consolidated the high-risk procedures at a single location.

The Shift to Virtual Encounters

Patients have embraced the availability of virtual alternatives compared to traditional in-person visits, but some providers have expressed reluctance to fully commit to virtual encounters due to reimbursement rates offered by commercial insurers. This is an instance where providers should lead—rather than follow—payers. In much the same way that we would not persist in using an out-of-date treatment method based solely on reimbursement, a strong argument can be made to base the pace of the shift to virtual care on patient welfare; payer policies will eventually catch up.

A concern raised early in the study, that a conversion to virtual visits would significantly reduce the use of ancillary services, turned out not to be an issue after all. An examination of data across a large number of Vizient members indicates that one of the postulates—fewer ancillary services per ambulatory encounter—did not come to pass. While fewer ancillary services occurred in association with virtual encounters, the average number of ancillary services per in-person visit increased by approximately 10% after the expansion of virtual visits—from approximately 95 per 100 visits in 2019 to 105 per 100 visits in summer 2020. The bifurcation of use rates suggests a differentiation in acuity between virtual and in-person visits. In other words, providers appear to be doing an effective job of triaging patients, handling lower-acuity visits virtually while conducting in-person encounters for patients more likely to need additional services. The weighted average number of ancillary services per 100 total evaluation and management encounters (virtual and in-person combined) trailed the 2019 baseline in the months immediately following the spring COVID-19 surge but had nearly matched it by October 2020. These data suggest that health systems consider expanding the use of a “virtual front door,” perhaps even triaging patients for in-person visits based in part on whether they are likely to need ancillary services.

Primary Care Practices

The rapid acceleration of the shift to virtual encounters that occurred during the pandemic promises to influence virtually every physician specialty but has the potential to have a profound impact on primary care. Our study found that primary care practices, particularly first movers, can improve care coordination and patient experience for the complex/chronic cohort and increase revenue and operating margin by intensifying the overall patient acuity for the practice.

The early-mover advantage, if there is one, comes from the fact that practice intensification is predicated on attracting and maintaining relationships with complex/chronic patients—a cohort that is not unlimited. In contrast to a more traditional primary care practice, in which physician panels are composed of large numbers of asymptomatic, largely healthy individuals with relatively few high-acuity complex/chronic patients scheduled around them, the intensified practice would focus physician attention on a smaller number of higher-acuity patients. The practice would retain the low-acuity/healthy cohort, but those patients would most often interact with advanced practitioners. The need to attract enough complex/chronic patients to make the model work while providing the level of service required to keep them loyal suggests the potential for first-mover advantages.

A natural path toward practice intensification within multispecialty groups is specialist-to-primary care referrals. Rather than being viewed as a funnel to increase specialty volume, primary care practices would be partners in longitudinal care plans, creating incremental specialist capacity to see new patients in the process. Larger primary care practices might consider an incremental rather than a wholesale approach to intensification, adding advanced practitioners slowly while monitoring the rate of complex and chronic backfill in physician panels. In any case, it is essential to resist the tendency to fill advanced practitioners’ time with support activities; the economic model only works if the advanced practitioners are seeing patients.

Conclusion

The short-term action plans suggested by this study can help hospitals and health systems position themselves for success as they emerge from the all-encompassing demands of the pandemic. As attention returns to strategy, there is a need for aspirational thinking. Disruption on the scale of a pandemic is likely—hopefully—a once-in-a-lifetime occurrence. There will be a national craving for normalcy that will not be limited to health care. Yet there will still be room to aspire…and to hope that some of the lessons learned from the pandemic will endure:

• That health systems will perform like systems, finally delivering the right care in the right place at the right time.
• That collaboration between providers will edge out at least some of the competition.
• That the provision of health care will come to be seen as a common good and not just another business.

Our study began by asking a simple question: What if not everything returns to normal? It concluded with a more provocative question: What if not everything should?

Power Huddles

Suicide Reduction Is a Health System Obligation

Marilouise Venditti, MD; Inua Momodu, MD, MPH, MBA, CPE; Julie Drew, LCSW, MPA; Lenore Hildebrandt, MEd

AtlantiCare

Background. AtlantiCare is committed to being a suicide safer health system aligned with the vision to make a difference in health and healing, 1 person at a time, through caring and trusting relationships. Transformation toward this aspirational endeavor began with senior leadership recognizing the need for a leadership-driven, safety-oriented culture committed to reducing suicide and suicidal behavior for individuals to whom we provide care. The impetus and the imperative to act was initiated in 2014 soon after our Patient Safety Committee reviewed 7 suicide-related serious occurrences and found zero deviation from standard of care and policies. How can we accept these outcomes even if the care was technically contemporary and excellent? Similarly, the suicide rate across southern New Jersey was increasing significantly and continues to increase in our primary service areas, throughout the state, and nationally. Suicide rates have been increasing over time and currently vary by geography and scope per 100 000: United States (13.9); NJ (8.0); and our primary service area counties (10.8), according to the most recent New Jersey Department of Health reporting period. The suicide mortality rate in southern New Jersey is actually rising faster than the national average.^1,2 While New Jersey’s suicide rate favorably ranks 49th among other states in 2019, 795 individuals died by suicide, 1 person dies by suicide every 11 hours, and suicide is the second-leading cause of death among youth in the state.³ Suicide continues to be a complex issue that warrants attention as a health care priority, especially as we navigate through the COVID-19 pandemic and opioid epidemic and their challenging effects. Our commitment and obligation to each person for whom we care: There is hope. There is help.

Intervention Detail. A suicide prevention charter and work plan aligned with the Zero Suicide transformational framework was defined. Interprofessional teams from across care settings identified evidence-based practices, collaboratively designed and deployed protocols, established staff education, and tracked data to improve care and outcomes for individuals wherever they receive care at AtlantiCare. The 7 key components of the Zero Suicide framework were adopted, including Lead, Train, Identify, Engage, Treat, Transition, and Improve. A distinct functional structure ensures communication and ownership across the organization. At the core is the Suicide Prevention Steering Committee, led by our chief medical officer. Subcommittees aligned with the components propel the work with broad stakeholder participation. Information technology and other key support areas are essential. Early on, global-, national-, and state-based research, as well as external educational opportunities, were important to establish our subject matter knowledge and equip ourselves and staff with evidence-based practices and tools. The elements of Lead, Train, and Identify were foundational to set the course and direction.

AtlantiCare-wide work plan activities include:

Lead:

• Participated in the Zero Suicide Academy, hosted by the National Action Alliance for Suicide Prevention.
• Conducted the Zero Suicide Workforce Survey, with over 2600 staff responses, to assess knowledge, practices, and confidence in providing suicide care.
• Completed the Zero Suicide Organizational Self-Study to assess current practices.

Train:

• Educated 90 staff on Cognitive Behavior Therapy for Suicide Prevention through the Beck Institute.
• Provided Zero Suicide education to 100 leaders (led by national experts).

Identify:

• Designed a suicide prevention protocol (including best-practice tools) relative to the individual served, risk level, and care setting.
• Adapted multiple electronic health records.
• Implemented standardized screening and assessment tools, including the Patient Health Questionnaire and Columbia Suicide Severity Rating Scale, while continually refining use of these tools.

Outcomes and Impact. The elements of Engage, Treat, Transition, and Improve were applied with a focus on individuals served, service and treatment providers, continuity of care, and ultimately the outcome and impact derived.

AtlantiCare-wide work plan activities and achievements include:

Engage:

• Educated all staff on suicide risk awareness and how to feel comfortable talking about suicide.
• Designed initiative to promote access to care across multiple clinical settings.
• Improved access to care with frequent contacts to promote engagement while collaborating with involved providers to meet individual needs.

Treat:

• Deployed standardized tools and approaches, including collaborative safety planning, Counseling on Access to Lethal Means, and Cognitive Behavior Therapy for Suicide Prevention.

Transition:

• Designed and implemented a Suicide Prevention Program, including dedicated suicide prevention staff to engage and support clients at risk for suicide across the continuum of care, in collaboration with the respective treatment providers.
• Provided services to 272 unduplicated clients through the Suicide Prevention Program since 2017.

Improve:

• Established effective data collection, management, analysis, and reporting through the Suicide Leadership Team and supported by the AtlantiCare information technology infrastructure and resources.
• Implemented and continually refine an e-scorecard to aggregate and trend suicide prevention protocol data across multiple electronic health record platforms to establish fidelity to model while accessing actionable information.
• Participated in a technical assistance consortium to develop and share best practices resulting in an impact story—“Safer Transitions”—being posted on the Zero Suicide website.
• Screened over 400 000 patients for suicide risk across key access care points (2018-2020); assessed over 16 000 patients at risk for suicide, of which approximately 4200 were at high risk.
• Significantly reduced the number of known deaths by suicide since full implementation of the Suicide Prevention Initiative implementation in 2016, with 14 quarters of zero deaths.

1. Stone DM, Jones CM, Mack KA. Changes in suicide rates - United States, 2018-2019. MMWR Morb Mortal Wkly Rep. 2021;70:261–268.

2. New Jersey Department of Health. New Jersey Violent Death Reporting System. Health Indicator Report of Suicide by County. Accessed April 2021. https://www.nj.gov/health/chs/njvdrs/.

3. Suicide Facts and Figures by State. New Jersey 2019. American Foundation for Suicide Prevention. Accessed April 2021. https://www.afsp.org/StateFacts/.htm.

Pharmacy-Led Medication Quality and Safety Programs

EunJi Ko, PharmD; Catherine Ulbricht, PharmD, MBA, CPPS

Brigham and Women’s Hospital

Background. The collaboration between the Brigham and Women’s Department of Quality and Safety and the Weiner Center for Preoperative Evaluation started in 2019, with a 6-month pilot program focused on medication reconciliation in orthopedic patients. The pilot proved to be effective as reconciliations increased in accuracy after pharmacist intervention. In addition, the collaboration achieved a second goal of reducing patient visit times, which allowed a more efficient allocation of resources. Following the pilot’s promising results, the program took on preoperative antithrombotic management. The initiative established a multitude of goals. The first was to reduce nurse practitioner (NP) workload, which would reduce visit durations and improve necessary patient access. Second, the program aimed to decrease surgery delays and/or cancelations using guideline-based recommendations. Lastly, the program sought to increase Brigham and Women’s Hospital (BWH) Anticoagulation Management Service (AMS) referrals for patients who were seen by BWH or Brigham and Women’s Faulkner prescribers.

Intervention Detail. After identifying eligible preoperative patients, pharmacists and pharmacy interns conducted electronic health record reviews to gather information essential to antithrombotic medication management. The information collected includes procedure and anesthesia type, renal function, prescribing physician, antithrombotic medication indication, and dose. Pharmacists and pharmacy interns contact the dispensing pharmacy and the patient for clarification when needed. Recommendations to continue or hold medications were directed by either the established preoperative plan or the Weiner Center for Preoperative Evaluation guideline for perioperative management of antithrombotic agents. Pharmacists communicate with surgeons, prescribers, and preoperative clinical staff about antithrombotic management plans using an electronic health record tool. Pharmacists also recommend continued follow-up through BWH AMS.

Outcomes and Impact. From a patient perspective, this program ensures proper adjustments of antithrombotic regimens prior to procedures to increase patient safety and reduce surgery delays. If antithrombotic agents are not held for the appropriate amount of time prior to an operation, the patient is placed at a higher risk for life-threatening bleeding during the procedure. If this risk is too high, the operation must be delayed to allow time for proper antithrombotic management, which can lead to scheduling issues and suboptimal patient outcomes. Alternatively, if the agent is held for an excessive amount of time or in situations where the risk outweighs the benefits, the patient is placed at a higher risk for formation of a thrombus, which may lead to life-threatening events. From August 2019 to February 2021, we ensured both the safety of our patients and a seamless coordination of care for 740 patients. This service also allowed us to identify patients who may benefit from AMS and send a referral to their provider. From a systems perspective, this program improves efficiency and reduces cost. Previously, NPs would spend an average of 15 minutes discussing antithrombotic adjustments with each patient. This saves time for both the NP and the patient, thus eliminating unnecessary visits and increasing additional patient access. With a weekly average of 25 patient visits, this change helped BWH save approximately $537.50 weekly and achieve a 204% return on investment during the 18-month pilot.

Leading Efforts on Equity-Informed High Reliability

Karen Fiumara, PharmD, BCPS, CPPS; Andrew S. Resnick, MD, MBA

Brigham and Women’s Hospital

Background. The Patient Safety and Risk Management Team at Brigham and Women’s Hospital (BWH) has led efforts on high reliability, resiliency, and safety culture since 2014, achieving many milestones on our journey. Integrating principles embraced by the commercial airline industry for years, we now have a coordinated approach to analyzing events, tracking risk/quality/safety issues, and staying current with capture of risk inputs throughout the organization. BWH leverages its equity-informed, high reliability framework to minimize the risk of error. The core reliability management system principles of effectiveness and resiliency are leveraged in all high-reliability case analysis and improvement work:

• Effective: If humans operate the system as desired, the system manages the risk.
• Effective and resilient: The system is resilient in the face of human error/choice.

We work to embed these principles throughout BWH as part of our safety culture:

• Implementation of systems to proactively anticipate risk while building resiliency against failure.
• Optimization of human performance through high-fidelity education, training, and minimization of distractions to maximize cognitive capacity.
• Management of our employees’ behaviors to help them make good choices.

Intervention Detail. Collaborative case reviews (CCRs), a more comprehensive/inclusive case analysis process, have replaced root cause analyses. A CCR is a sociotechnical process that identifies and categorizes contributing factors to risk and analyzes these factors to suggest and test modifications for both effectiveness and resiliency. Standardized taxonomies apply to risk, systems, human behavior, and performance. BWH has started to identify methods to manage human performance factors (knowledge, skills, abilities, cognition, and experience) differently than human behaviors. Our providers are using standardized taxonomies and reporting methods and have been integrating these into Quality Assurance Risk Management reports for The Board of Registry in Medicine since 2018.

Outcomes and Impact. Using these successful methods, case analysis reviews increased from 23 in 2017 to over 200 from 2018 to 2020, resulting in nearly 1000 action items. In 2019, more than 600 colleagues participated in CCRs, and 65% of them work on the front lines in nonmanagerial roles. All these efforts help instill our high-reliability principles across BWH. Since 2014, over 3560 BWH leaders have participated in hands-on training in high reliability and just culture methods. More than 6500 colleagues have been exposed to these concepts through multiple mechanisms, including presentations to staff, the Safety Matters campaign, and quality rounds. The Patient Safety, Risk Management, and Quality teams have formed an integrated, cohesive, and proactive approach to analyzing risk. We meet regularly to collaborate and update the risk register.

Redesign of Quality Focused on People, Process, and Patient Care

Karrie N. Moses, MSN, RN, NE-BC; Ross M. Cleveland, III, MBA

Carilion Clinic

Background. In summer 2017, Carilion Children’s experienced a patient death due to pediatric sepsis. However, that is not the whole story. Due to an outdated quality structure, there was a delay of several days before implementing stop gap measures and identifying a long-term solution for preventing patient harm. Therefore, we started our journey to zero harm by completely redesigning our quality team and structure that aimed to prevent any future patient harm. Prior to this event, we did not benchmark our quality data with other children’s hospitals and were measuring success only against ourselves. We based our restructure on a foundation of accountability by benchmarking with other children’s hospitals utilizing data from the Vizient Clinical Data Base and practice management recommendations from Children’s Hospitals’ Solutions for Patient Safety. We implemented a continuous performance management structure by forming a weekly interdisciplinary quality team led by the chair of pediatrics and unit leaders, with each department reporting to the chair. We utilized escalating huddles daily to address customer needs in real time and designated a quality unit director for each department without adding additional employees/full-time equivalents. We collaborated with clinical advancement and patient safety to align our goals with the systemwide quality plan. We employed data transparency monthly at the bedside staff level and aligned our quality goals with annual performance management for all staff members. Most importantly, we shifted from a fear of error or near-miss reporting to a just culture by rewarding reporting for continued sustainability. The following summary showcases how aligning goals, leadership accountability, and a shift in error reporting can prevent patient harm for other institutions.

Intervention Detail. After recognizing the issues within our process and reporting structure following the sentinel event, the integration of distinct tools and approaches became warranted. Implementation of the interdisciplinary quality huddle incorporated a safety list to help prioritize projects based on the level of impact the project had on the care being provided throughout our children’s hospital. The integration of safety lists allows for shared awareness on a variety of issues. Furthermore, the list points out certain items that might need more attention and effort with collaboration from all stakeholders to solve the problem and prevent reoccurrence. Once the interdisciplinary huddles were implemented, each unit within our children’s hospital added escalating huddles into their daily rounds to ensure a more pinpointed approach in addressing concerns to the appropriate unit director. Integration of the following leadership method ensured that operational leaders were made aware of problems impacting the frontline staff and offered direction on resolutions for any concerns that had been escalated. The escalating huddles further enforced the application of Ask, Request, Concern, Chain, a culture tool that encourages staff to voice any concerns without the fear of reprimand or discipline due to reporting an error. As the process evolved, senior leadership and frontline staff came to accept the just culture philosophy that rewards those who report all safety events throughout our children’s hospital. The Ask, Request, Concern, Chain tool and process translated to a more sophisticated reporting mechanism known as SafeWatch, to document all near misses and great catches that staff are escalating and documenting as we strive to become the safest children’s hospital.

Outcomes and Impact. We implemented an inpatient-specific pediatric sepsis screening tool and alert that resulted in a decreased sepsis length of stay index. According to the Vizient Clinical Data Base, our sepsis length of stay index decreased from 2.13 in Q2 of 2018 to 0.89 in Q2 of 2021, with a low of 0.54 post-implementation. According to the Vizient Pediatric Quality and Accountability Study, we ranked 9th out of 85 participating Vizient hospitals for 2020 in the efficiency category, based on 2021 methodology. Additionally, according to Vizient data, we decreased our mortality index trend to less than 1.0 from Q1 of 2018 to Q2 of 2021 for non-neonates. We improved employee engagement scores related to quality, rising above the system goal. In 2020 and 2021, we focused our efforts on medication safety and promoting a just culture environment. We achieved our stretch goal by reporting 152 medication near misses over a 12-month span. Our reporting efforts improved our compliance in medication pump guardrail utilization by 8% for Alaris pumps and 10% for Medfusion pumps. Additionally, we won our hospital’s “Great Catch” award for identifying pediatric/neonatal medications that were missing from the medication pump libraries, resulting in medication pump upgrades to add those medications. We decreased adverse drug events from 16 in 2019 to 4 in 2020 and 2021 with zero serious safety injuries. In 2021, we shifted our focus to central line-associated blood stream infections by implementing both insertion and maintenance bundles beginning in October 2020. Since bundle implementation, we have had 1 central line-associated blood stream infection. Our work is focused on our employees and our patients as we strive for zero harm throughout our children’s hospital, and it is a journey founded upon the framework of a quality structure based on a just culture environment and increased accountability.

Communicating With COVID-19: A Transitions of Care Model

Elizabeth Anderson, RN, BSEE, Lean Six Sigma Green Belt, Duke University Health System; Lanie Dreibelbis, BSISE, Duke University Health System; Kristian Knutsen, BSISE, Duke University Health System; Katie Flanagan, MSW, LCSW, ACM, Duke University Health System; Sharon Farrow, MBA, BSN, CCM, Duke University Hospital; Jen Massengill, MSN, RN, CNML, Duke University Health System; Elizabeth Hankollari, MD, Duke University Hospital; David Gallagher, MD, Duke University Health System

Duke University Health System

Background. Duke University Hospital was implementing a series of transitions of care interventions focused on reducing readmissions in general medicine patient populations when the COVID-19 pandemic hit. As a result of the pandemic, hospitals across the globe experienced an increased need to optimize patient flow to improve bed availability and throughput. Due to this urgent need, the project team quickly pivoted and focused interventions on the COVID-19 inpatient population. The project team learned that these patients have many unique postdischarge needs with unprecedented medical complexity. After completing data reviews and gathering patient feedback to guide the interventions, the team focused on standardizing communication and transitions of care for these patients. As COVID-19 is a novel disease, we found significant challenges to traditional communication and discharge pathways. Discharging actively infectious patients presents additional barriers, including care access, routine labs/infusions, dialysis options, and caregiver strain. The project was championed by senior leaders and key stakeholders who supported the work across hospital medicine, nursing, case management, primary care, and population health. The project was guided by project managers using iterative Lean Six Sigma continuous improvement processes. Using data gleaned from the target population and associated care teams, the multidisciplinary group identified 3 major areas of focus: communication, discharge planning, and inpatient to outpatient handoff processes. Focused interventions were implemented in September 2020 and the continuous improvement cycle was used to check and adjust process measures with feedback from frontline staff. This work is meaningful because it serves to improve discharge pathways and communication for current or recovering COVID-19 patients and it can easily be translated and applied to other patient populations. Significant impact was also noted as targeted interventions were developed for those in our cohort with an actively infectious disease—a previously unexplored area without established standards of care.

Intervention Detail. The multidisciplinary team conducted a value stream analysis and interviewed various team members across the continuum of care, mapping patients’ journeys and identifying care gaps and process steps that lacked standard work. Vizient benchmark analyses were conducted to glean opportunities for improvement regarding length of stay, unplanned readmission rates, and follow-up appointment scheduling rates. After analyzing pertinent data, the significant gaps in care identified were communication, planning for transitions at discharge, and the handoff from inpatient to outpatient care teams. Hospital leadership supported adding a new care transitions manager position to help the team implement the following key interventions:

• Conduct daily multidisciplinary huddles with providers, nurses, pharmacy, utilization management, and case management to discuss/develop potential solutions for discharge barriers and readmission risk factors.
• Engage patients to discuss discharge readiness and review educational materials.
• Assist with patient portal enrollment to support access to important health information, electronic educational materials, and appointments.
• Interview readmitted patients to discuss root cause of readmission and identify preventable gaps in discharge planning.
• Complete a discharge checklist to facilitate a safe and efficient discharge.
• Ensure patient has a follow-up appointment scheduled prior to discharge.
• Utilize readmission risk score (incorporated into electronic medical record) to guide interventions, such as discharge medication reconciliation by pharmacy.
• Refer patients to outpatient systems such as population health resources like DukeWELL (conduct expedited follow-up calls and outpatient wraparound services), project HOPE (skilled nursing facility transition team), and Duke Health System Resource Center (follow-up mechanism for continued support of discharge plan).
• Improve handoffs by ensuring consistent discharge summary routing to outpatient providers using electronic reminders within the electronic health record to the discharging provider.
• Streamline discharge summary formatting so that the most relevant information is highlighted for outpatient providers.
• Increase follow-up care pathways by promoting telehealth use and engaging the complex care primary care nurse.

Outcomes and Impact. While the project is still ongoing, the team has noted the following trends (percentages represent relative changes compared to our overall general medicine hospital population):

• An 18% improvement in average length of stay (a 1.5-d reduction).
• A 16% improvement in the 30-day readmission rate through the emergency department.
• A 7% increase in patient enrollment for access to their electronic health record and patient portal information, such as follow-up appointments or education.
• A 51% increase in referrals to our DukeWELL population health care managers.
• A 22% improvement in discharge summary routing to the primary care physician.
• An improvement to 90th percentile performance on the Hospital Consumer Assessment of Healthcare Providers and Systems discharge information section.

In-depth chart reviews and patient readmission interviews were performed and results shared with project team members to inform them of necessary workflow and process changes. Provider-level data was also distributed, allowing physician leaders to monitor individual performance measures (eg, discharge summary routing). In addition to quantitative results, there was positive qualitative feedback from the care team. The daily multidisciplinary huddle enhanced teamwork and communication while also focusing the team on the planned discharge date and key milestones. A frontline nurse shared this feedback: “I try to attend the discharge huddles every day; they give me the big picture for the patients’ care. After learning of the plan, there is a chance to express concerns with the physicians, as well as the case management team. I will continue to try and go to these as frequently as possible.” Moving forward, outcome data will continue to be monitored to fine-tune interventions through iterative process improvement cycles. Future collaborations are currently in the design phase to enhance education modules for caregivers and integrate a postdischarge clinic for the COVID-19 population. Expansion of this pilot to other high-risk populations with complex needs will also be explored as data is collected and analyzed.

DUHS Development of CARE Hub Supports COVID-19 Response

Mary Martin, BA, MPA, FACHE; Matthew Rougeux, MHA; Miranda Mathis-Harris, BSN, MBA, CNML

Duke University Hospital

Background. Duke University Health System experiences the same capacity and throughput challenges as other academic medical centers nationwide. To take on these challenges, we established a novel approach to programmatic performance improvement with a program called CARE hub—with the intention of it being a sustainable way to approach all current and future operational challenges. Foundational principles included: (1) make the best use of the space we have, making it more efficient; (2) become proactive versus reactive; (3) care for all patients who need our care; (4) ensure right patient, right bed, right time; (5) support the strategic plan and growth of service lines; and (6) leverage pilots and small changes quickly to get to meaningful change quicker. CARE hub also established that innovation (in both process and technology) would be critical to success. The first areas of focus were system balance, bed mix design, bed management, discharge planning, transfer center, and establishing a platform for analytical support tools, GE Tiles. As the first waves of projects were concluding their pilot phases and preparing for rollout, COVID-19 became a more imminent threat and the organization began scenario planning for case counts being seen in European cities. COVID-19 did not impact North Carolina and the Southeast region to those extents. However, there were significant impacts that reduced Duke’s ability to continue serving in the same capacity. These pressure points proved to be an innovation accelerant for the program. CARE hub has continued to innovate, take on challenges, produce results, and deliver on new technology.

Intervention Detail. CARE hub is a strategic program focused on delivering meaningful and impactful outcomes. The success focuses on purposeful governance, process redesign, and optimized technology. Purposeful governance: A tiered governance structure was developed to support robust accountability and quicker decision-making. Work groups were developed that report to a core team, which reports to a steering committee monthly. These meetings are not a reporting mechanism, but an action mechanism where decisions are made, actions are decided, and situations are evaluated. To support accountability, banner metrics are established as succinct measures of success. Process redesign: This included a mission to create processes that reduced the burden on frontline caregivers. Pilots were utilized to evaluate small changes rapidly and identify successful working processes. Requiring all new processes to create standard work ensures that the designs are sustainable. In addition, process redesign includes not only what is done, but how it is done. A good example is the CARE hub command center, the co-location of resources to improve collaboration in a physical space. Optimized technology: This ensures that technology is supporting the people and processes to deliver care to our patients—not inhibiting them. CARE hub developed a suite of real-time Tiles (or applications) that are accessible anywhere—from hospital computers and pin stations, to kiosks in offices and conference rooms—even on mobile devices. Tile alerts are actionable and help monitor for deviation from standard work. Additionally, the tiles were created with minimal dependencies on new workflows or electronic medical record builds.

Outcomes and Impact. There are many notable improvements since the start of the CARE hub program that have become part of organizational standard work and accelerated in the past 12 months. While daily operational huddles were established, current huddles use the real-time data information from the Tiles. An existing daily, tiered, huddle structure was leveraged and now the Tiles are being incorporated from department- to system-level meetings to ensure that everyone is looking at the same information at the same time. With atypical uncertainties in volume, a new approach to budgeting was established utilizing a digital twin that allows scenarios to be developed and put through the model to determine how much growth can be achieved given the then-current bed constraints. Hospital census has seasonal variability that can be anticipated (such as flu season). Simultaneously, the need to flex bed space more rapidly was possible through the structures and proactive analytics to enable Duke Health to provide high-quality care while reducing impacts to normal daily operations. CARE hub also delivered significant outcomes in its first 16 months (especially given that nearly 12 of those months were during a pandemic state), including: (1) a projected annualized increase in transfer center patients of 6% (a $2.4 million annual impact); (2) discharge planning that improved excess days by 6% to 12% (a $3.3 million to a $5.9 million annual impact); and (3) bed management that delivered a 10% reduction in cold bed time. Continued improvement opportunities and additional features of the Tiles are continuing to drive excess day improvements. In the new bed mix design, the digital twin created a scenario that will: (1) increase on-service placements; (2) reduce boarding time within the operating room and the emergency department; (3) co-locate services to better support physician workflow; (4) allow for strategic service line growth; (5) capture additional transfer center patients; and (6) allow for allocation of specialty units to be available for the next acute change in health care.

Adapting Telehealth Workflows for People Living With Dementia

Laura Medders, LCSW, Emory Healthcare; Carolyn K. Clevenger, RN, DNP, AGPCNP-BC, GNP-BC, FAANP, FGSA, FAAN, Emory Integrated Memory Care Clinic

Emory Healthcare

Background. In March 2020, like many health care systems, Emory Healthcare and the Emory Integrated Memory Care Clinic (IMCC) were forced to adapt workflows to protect staff and patients from the risk of COVID-19. Like many practices, this resulted in a sudden, unexpected conversion to telehealth appointments for patients. The IMCC is a primary care clinic for people living with dementia, and our patients, families, and providers shared concerns about continuing in-person appointments for our vulnerable patient population. At the time, community spread was high, and people living with dementia (PLWD) had difficulty remembering the new social distancing and masking requirements. However, the need to access primary care was increasing. PLWD also often experience a precipitous cognitive and functional decline due to the lack of traditional community support. Their care partners often face increased social isolation and caregiver burden as a result. In addition to the patient health concerns, our health care system and department faced a decline in revenue due to COVID-19’s impact. Our team was tasked with rapidly increasing total relative value units to decrease the gaps in revenue. IMCC providers and other team members rapidly adapted the in-person workflows to telehealth to meet the increased patient need for clinical support and to improve financial sustainability. While the health system worked to adjust to telehealth in a general sense, the IMCC team developed solutions for how to gather feedback from the care partner before a visit—including administering standardized functional and behavioral assessments and adapting neuropsychological testing for telehealth. As a result of these changes, the IMCC was able to increase the number of patient encounters to a pre-COVID-19 rate while also increasing the overall number of Medicare Cognitive Assessment Visits conducted.

Intervention Detail. Prior to March 2020, the IMCC did not utilize telehealth. By May 2020, all clinic appointments were telehealth. New telehealth appointment types specifically for the IMCC were created, and new workflows for all team members were created and implemented. The IMCC made improvements to Emory Healthcare’s system workflow to help prepare our families living with dementia for a telehealth encounter. The department created a Zoom Prep Team that contacted patients before a telehealth appointment to address technical difficulties and increase their comfort level with the telehealth platform. Standard work was created to help troubleshoot recurring Zoom challenges. As the public became accustomed to virtual meetings, this prep team was dissolved. For 99 483 encounters requiring cognitive assessments, multiple testing processes were adapted to successfully conduct the telehealth session. The assessment’s visual and drawing tasks were altered to accommodate the virtual environment. Rather than handing printouts to patients, the neuropsychological testers displayed figures on the shared screen for patients to draw on their paper at home. Oral tasks remained the same but the virtual environment did present some challenges. Testers were specifically trained on Zoom with information technology support to accommodate patient technical troubles. The functional assessment and neuropsychiatric inventory previously given to patients at in-person check-in were sent via email to caregivers prior to the visit using a third-party software that interfaced with the electronic medical record. This software was also used to send a dementia severity rating scale to assess the patient’s cognitive status if the patient living with dementia was too advanced in the dementia process for neuropsychiatric testing. Without the opportunity to speak with care partners privately during the visit, the IMCC also utilized prolonged, non-face-to-face encounters to gather information pre- or post-encounter to ensure an accurate and candid assessment of a patient’s health status.

Outcomes and Impact. As a result of the decision by the Centers for Medicare & Medicaid Services to broaden comparable reimbursement for telehealth services, the IMCC developed telehealth solutions for new and established appointments, cognitive assessments, annual wellness visits, and psychotherapy visits. Nonbillable services like support groups and educational classes for care partners were also converted to a virtual platform to decrease care partner isolation and burden. For many families living with dementia, the telehealth encounters have been a welcome solution to the all-day challenges of bringing a PLWD to an in-person appointment. The IMCC hopes Medicare will continue to reimburse telehealth encounters at a comparable rate to in-person visits to better provide person-centered care to this specific population. By implementing these new workflows, the IMCC was able to meet its total relative value unit goals consistently by July 2020 and decrease COVID-19’s negative impact on revenue. In April 2020, the IMCC went from seeing an average of 11 new patient appointments a month to seeing 1 new patient, went from seeing an average of 31 established appointments per week to an average of 12 established appointments, and billed zero Cognitive Assessment Visits. With the implementation of multiple rapid experiments and process improvements, the IMCC increased utilization by October 2020 to 18 new patient encounters per month, increased appointments to an average of 13.5 established appointments per week, and increased to 16 Cognitive Assessment Visit (99 483) encounters per month.

COVID-19 Post-Hospital Discharge Survivorship: A Health Equity Imperative

Nicole M. Franks, MD, Emory University Hospital Midtown; Jasmah Hanna, MS, Emory University Hospital Midtown; Theresa Johnson, PhD, PA-C, Atlanta Veterans Administration Medical Center; Zanthia Wiley, MD, Emory University School of Medicine

Emory University Hospital Midtown

Background. COVID-19 illness affects Black, Latinx, and other minority populations disproportionately, resulting in increased hospitalizations, morbidity, and mortality.^1–10 Underrepresented Atlanta COVID-19 minority populations have experienced similar outcomes; although 32% of residents are Black, 79% of COVID-19 patients hospitalized in Emory Healthcare hospitals from March 2020 to April 2020 were Black.² Emory’s COVID-19 Health Equity Interactive Dashboard also highlights significant health care disparities among COVID-19 patients in Atlanta and surrounding counties, with corresponding worse outcomes for minority populations.¹¹ Well-established social and health inequalities place many minority patients at risk of contracting COVID-19. Social determinants of health (SDOH), including decreased access to health care, increased occupational exposure to COVID-19 (eg, essential and frontline workers), and low socioeconomic status are all risk factors for COVID-19. The Centers for Disease Control and Prevention found that among 126 137 patients with an index COVID-19 hospitalization from March 2020 through July 2020, 9% of patients who survived were readmitted within 2 months of discharge; 23% were non-Hispanic Black and 21% were Hispanic.¹² The immediate and long-term care of COVID-19 patients discharged from the hospital has proven challenging, and evidence shows that 76% of patients hospitalized with COVID-19 had at least 1 ongoing symptom 6 months after their acute illness—including fatigue, shortness of breath, anxiety, depression, sleeping disorders, and joint pain.¹³ Many health care systems are opening outpatient clinics to focus on these lasting disabilities; however, these clinics are not accessible in many low-income settings.¹⁴ Interventions to improve the care of patients being discharged to socially vulnerable settings are essential.

Intervention Detail. This retrospective study described clinical characteristics, clinical outcomes, and SDOH of COVID-19 adults with all-cause 30-day hospital readmissions to 8 Atlanta hospitals from 3 health systems. Clinical data of polymerase chain reaction-confirmed COVID-19 positive adult patients (> 18 y old) with unplanned hospital readmissions within 30 days of index admission was extracted from the electronic medical records (via the clinical data warehouse and manual chart reviews) of patients admitted to 8 Atlanta hospitals from March 1, 2020, to December 31, 2020. Patient demographics, clinical characteristics, comorbidities, diagnostic workup, COVID-19- and non-COVID-19-specific treatment, clinical outcomes, and discharge characteristics were extracted from each patient’s index admission and readmission. We performed descriptive and multivariate analyses using clinical characteristics and outcomes data. Patient residence zip code data was utilized to analyze community-level SDOH such as employment; health care access; housing; education; and median household income by race, age, and gender. Data from each hospital was integrated into a central database to create a data repository. Services implemented to address disparities in health care access for discharged COVID-19 patients include care transitions support, post-hospital discharge follow-up within 24 to 48 hours, and telemedicine video follow-up assessments and clinic visits.

Outcomes and Impact. A cohort of 7155 polymerase chain reaction-positive, COVID-19 hospitalized patients was analyzed, where the average age was 62 years for all races and the number of total patients readmitted within 30 days was 463 (6.5%). Most readmitted patients (37.6%) were between the ages of 45 to 64 years, men (54.2%), and identifying as non-Hispanic Black (Black) or Hispanic (68.78%). Black females were more likely to be readmitted (50.8%) in comparison to females of other racial/ethnic groups. Uninsured patients comprised 3.2% of the readmitted cohort and those insured via Medicaid comprised 12.3%. Readmissions, coupled with comorbidities of hypertension and diabetes, were statistically significant (P < 0.05) for the total population (all groups). However, in comparison to readmissions for Whites, patients in the group comprised of Asians, Hispanics, Blacks, and the “other” group with hypertension had a greater percentage of readmissions (83.9% and 88.9%, respectively). The comorbidities with the highest odds of readmission included hypertension (odds ratio [OR], 1.4; 95% confidence interval [CI], 1.1–2.0; P = 0.04) and stages 3 and 4 chronic kidney disease (OR, 2.1; 95% CI, 1.4–3.2; P < 0.001). Patients with a high-risk Charlson Comorbidity Index had the highest odds of readmission (OR, 4.8; 95% CI, 2.1–11.0; P < 0.001). Patients who received antiviral treatment during their index hospitalization had decreased odds of readmission (OR, 0.5; 95% CI, 0.4–0.8; P < 0.001). Additionally, patients who required intensive care or mechanical ventilation during their index hospitalization had decreased odds of readmission (OR, 0.5; 95% CI, 0.4–0.7; P < 0.001) and OR (0.3; 95% CI, 0.2–0.5; P < 0.001). Of the 458 readmitted patients with available zip code data, the analyzed SDOH were significant for race only. The impact of this methodology for assessing predictors of readmissions for COVID-19 patients can inform targeted interventions to address health disparities. Further evaluation of these identified predictors and impact of health care access interventions is ongoing.

1. Gold JAW, Kong KK, Szablewski CM, et al. Characteristics and clinical outcomes of adult patients hospitalized with COVID-19 — Georgia, March 2020. MMWR Morb Mortal Wkly Rep. 2020;69:545–550.

2. Killerby ME, Link-Gelles R, Haight SC, et al. Characteristics associated with hospitalization among patients with COVID-19 – metropolitan Atlanta, Georgia, March-April 2020. MMWR Morb Mortal Wkly Rep. 2020;69:790–794.

3. Azar KMJ, Shen Z, Romanelli, RJ, et al. Disparities in outcomes among COVID-19 patients in a large health care system in California. Health Aff (Millwood). 2020;39:1253–1262.

4. Price-Haywood EG, Burton J, Fort D, et al. Hospitalization and mortality among black patients and white patients with COVID-19. N Engl J Med. 2020;382:2534–2543.

5. Mahajan UV, Larkins-Pettigrew M. Racial demographics and COVID-19 confirmed cases and deaths: a correlational analysis of 2886 US counties. J. Public Health (Oxf). 2020;42:445–447.

6. Millett GA, Jones AT, Benkeser D, et al. Assessing differential impacts of COVID-19 on black communities. Ann Epidemiol. 2020;47:37–44.

7. Henning-Smith C, Tuttle M, Kozhimannil KB. Unequal distribution of COVID-19 risk among rural residents by race and ethnicity. J Rural Health. 2021;37:224–226.

8. Kim SJ, Bostwick W. Social vulnerability and racial inequality in COVID-19 deaths in Chicago. Health Educ Behav. 2020;47:509–513.

9. Van Dorn A, Cooney RE, Sabin ML. COVID-19 exacerbating inequalities in the US. Lancet. 2020;395:1243–1244.

10. Hooper MW, Napoles AM, Perez-Stable EJ. COVID-19 and racial/ethnic disparities. JAMA. 2020;323:2466–2467.

11. Emory COVID-19 Health Dashboard. Accessed January 10, 2021. https://covid19.emory.edu.

12. Lavery AM, Preston LE, Ko JY, et al. Characteristics of hospitalized COVID-19 patients discharged and experiencing same-hospital readmission – United States, March – August 2020. MMWR Morb Mortal Wkly Rep. 2020;69:1695–1699.

13. Huang C, Huang L, Wang Y, et al. 6-month consequences of COVID-19 in patients discharged from hospital: a cohort study. Lancet. 2021;397:220–232.

14. Cortinovis M, Perico N, Remuzzi G. Long-term follow-up of recovered patients with COVID-19. Lancet. 2021;397:173–175.

Team-Based Approach to Improving Mortality Index

Doreen Timm, MSN, RN, APN/PCNS-BC; Rebecca Pedersen, MD, CPE, FAAFP; Deb Kent, MBA, BSN, RN, CCDS

FHN

Background. FHN previously participated in several Vizient improvement projects to examine ways to decrease the mortality index. Strategies such as utilizing the Vizient Clinical Data Base, tracking tools, and hospice respite were considered based upon feedback and lessons learned from other Vizient participants. We have pursued numerous strategies that have proven successful in our rural community, despite the challenges of COVID-19. We are excited to share those strategies with other Vizient participants, as we decreased our mortality index from 1.3 to 0.33 in 2020.

Intervention Detail. FHN began work on our mortality index in May 2018. Clinical documentation improvement (CDI) specialists presented mortalities in a weekly meeting with provider leaders and the quality department director. While some awareness was gained, little movement on the actual index occurred. After 6 months, the mortality index criteria was fully dissected and its far-reaching breadth was discovered. After utilizing a suggestion from a Vizient member, the CDI team identified factors related to mortalities and broke the data into 3 buckets—patient data, quality of care, and throughput. Process improvement activities were then driven from the data using a team-based approach. The weekly mortality meetings were utilized to address findings and drive corrective actions. Focus and team members evolved over time. CDI, registration, clinical quality department, and case management each led different initiatives. Clinical providers, including the chief quality officer, chief medical officer, and hospitalist director, consistently attended weekly meetings and were critical to driving provider engagement. While a steady downward shift was seen in our index, our data showed a consistent need for throughput options. In-house hospice respite was proposed and a multidisciplinary team was developed. During summer 2020, workflows were developed to enhance clinical processes using the electronic medical record. This required collaboration with CDI, clinical teams, case managers, coders, and billing teams. The first patient entered the revised workflow in October 2020. To date, 34 patients have benefited. Interestingly, this new process went live at the same time as a COVID-19 patient surge. Typically, this would result in a failure of the new process; however, our clinical teams embraced the concept of hospice respite. Providing the extra support that hospice gives our patients and families was a natural supplement to the care we offered during the surge.

Outcomes and Impact. When we began in May 2018, our mortality index was 1.98. More gripping was a few monthly indexes greater than 2.0 during the first year of mortality reviews. Once we later understood the index criteria, we started analyzing and addressing our patients’ expected mortality scores by model groups. We utilized the Vizient Clinical Data Base calculator at a 95% accuracy rate, allowing us to better predict our scores. However, our scores remained inconsistent until we identified factors that impacted the reliability of our index and addressed them in process improvement initiatives. We discovered that over 85% of our mortalities had patient data opportunities, which we addressed with CDI queries, correction of the data, and staff education. More difficult to address was our throughput opportunity rate of 95%, of which 72% was related to discharge options. After the expansion of our team and with a focus on clear goals and expectations, we began to see a consistent downward movement of our index and observed mortalities. As more clinicians became familiar with the nuances of our mortality management culture—including end-of life-discussions, correct hospitalization status, discharge options, etc., we have achieved sustained results. Our expected mortality has been higher than our observed for the past 6 months, our mortality index for 2020 was 0.93, and our last mortality index score for December was 0.33. Continued work is being done to ensure reliability and attention to this topic. We continue to track data, quality, and throughput opportunities, as well as our observed and expected mortality rates. CDI continues to utilize the Vizient Clinical Data Base calculator on a regular basis with its reviews. In addition, weekly mortality meetings continue to identify new opportunities and factors that need reinforcement.

Merging Health Care Organizations: Success Starts in the Trenches

Grant Sinson, MD; Julie Kolinski, MD; Carrie Alme, MD

Froedtert & Medical College of Wisconsin

Background. Over previous decades, the consolidation of health care delivery organizations has increased. A 2018 review of community hospitals in the United States reported that 67% belonged to a multihospital system.¹ The forces driving this change are complex and unlikely to change in the near future. It is possible that ripple effects of the COVID-19 pandemic will be an additional force supporting this trend. However, these forces do not consider the impact on the administrative workforce of hospitals. This is particularly true with physician advisor (PA) programs and their relationship with utilization review (UR). PA programs have evolved over many years. Unlike medical fields of study or specialties, there is no residency or fellowship to become a PA. While numerous courses exist, much of the training requires learning the nuances of one’s own institution. Froedtert & Medical College of Wisconsin, a 700-bed teaching hospital, has a PA program that includes 1.5 full-time equivalents (shared by 3 physicians) and a separate moonlighter group (6 physicians) for additional coverage of UR. As part of a larger enterprise, we consolidated with a 200-bed community hospital and a 70-bed community hospital that employed 1 PA and used a third-party company to provide the remainder of UR coverage. With the departure of that PA, we were asked to engage with these hospitals and assist in building a new program. The important parameters were to recruit only physicians already employed by the system and remove their dependence on third-party reviews.

Intervention Detail. A natural skepticism on the part of our community partners required first engaging with community hospital leadership. Not all hospital leaders have a detailed understanding of the scope of a PA program, much like other clinical support programs that have morphed to meet specific needs. Hospital leadership understanding is essential to access required resources. Additionally, leadership became a necessary advocate when trying to educate reluctant physicians about the PA program. Next, we met with frontline staff to learn about their unique systems. We walked through all key workflows within the initial UR by case managers, PAs, and denial management. How were they doing it? Is there anything that they did not like about their current process? The power of managing all UR internally is in the interpersonal relationships established. In the case of UR, it is between a PA and a provider. Providers respond much better to a PA who is part of the same institution and even better if they know the PA personally. Consequently, we prioritized recruitment of an on-site PA team. Once our roadmap was completed, we reengaged with the current community hospital teams. Together, we made plans for any changes in the process. Their confidence in process changes was crucial to onboarding the inexperienced PA group. With leadership’s help, an on-site lead PA was recruited to play the important role of a contact person who knows the system and can help recruit associates. With the process delineated, a 1-day course for the new PA group was organized with specific learning objectives. It was vital to support program autonomy by allowing the local PA-case management relationship to develop. However, hitting the ground running required the group’s understanding that an experienced PA was always available by email, text, or page to help.

Outcomes and Impact. Implementation of the stepwise process allowed seamless transition to the new PA program—a process we believe would transition to other clinical support programs within organizations looking to merge. With the support of the chief medical officer, a semiretired physician was identified as the local lead PA, and 6 recruited providers completed the PA team. The completion timeline was 3 months. The PA program has remained autonomous and internal since it began, with no reliance on external reviewers. Overall enterprise savings by eliminating external reviewers is $400 000 annually. Approximately once per month, their PA team will reach out for additional input. The biggest hurdle was recruitment of internal physicians, which required considerable education about the PA role. There were misconceptions regarding both the value of the job and its complexity. Direct physician-to physician discussions helped overcome these obstacles. Throughout the entire process, respectful listening to the views and goals of all parties to obtain local buy-in; developing a focused, stepwise roadmap; and overcoming barriers facilitated successful merging of different administrative groups within a growing health care system.

1. American Hospital Association. Trendwatch Chartbook 2020. Accessed January 3, 2020. https://www.aha.org/system/files/media/file/2020/10/TrendwatchChartbook-2020-Appendix.pdf.

Don’t Image Without a Safety Net: Avoiding Harmful Missed Follow-Ups

Timothy E. Klatt, MD; Stacy D. O’Connor, MD, MPH, MMSc, CIIP

Froedtert & Medical College of Wisconsin

Background. Published reports show that 3% to 8% of imaging studies discover a potentially abnormal finding that requires follow-up.^1,2 Compliance with follow-up recommendations is highly variable. Lack of follow-up can result in delayed diagnoses, poor patient outcomes, increased health care costs, decreased health system credibility, and malpractice risk. A safety net that ensures completion of indicated follow-up improves patient care and promotes provider wellness by alleviating a significant burden on both ordering providers and radiologists, who are often both legally responsible. Even though all health systems face these risks, to date, only a few have implemented an organizational solution. These programs, managed by nurses, are usually limited to imaging ordered through the emergency department (ED). Our safety intelligence event reporting identified 4 patients whose diagnosis of lung cancer was delayed by missed follow-up of lung nodules. None of the index imaging studies were ordered through the ED. To address this and move our organizational culture toward high reliability, we implemented an innovative radiology findings safety net program that ensures follow-up of all potentially abnormal findings identified within the more than 700 000 imaging studies performed annually throughout our clinical enterprise, which includes a 724-bed quaternary care academic hospital, 2 community hospitals, and 45 outpatient settings in which over 425 000 patients partner with us in their care annually. At baseline, our patients followed up at higher rates than those published. Even so, significant gaps remained. For example, 41% of those with lung nodules for which follow-up was recommended did not receive this imaging. For lung nodule follow-up alone, assuming Medicare reimbursement, potential revenue exceeded $675 000. To realize the numerous benefits, our program was designed to be managed by a radiology clinical liaison, paid $66 560 annually.

Intervention Detail. The safety net includes all cases with follow-up recommended within the next 2 weeks to 2 years. The radiologist begins the process by including the phrase “radiology safety net” in the report. The study is then automatically entered into a Research Electronic Data Capture database for use by the liaison. If the study was ordered in the ED, the safety net activates the next weekday and communicates the need for follow-up to the patient’s primary care provider (PCP) or oncologist (oncologists often serve as cancer patients’ PCPs). If the patient is not established with a PCP or oncologist, the safety net helps them establish with a PCP via our standard new patient scheduling process. Our health system communications begin with messaging from our electronic medical record. When the patient’s PCP or oncologist is external to our organization, the liaison calls their office and faxes them the report. Once a provider has accepted responsibility for ED follow-ups, the cases are added back to all other identified cases. Shortly after the expected action date has passed, the liaison reviews the medical record (or communicates with the outside provider) to determine whether follow-up has been completed or deemed clinically unnecessary, or if the patient made an informed decision to decline follow-up. If one of these has occurred, the safety net process closes the case. If none of these conditions have been met, we reach out to the ordering provider. At this point, non-PCP ordering providers are offered the option of having the safety net contact the patient’s PCP or oncologist, as they may best understand the entire clinical picture. Workflow redundancy ensures that providers are contacted at least twice before closing a case.

Outcomes and Impact. Collection of potentially abnormal findings began on April 13, 2020, although due to technical factors, tracking of follow-up completion did not begin until March 1, 2021. Of 3482 trackable recommendations, 1527 (44%) have not yet become due. Using most recent data from July to September 2021, an average of 454 recommendations were made each month from an average of 51.08 nonmammographic imaging exams (0.9%). For the 413 recommendations stemming from inpatient imaging exams, 354 (86%) were included somewhere in the discharge summary, although only 102 (25%) were included in the “follow-up needed” section of the summary. Forty-one patients were offered access to PCPs through our new patient scheduling process, although only 10 (24%) accepted. All recommendations from inpatient imaging were deferred to the patient’s PCP, suggesting a potential alteration of the program to automatically route these recommendations to the PCP. The safety net reminded providers of incomplete follow-up 96 times (5% of recommendations)—30 for ED imaging, 4 for inpatient imaging, and 62 for outpatient imaging. This supports our design of tracking follow-up from exams throughout the enterprise rather than focusing solely on the ED. Safety net reminders resulted in 53 follow-ups within our enterprise, 29 follow-ups external to our organization, 8 follow-ups deemed not clinically indicated, and 6 follow-ups that patients made an informed decision to decline. Over the past 6 months, follow-ups within our organization, using Medicare reimbursement values, generated $24 138—although this does not include a cholecystectomy to be performed for an indeterminate gallbladder polyp or care for rectal cancer with a timely diagnosis after a safety net reminder. Reminders also resulted in discovery of a likely parotid gland tumor for which the patient declined follow-up, biopsies for postmenopausal bleeding (both benign), and symptomatic improvement of patients after antibiotics for missed cystitis and colitis. Lessons learned from this project include that 95% of follow-up recommendations from radiology exams were addressed without safety net intervention and that outpatient imaging required safety net reminders twice as often as ED imaging. Safety net implementation did not result in a large influx of new patients to our PCPs. Our program, which covers imaging throughout our organization and contacts both internal and external providers, can be successfully run by a clinical liaison. It is achievable and enhances patient care, and we anticipate that it will be financially self-sustaining.

1. Anthony SG, Prevedello LM, Damiano MM, et al. Impact of a 4-year quality improvement initiative to improve communication of critical imaging test results. Radiology. 2011;259:802–807.

2. Lacson R, Prevedello LM, Andriole KP, et al. Four-year impact of an alert notification system on closed-loop communication of critical test results. AJR Am J Roentgenol. 2014;203:933–938.

Moving to EXTREME ACCESS for Our Patients

Doug Marx, DO

Froedtert & Medical College of Wisconsin

Background. Maintaining adequate access for patients in primary care, specialty care, and ancillary services is fundamental in achieving consumer satisfaction and growth for the organization. As consumer expectations rise, health systems face the challenge of creating patient-centric access that is timely, geographically convenient, and flexible. Historically, like most physician organizations, we reported retrospective access: (1) percentage of sites meeting < 7 days third-next available new patient appointment (TNAnew) and (2) Clinician & Group-Consumer Assessment of Healthcare Providers and Systems (CG-CAHPS) Dimension 1 around access to care. While this approach is useful, it does not drive performance that meets the new consumer’s needs. Our strategic imperative to grow primary/specialty care in the community forced us to think about access in a new way. We began from a position of power (90% of sites < 7 d TNAnew and 85th percentile CG-CAHPS access to care). Our finance and analytics team recognized that the existing tools did not meet our operational need to create consumer-focused access, so the team partnered with operational leaders and practice executives to take a fresh look at our approach. The decision was made to pivot from passive reporting to active management and creation of appointment slots. We tasked the finance and analytics team with developing a new set of tools and benchmarks. We called the project the “Extreme Access” initiative. The fundamental principle of Extreme Access is to focus on the prospective creation of appointment spots to meet consumer demand and grow the practice.

Intervention Detail. This work led to the evolution of our analytics tool kit. Process: We challenged our local leaders to begin focusing on the creation of access on a week-to-week basis. Leaders received a detailed report on their clinic utilization every Friday. By Wednesday, they met with their teams and developed tactics to address the information provided before the next report is published again on Friday. Weekly report: The weekly report, distributed to operational leaders (directors and medical directors) every Friday, told a story in terms of the number of slots available, slots scheduled, and fill rate (available/scheduled) over the course of 3 weeks. The 3 weeks reflected the last week, current week, and upcoming week. The report also showed these metrics on a daily basis to identify trends. Three views of the report (executive, regional, and site) provided detail to drive performance and accountability. This information helped identify trends and areas of opportunity for clinics to address patient demand and accessibility to our sites. For example, on Thursday and Friday, a site is showing a high fill rate but has low availability for patients. On Monday and Tuesday, there is high availability with lower fill rates. This allows leaders to consider level-loading the availability of providers and advanced practice providers (APPs) to later in the week when patient demand is so high. Weekly tactics: Looking forward to utilization in future weeks, leaders were expected to submit weekly tactics to create additional appointment slots to meet patient demand in areas below a predetermined benchmark/baseline. Operational leaders were tasked with developing real-time tactics to create additional capacity (slots). This work led to a cycle of continuous improvement with weekly oversight by executive leaders and engagement of local leaders to problem solve, optimize, and collaborate across sites. The local leaders identified themes that fell into several common buckets: template optimization, weekday level-loading, utilization of Saturday hours, and data-driven operational decision-making. From there, the local teams managed a set of tactics to drive performance, including managing paid time off, reducing no shows, handling strategic overbooking, and ensuring top-of-license clinical work. Our executive leaders reviewed performance and held directors and managers accountable for results.

Outcomes and Impact. In less than a year, we saw a steady and sustainable increase in the number of available slots across the community arm of the practice. Examining slots and fill rate provided us with a good indication of demand for services. The traditional metrics of access held stable (> 90% of sites < 7 d TNAnew and 85th percentile CG-CAHPS satisfaction with access to care). Additionally, the new metrics gave us the opportunity to examine whether we were meeting demand and adding scheduling capacity that would translate into new growth. After 5 months, we went from 5% to 73% of sites that were performing better than baseline when looking at the week ahead, as defined by the weekly number of available appointment slots. Work week level-loading improved across the practice, resulting in availability that is more patient-centric than physician-/APP-centric. Physician availability increased significantly, showing a 23% rise in the number of physicians with TNAnew < 7 days. Because our APPs already had adequate access, APP access rose more modestly, with a 17% increase in the number of APPs with TNAnew < 7 days. We successfully hardwired practices to impact physician availability. We are no longer completely dependent on APP access for new and existing patients. Physician availability translates directly into practice growth and larger panel size. Reflecting upon our focus on metrics to support the local operations teams allowed us to achieve optimal performance for our patients—taking a practice from having good access to Extreme Access.

Take Flight: Modeling Reliable Physician and APP Critical Care Staffing

Sarah Vanderlinden, PA-C, DFAAPA, Froedtert & Medical College of Wisconsin; Julie W. Raaum, FNP-BC, AAHIVS, Medical College of Wisconsin

Froedtert & Medical College of Wisconsin

Background. Increased workload due to the complexity of critically ill patients demands an ever-evolving workforce to deliver high-quality, safe, and reliable care 24 hours a day, every day. Patients and their families expect it and depend on it. We looked to the airline industry for lessons learned in delivering high-reliability service in complex environments. When a passenger flies, they expect that the process will be predictable and executed the same way every time by team members—from check-in to security screening, take-off to landing, and all the way through to luggage collection. This process has become so predictable that it is easy to overlook how complex it really is. But there is a complexity behind matching the workload with the workforce in a calculated and deliberate manner. Our health care system has experienced growth in inpatient workload from the addition of inpatient capacity and increased acuity from complex specialty populations. These programs generate critically ill patients needing critical care environments. We have 5 critical care teams composed of attending physicians, physician residents and fellows, and an increasing number of advanced practice providers (APPs). The teams, grown over the years in silos, adapted to local workload and staffing changes. Historically, there has been variation in workforce approvals, inconsistent critical care staffing models, and no tools to describe workload. By creating tools to assess workload and workforce and identifying data sources and definitions, we developed a model to help us predict our critical care physician and APP staffing models. As the critical care patient population grows, the health care system needs a standard and reliable staffing model to anticipate provider staffing; inform operations; and integrate return on investment, including cost and quality patient outcomes.

Intervention Detail. Tasked to describe the current state of each critical care service staffing model for physicians and APPs, the project team was also charged with creating a formula for critical care physician and APP staffing to inform workforce planning. The project team, sponsored by executive leadership, consisted of advanced practice directors, the transformation director, and practice finance. A review of the literature was curated, including assessment of current state for each critical service, and review of best practices for physician intensivists; however, there was little describing the construct of critical care teams or outcome-driven formulas. Consensus was achieved on a patient ratio range of patients per physician/fellow and for APP/resident across all services. The project team led focus group interviews with physician, APP, and administrative leadership to complete a descriptive analysis of workload. To better understand each critical care team’s operations and workload, a macro work assessment was completed with the aim of describing the work completed within a 24-hour period by each team member based on role. Common themes were identified. The macro assessment was subsequently translated into a micro work inventory to assign estimated times to key workload tasks. In addition to the descriptive analysis, data was also used to calculate census for workload factor. To date, there is not a uniform method to identify team census due to variation in critical care team models. Consensus was achieved across the services that census would be counted by the number of patient touches in a 24-hour period abstracted from provider notes in the electronic medical record (EMR). In addition, modified sequential organ failure assessment and critical care time were added as metrics to inform workload, as they describe patient acuity. An acuity factor (high, medium, low) was applied to reflect workload. A staffing model and formula were developed to identify the workforce based on the team workload using census and acuity.

Outcomes and Impact. The project outcome is the development and adoption of internally validated tools, definitions, and data points. This organization conducted reliable, reproducible, and comprehensive analysis of critical care physician/APP team workload and workforce to inform workforce planning. Key highlights include: Evaluation of the workforce: Roles evaluated include attending physician, physician fellows and residents, and APPs. For purposes of this project, staffing was reported in time frames of 28 days, defined as a “block.” A 24-hour day was divided into day coverage and night coverage and provider workload contributions were by full-time equivalent status. Evaluation of the workload: Standard work data and definitions were established using work inventory tools and facilitated focus groups. Project leaders guided subject matter experts to categorize findings by common themes and coached the subject matter experts to consensus on categories of workload common to all critical care services, including patient management, provider team communication and collaboration, and interprofessional and family communication. EMR metrics in each category were identified to support measurement for each category. Visualizing the workload of each team member across the team enhanced understanding of roles and provided data to assess changing workloads or staffing models. Moreover, this provided a standardized method to evaluate workload by role. Service census and acuity were also factored into workload. A census surrogate to “heads-in-beds” was obtained using notes written by the critical care teams from the EMR. All patients do not require the same level of care, but no standard way of stratification existed. The workload level was divided into 3 acuity categories, and acuity was used to adjust the census to support the staffing model. Critical care billing time captured as current procedural terminology codes 99219 and 99292 were also surrogates for workload time and measurement. The staffing model formula, Workforce + Census × Acuity, is inclusive of attending physicians and APPs and supports forecasting for workforce planning.

Improving and Transforming End-of-Life Care by Implementing a Hospice Primary Team Pathway

Balaji Narayanan, MS IE, Froedtert & the Medical College of Wisconsin; Paul Stellmacher, MD, Medical College of Wisconsin

Froedtert & the Medical College of Wisconsin

Background. Access to end-of-life care continues to be limited for severely ill and terminal patients across the United States. Patients’ short prognosis due to their clinical condition and the inability to transfer them home or to residential hospice facilities can often lead to patients and families not being able to access their hospice benefit. As a level 1 trauma center with high-functioning oncology, pulmonary critical care, and cardiothoracic/vascular service lines, we were unable to provide high-quality, end-of-life care to patients—and more importantly to families that needed bereavement support. This critical area of need triggered the business case to develop a virtual hospice model in which end-of-life resources, such as hospice physicians, nurses, social workers, child life specialists, and music thanatologists are provided to eligible patients and families.

Intervention Detail. (1) Implement primary team pathway in critical care units where the intensivist is trained to be the hospice physician, providing continuity to patients and families; (2) partner with local hospice agency to bring bedside hospice care to eligible patients; (3) implement a performance dashboard to monitor usage of end-of-life care resources and optimize workflows from referral to admit; (4) develop training on symptom management, clinical workflows, and serious illness conversations for primary physicians—thereby empowering additional providers to offer end-of-life resources; and (5) high-functioning mortality review team provides near real-time feedback on missed opportunities and coaching to frontline providers.

Outcomes and Impact. Between FY19 (July 2018 through June 2019) and FY20 (July 2019 through June 2020), our organization:

• Improved admissions to hospice by 39%.
• Served 88 additional patients and families.
• Improved awareness, referrals, and penetration of end-of-life resources (palliative care by 3% and hospice by 32%).
• Increased average census for hospice patients by 44% (2.01 to 2.89).
• Reduced time from hospice referral to admission by 45% (15 h to 8.3 h).
• Improved patient and family satisfaction to achieve top decile performance.
• Improved mortality domain ranking from top quartile to top decile (22 in FY19 to 9 in FY20), as measured by the Vizient Comprehensive Academic Medical Center Quality and Accountability Performance Scorecard.

Don’t Burn the Supplies! Baking Burn Rate Calculators From Scratch

Amanda Puls, BSN, RN, CVAHP; Jack Koczela, MBA; Michael Tabbert, EE BEM MEM; Paul Darko, BSME, MBA; Alissa Strange, BS

Froedtert Health

Background. Our facility already had a self-distribution supply model in place prior to the COVID-19 pandemic. With an integrated service center, we were able to efficiently maintain our inventory. What the pandemic did not prepare us for was the unpredictable burn rates that fluctuated by item, department, patient population, protocols, and severity of the virus. The pandemic left us wondering how fast items were going out, how long it would take for orders to come in, and how much we would get at any given time. Our C-suite quickly transformed into an enterprise incident command (EIC) structure, requiring an hourly pulse on the critical supplies we had available and the status of the supply stream. Multiple models were being developed on the fly by third-party organizations. Our supply chain team developed its own data model and burn rate calculator to allow our EIC to make real-time decisions to protect and care for our patients, staff, and community.

Intervention Detail. Using enterprise resource planning and electronic health record (Epic) data, our Supply Chain Value Analytics and Informatics teams developed a data model that contained both personal protective equipment (PPE) and critical supply data needed to support the pandemic. The first steps were to categorize our items within our enterprise resource planning system as PPE or a critical COVID-19-related supply. Using United Nations Standard Products and Services Codes as our identifier, both clinical staff within and outside of supply chain made these identifications. Writing multiple database queries, a data model was developed showing daily utilization, receipts, allocations, stock on hand, and other supply trends of these categorized supplies. Additional subcategories were developed to handle multiple SKUs under 1 category. Items like masks, gloves, and oxygen tubing were all categorized with dozens of SKUs each. This gave us an overall burn rate of not just the individual SKU but also a roll-up to the category as a whole. This type of categorization provided potential substitutions during critical moments. Taking the data model further, our team drilled down to utilization per COVID-19 patient and non-COVID-19 patient. Finally, the calculator was made interactive for the EIC or any other staff member attempting to visualize how our supply would be affected by change. Interactive functions were put in place to enter any amount of COVID-19 surge, both by single patient and/or a percentage. Fluctuations in patient volumes and the effect on supply utilization could be compared side by side in each of our service spaces, such as surgery, ambulatory, and inpatient. In summary, data points developed and monitored included:

• Average daily use
• Stock on hand, both with and without projected increases
• Quantity on order
• Average daily receipts
• Use per COVID-19 patient
• Use per non-COVID-19 patient
• Percentage increase of utilization week-to-week in all supply categories
• Specific department utilization
• Over-time visualizations

Outcomes and Impact. The calculator was a success on multiple levels. Our supply chain team utilized it daily to identify potential stockout dates and substitutions. It drove our daily priority of work and escalation needs, both with our suppliers and our EIC. A supply status report was made to our EIC daily and was extremely visual to our C-suite. Similar to the rest of the country, summer 2020 brought a decline in COVID-19 census. Knowing we needed to prepare for what was to come, our calculator allowed us to successfully prepare our organization for the second surge of the pandemic. The surge was beyond our original estimation, as it was for many organizations. We still were able to maintain a level of supply to support our staff and patients, along with the ever-changing Centers for Disease Control and Prevention guidelines. With our data, we were able to share the current PPE and critical supply status with all of the organization’s staff via our intranet. Color-coding provided an easy visualization to help determine which supplies needed careful monitoring and judicious distribution. This supported our organizational value of transparency. Realized benefits include: (1) a published calculator that is easy to refresh daily; (2) transformation into a predictive model that supports surges or declines in the pandemic; (3) department-specific information to identify potential outliers in PPE or critical supply utilization; (4) the ability to project utilization with and without reprocessing; (5) a method to report required items to both state and federal bodies; and (6) the ability to open new COVID-19 units knowing exactly how many supplies are available. Lessons learned include: (1) collaborate with your clinical leaders early on in the process; (2) establish a route of escalation to use when outliers are identified; (3) review your units of measure and review them again as necessary; and (4) develop a process to accommodate returns (negative values) to your inventory.

Face Down, Sats Up: Proning Breathes Life Into Systemwide Mobility Program

Allison Wier, DNP, RN, CMSRN, AGCNS-BC, APNP; Erin DeJarlais, PT, DPT; MaryAnn McKenna Moon, APRN, MSN

Froedtert Hospital

Background. The emergence of the novel coronavirus, SARS-CoV-2, commonly referred to as COVID-19, has had a tremendous impact across the world, affecting over 270 million people.¹ As the virus evolved, so has our knowledge of treatment as care providers learning what was most successful in improving patient outcomes. Our acute care setting was inundated with the rapid respiratory decline of COVID-19-positive patients, leading to an increase in rapid response team (RRT) calls with intensive care unit (ICU) transfers. Nursing staff and intensivists receiving the deteriorating, post-RRT, COVID-19 patients placed them in the prone position, resulting in decreased patient respiratory demands. In March 2020, the organization implemented a clinical practice guideline recommending prone positioning for all COVID-19-positive patients. Although prone positioning had previously been utilized for severe acute respiratory distress syndrome in ICU patients, prone positioning was not applied as a treatment modality in the acute care and emergency department (ED) setting. The demands of educating patients and completing the proning in a timely manner placed even more constraints on the nursing staff. Froedtert Memorial Lutheran Hospital incident command, in collaboration with the director of advanced practice and the rehabilitation services department supervisor, recognized the need for mobility experts to optimize proning of patients across the continuum of care. With assistance from process improvement, the Prone Team deployment started a few weeks into the first COVID-19 surge. Educational services and ICU registered nurses experienced in proning were assembled to train the therapy professionals who were available for redeployment, due to a decrease in non-COVID-19 patient census. Bringing furloughed therapy staff from the ambulatory and inpatient settings required physical therapists, occupational therapists, speech-language pathologists, athletic trainers, and rehabilitation technicians to assemble the expert base for the Prone Teams.

Intervention Detail. Interdisciplinary Prone Teams were created to support our complex patients in the acute care, ED, and ICUs. The teams also completed simple tasks that helped limit nurses’ COVID-19 exposure while decreasing personal protective equipment usage. The Prone Team provided advanced mobility for complex patients to ensure they continued to mobilize at pre-pandemic standards. The team also mobilized patients outside of skilled therapy sessions by going room to room on COVID-19 units while screening for patients who would benefit from therapy consults. Clinical nurse specialists alongside the Prone Team developed education for patients about the benefits of prone positioning during their inpatient stay and after discharge. To improve patient outcomes, prone positioning criteria was implemented with a standard, consistent approach that started in the ED and continued through the transitions of care. The Prone Team also improved patient compliance by educating and demonstrating adjustments to make the prone position more comfortable. Rehabilitation staff developed an educational packet about exercise and activity for patients with COVID-19 to promote further activity and mobility. During a second COVID-19 surge in fall 2020, a modified Prone Team approach was initiated. The Prone Team and clinical nurse specialists rolled out hands-on education to registered nurses and certified nursing assistants on how to safely prone patients. This provided knowledge to non-COVID-19 unit staff enabling them to prone a decompensating patient before transferring them to a COVID-19 unit. The education also led to the development of proning champions on COVID-19 units that resulted in empowered staff. To solidify this education, the Prone Team created a document outlining the benefits of the prone position to assist in identifying appropriate patients. Nursing staff developed signs to post in patient rooms and utilized safety rounds and assignment boards to enhance communication on which patients needed proning.

Outcomes and Impact. Assessment of the effectiveness of the Prone Team, leaders, and clinicians was done through utilization of data from internal surveys; the Vizient Clinical Data Base; Resource Manager, a Vizient Clinical Data Base tool; Get With the Guidelines (from the American Heart Association and the American Stroke Association); and Clarity, an Epic-developed database that supports its electronic medical record. Nursing staff from COVID-19 units and Prone Team members were surveyed using a Likert Scale to assess the effectiveness of the Prone Team intervention. Over 79% of respondents reported the teams were extremely effective or very effective at mobilizing patients, reducing nursing workload, and preventing deconditioning. Throughout the Prone Team activation from February 2020 to June 2021, Vizient Clinical Data Base data noted clinical impact, with a 20% reduction in mortality, a 27% decline in RRT and medical emergency calls, and a cumulative 17% reduction in RRTs resulting in ICU transfer utilization. During this time frame, the patient case mix index remained between 1.95 and 2.23. Froedtert Memorial Lutheran Hospital maintained an ICU transfer reduction of 24%, even with an increase in case mix index during the first surge (March 2020 through May 2020). Once our first surge subsided, we were able to establish a Prone Team reactivation plan that we used as we experienced a second COVID-19 surge and planned for subsequent surges. The Prone Team taught us essential lessons to advance the culture of mobility for complex patients in the ICU and acute care settings. A systemwide mobility program has since been implemented to enhance patient outcome measures. The interprofessional collaboration between nursing staff and Prone Team members was crucial to our high-quality COVID-19 patient outcomes. The prone positioning initiative, enriched by the Prone Team, will have a lasting impact on the organization.

1. World Health Organization. WHO Coronavirus (COVID-19) Dashboard. 2021. Accessed December 16, 2021. https://covid19.who.int/.

Reducing Harm During COVID-19: Advancing High Reliability Through CNO/CMO Partnerships

Jason H. Gilbert, PhD, MBA, RN, NEA-BC; Michele Saysana, MD

Indiana University Health

Background. Since the early 2000s, our health care system has been on a mission to reduce patient harm events through implementation of evidence-based practices and high-reliability quality and safety principles. Between 2015 and 2019, we achieved a more than 50% reduction in harm events—specifically central line-associated bloodstream infections (CLABSI), catheter-associated urinary tract infections (CAUTI), and hospital-acquired Clostridioides difficile infections. Traditional structures included a coordinating council for the system with subcouncils focused on different harm events. The councils convened in-person on regular cadences to review performance metrics, appraise evidence, evaluate new products, plan interventions, and spread best practices to other system hospitals. For several years, this structure and process led to improved harm outcomes across the system that met or exceeded goals. When system harm goals were established for 2020, the strain that COVID-19 would place on health care teams and care models was not yet imagined. The impact of the pandemic on the clinical environment brought rapid influxes of highly acute patients, frequent changes in care protocols, supply chain disruptions, adaptation of traditional care models to meet patient demand, and clinician stress and anxiety—all of which created an urgent need to evolve traditionally successful structures and processes used to lead quality and safety improvement outcomes. This presentation focuses on the innovative, yet inimitable structures and processes established during the pandemic that are designed to coordinate quality and safety outcomes across a 17-hospital health system. This was achieved through further integration of high-reliability principles, narrowing focused efforts in rapid cycle improvement, and empowering the leadership structure through regional chief nursing officer (CNO)/chief medical officer (CMO) dyads. This effort led to a reduction in harm events by 11% in 2020 and allowed the system to meet the established harm goal, despite the effects of COVID-19.

Intervention Detail. The overall response to the pandemic necessitated an evolution of the quality and safety structure to become increasingly agile with accelerated interventions and dissemination of results. The evolved structure included daily huddles attended by the regional CNO/CMO dyad, led by the system chief medical executive or chief quality officer and the chief nurse executive. This comprehensive effort empowered the regional leadership dyads through enhanced, transparent communications; relationships based on trust and information sharing; and targeted resource support. High-reliability principles embedded into the structure and process included:

• Preoccupation with failure: This includes scenario planning and execution for anticipated or realized issues such as staffing, space, supply, or equipment shortages; testing protocols; ventilator management; and new care delivery models.
• Reluctance to simplify: Bidirectional communication with quick and efficient decision-making informed by the front line with swift dissemination of information was employed to address the complexity and novelty of caring for COVID-19 patients.
• Sensitivity to operations: Working collectively ensured connection to the big picture, rapid identification of needs, and nimble resource deployment to hospitals experiencing operational challenges.
• Deference to expertise: Frontline clinicians were included in the structure through a clinical care council comprised of interdisciplinary professionals who worked to inform solutions to clinical issues and identify safety issues.
• Commitment to resilience: As variable patient surges occurred throughout 2020, so too did harm events. The structure allowed for resiliency through rapid identification of harm events, analysis, mitigation, and shared learning to course correct rapidly.

Data analysis included weekly review with regional CNO/CMO dyads of number and type of harm event (CLABSI, CAUTI, and hospital-acquired C. difficile infections), along with qualitative findings of root cause analyses conducted after a harm event. Key findings, learnings, and results of interventions were shared weekly.

Outcomes and Impact. Since 2016, the system realized a 59% reduction in CLABSI, CAUTI, and hospital-acquired C. difficile infections. When our first patient was diagnosed with COVID-19 in March 2020, much of the focused work to prevent harm shifted to clinical care of patients with COVID-19. As initial COVID-19 volumes abated in summer 2020, the regional CNO/CMO huddle shifted to a weekly cadence to maintain focus on harm and share best practices to prevent hospital-acquired infections. This huddle remains in place today, at the request of the CNO/CMO dyads, to increase communication and collaboration surrounding clinical care issues. Despite the multiple challenges faced in 2020 during the pandemic response, the system realized a further reduction of 11% in CLABSI, CAUTI, and hospital-acquired C. difficile infections, meeting system harm reduction goals. The implementation of high-reliability principles with the CNO/CMO dyads across the system allowed for agility in identifying and mitigating harm event trends, while also informing care improvements locally and nationally. For example, in late 2020, many regional hospitals experienced increased C-difficile and CAUTI infections. The problems were identified and mitigated within 2 weeks. Mitigation plans were then quickly spread throughout the system to prevent other harm events. In completing root cause analyses for CAUTI, we found that our national resource for culturing procedure did not contain the most current, evidence-based practice. The CNO/CMO dyads submitted this finding to the resource company, which in turn updated the nationally used procedure. The CNO/CMO dyads also collaborated to develop strategies and procedures to care for prone patients, given the increased number of COVID-19 patients requiring this care modality. The team continued to meet in this frequent, focused manner in 2021 to further reduce harm events and improve mortality and the patient experience.

CDI—Can Do It: A Success Story About Improvement in “Expected” Mortality

Kearstin Jorgenson, MSM, CPC, COC; Sathya Vijayakumar, MS, MBA; Kory T. Anderson, MD

Intermountain Healthcare

Background. Providers have not historically understood the “why” of clinical documentation improvement or the basic principles of risk adjustment. The Vizient Clinical Data Base compliments the dissemination of foundational principles in these 2 areas but specifically demonstrates improvement opportunities in each service line. Common themes that emerged from the Vizient Clinical Data Base data were opportunities to improve metrics for mortality and patient safety indicators. Data highlighted educational opportunities for the importance of comorbidity education and inclusion/exclusion elements specific to measures. Addressing these critical issues would help our frontline caregivers provide the best care and positively impact mortality outcomes. Hence, our Intermountain Physician Advisory Services undertook the task of empowering our caregivers with their respective data and sharing documentation best practices for their respective service lines. The Office of Patient Experience collaborated with the chief medical officer and administrator of our trauma facilities in 2020 to roll out the education. These senior clinical leaders helped emphasize the importance of provider attendance during the challenging COVID-19 pandemic. Medical directors messaged documentation as a critical safety element to ensure care teams were set up for success as each patient’s story was accurately recorded. During the pandemic when the training was remote, it was critical that we implement this initiative successfully.

Intervention Detail. Caregiver-specific scorecards using data from the Vizient Clinical Data Base physician insight card were generated as part of the core curriculum using these steps: (1) a list of caregiver names was obtained from the department requesting training; (2) case volumes by diagnosis-related group for the listed physicians over the previous year were obtained from the system enterprise data warehouse; (3) the diagnosis-related groups that constituted 80% of the case volumes for these physicians were used as an advanced restriction on the physician insight card; (4) discharge and principal procedure physician were selected for physician role; and (5) bad data, normal newborn, nonviable neonates, pediatrics age, and hospice were all excluded. Intermountain Physician Advisory Services educators shared unblinded scorecards with physicians during remote training. The strategy was to implement peer-to-peer education to foster an optimal learning environment. All participants were encouraged to use a web camera during these sessions to improve engagement. For each Vizient service line, the top 10 most underdocumented risk variables were shared with pocket-sized tip cards highlighting the missing variables. The goal of these training sessions was to empower our caregivers with the information and tools necessary to document all relevant conditions that are most underdocumented based on Vizient Clinical Data Base data. Throughout 2020, web conferencing sessions were scheduled at 4 trauma facilities, engaging 541 providers in multiple Vizient service lines. Each facility’s respective Medical Executive Committee approved the education sessions for providers during the pandemic. The medical director or designee attended each session to vocalize both support of the content and an expectation that participants engage in the presented improvement tactics. The focus was on providing tools, including reports, tip cards, and one-on-one sessions, to give all participants a chance to learn how they could get credit for the excellent care they provided.

Outcomes and Impact. The 4 Intermountain trauma facilities that participated in the education sessions improved overall expected mortality from a baseline of 1.52% in 2018 to 1.96% in Q4 2020. COVID-19 cases were removed in Q4 2020 due to the significant increase in expected mortality with this diagnosis. It is also important to consider that many of the trauma facilities operated with a different patient mix in 2020, as they focused on COVID-19 while load-balancing with other community facilities. Regardless of the circumstances, there was a demonstrable improvement in the expected mortality based on the remote education sessions provided in 2020. The key outcomes of this engagement include: (1) 541 providers (physicians and advance practice providers) at 4 trauma facilities were empowered with their data and strategies to improve documentation; (2) a statistically significant increase in coding and documentation, as demonstrated by the increase in expected mortality; (3) our Physician Advisory Services team was able to devise a strategy to share this data at a regular cadence, allowing our caregivers to see the impact of their efforts; (4) this engagement model is now being applied to our other system hospitals—specifically the community and rural facilities; (5) the medical directors have asked for clinical documentation improvement Core Curriculum 2.0 to ensure sustainment of gains beyond 2020; and (6) making this education an in-house curriculum saved Intermountain Healthcare an estimated $9.2 million in consulting fees to external entities.

Leaping to A’s: Using Medication Safety Improvement and Vizient Leapfrog Templates

Elizabeth McKnight, MA, MS, CPPS; Sathya Vijayakumar, MS, MBA; Harmony Schneider, PharmD, BCPS; Farukh Usmani, MD, MSHSM

Intermountain Healthcare

Background. In 2016, Intermountain Healthcare hospitals earned a lackluster 2 A’s, 4 B’s, and 4 C safety grades on its Leapfrog safety report card—“not something you’d necessarily want to put on your refrigerator as a report card,” our CEO Marc Harrison joked. We prepare as many as 15 Leapfrog surveys annually—including those for a pediatric hospital and 2 ambulatory surgery centers—to provide a snapshot of our safety progress compared to the previous year. Our system has strengthened its strategic safety mission since 2016 and adopted a centralized, standardized approach to Leapfrog survey submission to help mark progress toward high reliability and keeping people safe—to meet our own expectations and compare strongly to national practices. As of spring 2021, our Leapfrog safety report card is holding steady at 9 A’s and a single B grade. Leapfrog uses a balanced scorecard of publicly reported and independently submitted safety metrics, structured around the National Quality Forum’s Safe Practices for Better Healthcare. Maternal outcomes and inpatient- and outpatient-specific procedural volumes contribute significantly to workload for the annual submission and have presented significant challenges to our small, central team accountable to submit this data. Since 2019, Vizient Leapfrog templates have systemized and streamlined the submission process and enabled a steady increase in submission quality and results, and thereby higher numeric scores. The combined measures of computerized physician order entry (CPOE) effectiveness, bar code medication administration (BCMA) compliance, and communication with patients about medication (Hospital Consumer Assessment of Healthcare Providers and Systems) comprise 14.3% of the safety grade. Working with our pharmacy and medical informatics teams allowed us to drive needed improvement through the systematic guidance and benchmarking provided by the Leapfrog process.

Intervention Detail. Our system approach to improve our Leapfrog safety grade included 3 core components: (1) using Vizient templates to standardize our maternity and inpatient- and outpatient-specific procedural volumes; (2) partnering with our pharmacy and medical informatics to optimize, validate, and improve our CPOE and BCMA compliance; and (3) setting realistic, achievable improvement targets with our hospital administrative teams annually. In this abstract, we illustrate how we accomplished moving from mostly B and C grades in 2016 to 9 A’s and 1 B grade by spring 2021. For an entity like Leapfrog, accurate data collection and submission is key to the result. For our procedural volumes and outcomes, making use of Vizient templates from the Vizient Clinical Data Base helped our submission team eliminate any errors in the data categories/collection process. Our second biggest opportunity was improving our CPOE compliance. An information services group, composed of information technology-savvy clinicians, physicians, pharmacists, and registration specialists, was used to complete the CPOE evaluation. Each question that Leapfrog rated “failed” was categorized and mitigation options were analyzed using the Lean Six Sigma Possible, Implement, Challenge, and Kill chart. Improvements were prioritized and then scheduled over the following year. Using this process repeatedly allowed Intermountain Healthcare to greatly improve our scores while continuing other ongoing maintenance and enhancement work. Thirdly, our Pharmacy and Medical Informatics Team improved capture of BCMA data and standardized the dashboard for BCMA compliance, thereby further improving our safety grade. Lastly, as a submission team, we sat with each of the submitting facilities’ administrative teams and set realistic goals for safety improvement. Engaging senior leaders at each facility gave us greater reach into operational efficiencies and helped achieve results.

Outcomes and Impact. The outcome of this work was a clear and lasting increase in the number of “A” hospital safety grades for Intermountain Healthcare. From 2016, when we received 2 A’s, 4 B’s, and 4 C safety grades, we increased the A’s and eliminated all C grades in 2021. The collaborative work performed by this team has objectively improved our CPOE decision support and helped reduce safety errors. Per Leapfrog’s scoring system, Intermountain hospitals increased from 85% to 94% review of inpatient medications via CPOE. CPOE scores for sensitive and specific flagging of potentially harmful errors increased, on average, from 46.8% to 54.9%, where 68% represents a perfect Leapfrog test score. All hospitals achieved full or substantial achievement ratings. More importantly, our self-reported percentage of medication order errors out of total medication events consistently dropped from 37.3% in 2018 to 31.7% in 2019 to 27.6% in 2020. Highly reliable use of BCMA during medication administration steadily improved during the same period. In 2018, the average system BCMA compliance score was 96%, but 2 hospitals had rates below 95%—suggesting that nursing did not yet have consistent practices at these facilities. By 2020, all hospitals were performing at 97% or better and the system average rose to 98% compliance overall. A system approach to streamline data collection/submission with Vizient templates; decision support work to improve CPOE and BCMA compliance, in collaboration with pharmacy and medical Informatics; and use of Lean Six Sigma methods, combined with strong backing by our hospital administrative teams resulted in a steady, incremental improvement in hospital compliance. Ultimately, our patient care has improved and patients are safer as a result of the work done by this team. We hope others can follow our model to reduce adverse medication events and improve publicly reported outcomes.

It’s the Journey That Matters: A Robust, End-of-Life Care Process to Improve “Observed” Mortality

Sathya Vijayakumar, MS, MBA; Kearstin Jorgenson, MSM, CPC, COC; Kory T. Anderson, MD

Intermountain Healthcare

Background. Vizient Clinical Data Base data has consistently shown that the greatest improvement opportunity for Intermountain Healthcare is in the mortality domain. Specifically, we focused on patients who enter the system through the emergency department (ED) and expire within the first 48 hours post-admission. This led the Quality and Intermountain Physician Advisory Services (IPAS) teams to come together and address these observed deaths. Three key issues stood out after an in-depth mortality review: (1) patients were inaccurately admitted to inpatient care despite being on comfort care measures only; (2) the hospice/palliative care team was not consulted early in the case; and (3) hospice care was not accurately captured. This data was shared with the hospice, care management (CM), utilization review (UR), and IPAS teams and a collaborative solution was developed to address the issues. To address the first 2 issues, our IPAS team partnered with the hospice, CM, and UR teams to develop a workflow to escalate patients meeting a certain set of criteria to a physician review. The reviewer would then follow a clear guideline indicating appropriate patient status and place the patient in the correct level of care. We defined 2 pathways for patients coming in from the ED versus critically ill inpatients. A concurrent physician review process was also implemented to ensure appropriate patient placement. A web-based database tool was developed using Research Electronic Data Capture to track at what points the process broke and where there was opportunity to improve. The second issue was addressed by partnering with hospice, CM, patient placement, registration, and health unit coordinators to develop an integrated workflow to ensure everyone did what they were supposed to do at the right time to capture the hospice disposition appropriately.

Intervention Detail. Two standard workflows were developed for end-of-life (EOL) care in collaboration with hospice, CM, IPAS, and UR teams for the ED versus acute care patients at the end of life. The process ensured that the patient and family were given all the options (aggressive intervention, palliative care, general inpatient care, or respite hospice or home on hospice) and were able to choose what they deemed best for them. The concurrent review process ensured the patient was in the appropriate status for the level of care chosen. A web-based tool was developed to track these patients while they were still in the system. When an EOL patient is entered in the tool by an ED care manager, it triggers an email to the next person in the process (in our case, the UR/IPAS and acute care managers) and alerts them to review the patient for appropriate patient status and level of care. The respective teams finish their reviews and document them in the tool. A daily huddle team, including representatives from each of these teams, was created to evaluate the level of care with clinical interventions and overall utilization management. This collaboration between IPAS and CM during the huddles worked effectively as an escalation pathway to physicians. This is a robust way to capture how many EOL care patients are being seen in our system and whether they are getting the right care. To ensure accurate capture of hospice discharges, we brought together leaders from patient placement, registration, CM, the nursing floor (health unit coordinator), and finance information systems to make sure the right labels were generated for hospice status and that the correct medical service was chosen in the electronic medical record. A workflow was created and shared with these teams to explicitly highlight each team’s role in this capture process.

Outcomes and Impact. The implementation of our novel workflows and the web-based tracking mechanism enabled the 4 biggest Intermountain hospitals to increase our discharges to hospice from the ED from 0.45% (January 2019 through April 2020) to 0.65% (May 2020 to November 2020). Concurrent reviews of EOL patients in the system led to the recategorization of 87 inpatient deaths because they were on comfort care only for the duration of Q3 2020 through Q4 2020. This has also helped improve our observed mortality by 0.26% (from 3.54% in Q1 2020 through Q2 2020 to 3.28% in Q3 2020 through Q4 2020). Of special mention is the marked improvement in early deaths (< 48 h), which dropped from 1.38% in Q1 2020 through Q2 2020 to 1.01% in Q3 2020 through Q4 2020. In the case of inaccurately captured hospice cases, we were able to retrospectively correct 23 cases for the Intermountain system in which EOL patients were counted as inpatient deaths, when they were really in hospice in 2019. The impact of the new hospice capture workflow kept 168 inaccurate patient deaths from affecting our inpatient mortality. This novel approach to address early deaths through implementation of a robust workflow and the engagement of a multidisciplinary team, including hospice, IPAS, CM, UR, registration, patient placement, and nursing units, has helped Intermountain Healthcare improve our observed mortality. We still have a long way to go to reach our destination in the top decile but we have already shown success, as demonstrated by our overall improvement in mortality ranking from the 38th percentile in Q4 2019 to the 69th percentile in Q3 2020.

Pharmacy Professional Development Practitioner: The Focus on Six = The Future of Pharmacy

Lisa Purdy, PharmD

IU Health—Riley Hospital for Children

Background. Pharmacy is adapting to play a larger role in the growing needs of today’s health care, including defining problems; assessing options; implementing solutions; and being relational, empathetic, and influential. At the same time, top-performing pharmacy clinicians are being promoted for good performance who may not necessarily have adequate leadership training. Pharmacy team members are expected to ensure an engaged, empowered, and sustainable workforce that drives retention. In order to push the envelope of what pharmacy can be in the future and elevate pharmacy team members to perform at the top of their license, we required a new role to emerge, one who can help guide team members in personal discovery and advancement. Modeled after the nursing professional development practitioner, the pharmacy professional development practitioner (PPDP) expands on the pharmacy educator role to cover 6 keys areas: (1) role development; (2) collaborative partnership; (3) research/quality improvement/evidence-based practice; (4) orientation/onboarding; (5) competency management; and (6) education. The role of a PPDP is multifaceted and includes serving as a leader, change agent, mentor, learning facilitator, champion for inquiry, advocate for the PPDP specialty, and partner for practice transitions. The PPDP promotes the clinical learner and the organization through professional staff development programs that support the needs of the organization, beginning with newly hired clinical employees and continuing through the expert clinician.

Intervention Detail. The PPDP focuses on 6 key areas outlined in the “Background” section. Through regular use of environmental scanning, the PPDP identified the department’s needs related to these 6 areas by assessing the practice gaps, as well as staying abreast of important quality outcomes and opportunities for improvement. The PPDP grows staff members personally and professionally through regular one-on-one meetings, career discussions, and individual development with the use of professional development pathways. The PPDP established an informal leader program, a department mentorship program, and created space for succession planning. By focusing on new hire orientation and onboarding, the PPDP established consistent quality training and increased employee retention. Through collaborative partnership, the PPDP focused on establishing connections between the pharmacy, pharmacy schools, and technician training programs to provide enhanced experiences for all learners. The PPDP is uniquely suited to establish and support pharmacy shared governance and related steering committees and coordinate between campuses, health care units, and schools. As part of the research portion, the PPDP fostered a spirit of inquiry through one-on-one check-ins and provided resources for evidence-based practice design. As a natural change agent, this role contributes to building a sustainable professional culture.

Outcomes and Impact. Since establishing this role, we have seen a statistically significant increase in our employee engagement results in 8 of 12 questions asked related to this role. The PPDP worked with pharmacy leadership to establish a cadence of reviewing technicians for promotion and within 9 months, we had 15 promotions. Within 2 months of starting the role development pathways, 47% of the staff started a pathway, gaining 8 new certifications, and 33 staff members entered the informal leadership program. Fast forward 6 months, and 80% of the department is now on role development pathways with 26 certifications in process, with many about to be completed. In a little over a year, we have realized a decrease in first-year turnover. We further anticipate increases in publications and professional organizational activity. The creation of a shared governance council will establish a voice for our frontline care givers, empowering them to make continued improvements in our department while enhancing communication across the department and the organization.

Empowering, Engaging, and Educating Frontline Clinic Staff in Continuous Quality Improvement

Saira N. Huggins, MS

Johns Hopkins Health System

Background. Ambulatory clinics in large health care systems may not always have defined resources for continuous quality improvement; therefore, most of the responsibility to improve outcomes and prevent harm falls to the nurse manager and/or operations manager. These managers are often focused on supporting daily clinic operations and may not always have sufficient time to ensure the systems in place to improve outcomes and/or prevent harm are sustained. Although high-reliability organizations recognize those closest to the work are the most knowledgeable, the traditional clinic hierarchy may not always see frontline staff as the experts. This may be because the frontline staff are not empowered, engaged, or educated in continuous quality improvement. Creating a continuous educational program for those frontline staff members who are not in formal leadership roles not only improved our performance in quality, patient safety, and patient experience metrics, but it strengthened the overall organizational culture within the local clinic setting. This is an important part of the 2021 Vizient Connections Summit because many health care leaders do not see the value of investing in frontline staff specifically to improve outcomes, prevent harm, and ensure every patient has the ideal experience. Executive leadership also may not realize the amount of responsibility that is already on the shoulders of the local leadership. Shifting the responsibility by empowering, engaging, and educating those closest to the work can lead to improvement and resiliency.

Intervention Detail. The first version of the program included 1 introductory session and continuing education sessions, as requested by participants. The 8-hour introductory session was open to all (no application necessary) and laid the foundation for quality, patient safety, and patient experience. The first 4 hours were spent discussing structure and metric definitions. The second half of the day was spent in a computer lab introducing the available resources (eg, dashboards, data, quality improvement tools). A post-survey was administered that asked participants what else they would like to learn about, and continuing education sessions (patient experience and Plan-Do-Study-Act [PDSA]) were developed and offered based on the responses. However, not all introductory session participants attended the continuing education sessions. While attendance at the introductory session was high, this first version lacked accountability, structure, and continuous maintenance. The second version of the program included: (1) an application process; (2) five 4-hour sessions on a rotating schedule (based on the PDSA methodology); (3) open office hours with leaders; (4) 1-hour monthly meetings for sharing of best practices and “barrier-busting”; (5) quarterly progress checks sent to managers; (6) the expectation to present at the end-of-the-year summit; and (7) defined, continuous responsibilities after 1 full year of the program. A pre- and post-survey was administered for each session, feedback was reviewed routinely by program leaders, and the enterprise-wide strategic objectives were analyzed on a quarterly basis to monitor the program’s impact. Also, a clinic-specific project was identified for each participant based on feedback from the clinic team and manager. The participant was expected to collect baseline data and define the key metric that would be used to measure improvement. Almost every project was trending toward improvement.

Outcomes and Impact. During the first version of the program, 35 participants attended the introductory session, 7 attended the patient experience session, and 15 attended the PDSA session. In the post-survey, 96% of participants stated they were interested in attending other sessions, but they would like to dive deeper into topics, have more time to navigate available resources, and a slower delivery. The enterprise-wide strategic objective performance was compared. Entity A improved by 2 points, while Entities B and C stayed the same. Ten participants from the first version of the program continued into the second version of the program and 17 new participants joined, for a total of 27 participants in the enhanced program. The enterprise-wide strategic objectives were not able to be used as a performance metric, but all program participant projects were trending toward improvement. Even without quantitative data, the organizational culture has shown improvement as the participants are actively involved in initiatives that managers historically took ownership of, such as the implementation of huddle boards.

Conquer Pharmacy Technician Recruitment Challenges: Establish an Innovative, Subsidized Workforce Pipeline

Karen Nolan, RPh

Lifespan—Rhode Island Hospital

Background. Working with a reduced workforce places increased stress on current team members who are expected to complete the same amount of work with fewer employees. A 1997 study found that 60% of employee absences are related to job stress, with approximately 225 million working days lost annually in the United States from stress-related absenteeism.¹ The Society for Human Resource Management estimated the average cost to hire an employee is over $4000, with an average of 42 days to fill a position.² In August 2021, the US Bureau of Labor Statistics identified 5.4 million job openings throughout the United States (a significant increase from 2.7 million in 2010), with many individuals unable to maintain steady employment because they lack the skills necessary for success in the workplace.³ Implementing an innovative, demand-driven solution, such as an employer-led workforce pipeline that connects the employer’s needs with education/training programs, can yield a stronger, more diverse pharmacy technician workforce. For the pipeline to be successful, the educational partner(s) must have a profound understanding of pharmacy; a strong education track record; soft skills training; and a flexible, evolving curriculum to match the employer’s changing needs—subsequent to advancements related to technology and/or the role of the pharmacy technician. Supplementing the didactic training with a hands-on externship elevates the training program, engages team members, and provides the employer a risk-free opportunity to observe key qualities (eg, work ethics, dependability, team skills, communication, critical thinking) of potential employee candidates (ie, externs). A high-quality pharmacy technician extern training program increases employee engagement by empowering team members to be mentors and actively participate in the recruitment process. Motivated, engaged team members can ultimately lead to increased employee retention and increased productivity.

Intervention Detail. Running an organization while understaffed can lead to high turnover and exacerbate stressors on an already overburdened workforce. This scenario can have serious effects on team health and well-being, patient safety and quality, and productivity and the financial bottom line. Building the right team requires 2 crucial prongs—getting the right individuals on board and retaining an engaged workforce. To successfully solve both challenges, pharmacy services at Lifespan, a comprehensive integrated academic health system, partnered with Lifespan Workforce Development, Community College of Rhode Island, and Rhode Island Department of Labor to establish a state-funded pharmacy technician training program designed to simultaneously strengthen our inpatient and outpatient pharmacy labor force. The program transforms unemployed/underemployed, entry-level Rhode Islanders into well-qualified, nationally certified, health care professionals. The pharmacy extern training program is 16 weeks and includes 180 hours of didactic learning, 120 hours of hands-on pharmacy experience at a Lifespan affiliate, and 42 hours of workplace readiness and leadership training necessary to obtain and sustain employment, as well as demonstrated progress along a management-driven career pathway. State grant funding covers all financial aspects of the training program, including expenses related to obtaining state licensure and national certification, as well as support services, such as childcare and transportation assistance. This helps remove obstacles so that all Rhode Islanders have equal opportunity to take advantage of the program. Increased demand for prescription medications and expanded pharmacy technician responsibilities are 2 major contributors to the current and projected increased demand for pharmacy technicians. Health care systems nationwide are experiencing significant effects related to pharmacy technician workforce shortages, necessitating innovative solutions such as building a subsidized workforce pipeline.

Outcomes and Impact. Since implementation in 2019, we have had 5 pharmacy extern cohorts with over 750 applicants, 126 students accepted into the training program (cohort class size has grown by 33% since conception), and 66.7% successfully graduating—of which 75% achieved national certification. Lifespan hired 40 externs as full-time pharmacy technicians across our affiliates. Technicians hired after graduating from the training program acclimate quickly to pharmacy operations, instantaneously have mentorship support from frontline team members as a result of connections developed during their externship, and were able to proactively assess that the position is a good fit (due to transparent, first-hand insight into the position), further increasing the potential for success. Frontline team members now have a more pronounced voice in the recruitment process and previous extern graduates take greater interest in mentoring current and future externs, helping them navigate their way through the training program, prepare for the national certification exam, and learn the day-to-day responsibilities of a pharmacy technician. Research shows that when team members feel empowered and have a sense of ownership in their job, engagement and retention increase—which is exactly what we are witnessing with our team. Additionally, because team members assist with coaching, training, and selecting externs to potentially join our group, they have a special interest in their success—resulting in increased team bonding, compassion, and respect. Establishing a pharmacy technician workforce pipeline without financial obligation is an innovative solution that health care organizations, regardless of size, can implement to resolve recruitment challenges, intensify team engagement, increase employee retention, and broaden the diversity of team members. Team diversity potentially diminishes disparities between health care workers and the patient populations they serve by tapping into the unemployed/underemployed populations within your state.

1. Cartwright S. Taking the pulse of executive health in the U.K. [and executive commentary]. Acad Manag Perspect. 2000;14:16–24.

2. Northon L, Dooney J, Esen E, et al. 2016 Human Capital Benchmarking Report. Society for Human Resource Management. 2016. Accessed January 28, 2022. https://www.shrm.org/hr-today/trends-and-forecasting/research-and-surveys/documents/2016-human-capital-report.pdf.

3. FRED. Total Unfilled Job Vacancies for the United States. 2021. Accessed January 28, 2022. https://fred.stlouisfed.org/series/LMJVTTUVUSM647S.

Utilizing Vizient Intelligence to Compare Use of Antimicrobial Agents

Khadeja Kausar, MHA; Monica Ghitan, MD; Sam Simon, PharmD, AAHIVP; Suri Mayer, PharmD; Rosanna Li, PharmD BCIDP; William Doxey, PA-C, MBA; Robert A. Press, MD, PhD

Maimonides Medical Center

Background. Maimonides Medical Center (MMC) is a 711-bed hospital that serves a core service area of 1.1 million people with over 43 000 discharges annually, including a higher percentage of older adult patients and patients with a higher severity of illness. MMC uses an Antimicrobial Stewardship Program (ASP) to promote the appropriate use of antibiotics—a requirement that several state and federal institutions monitor. MMC’s program interventions are designed to improve and measure the appropriate use of antibiotics, optimize clinical outcomes, minimize toxicity/adverse drug reaction, and limit the emergence of antimicrobial resistance. MMC’s ASP consists of infectious diseases pharmacists and infectious diseases physicians. The ASP reports to the Antimicrobial Stewardship Committee, a subcommittee of the Pharmacy and Therapeutics Committee. Using a combination of persuasive, restrictive, and structural interventions, the ASP implemented an authorization process and provides information to prescribers, creates education programs and audits, and restricts formulary. While the ASP successfully reduced the mean days of use for several antibiotics, this presentation’s focus is to illustrate a reduction in usage of ceftriaxone and meropenem for patients with community-acquired pneumonia or urinary tract infection. Comparing the first quarters of 2018 and 2021, the ASP’s intervention successfully reduced the mean days of use for meropenem and the percentage of the clinical population that received the antibiotic. The comparator hospitals also saw a small reduction in mean days of resource use. However, the percentage of the clinical population that received meropenem was slightly higher. In the case of ceftriaxone, the percentage of the population prescribed ceftriaxone increased at MMC from 2018 to 2020. However, focused interventions had a successful impact in 2021.

Intervention Detail. The ASP provides information to prescribers, creates education programs and audits, restricts formulary, and uses an authorization process. Efforts to reduce overutilization include transitioning to an oral antibiotic early in the disease course when appropriate; improving clinician communication and education through ASP rounds, prospective audit, and feedback; standardizing infectious disease curbside consult; and increasing access to MMC guidelines via a smartphone app and the MMC intranet website. Usage and outcomes data for patients with a diagnosis of community-acquired pneumonia or urinary tract infection who were prescribed meropenem and ceftriaxone were abstracted from the Vizient Clinical Data Base for the first quarters of 2018 through 2021 for MMC and comparator hospitals. Additional variables, such as mean days of resource use and percentage of clinical population prescribed the previously mentioned agents, were calculated using built-in functionality of resource usage in the Vizient Clinical Data Base for both MMC and comparator hospitals.

Outcomes and Impact. The MMC ASP successfully reduced the mean days of use for several antibiotics. The mean days of use for meropenem declined consistently from 7.79 days in 2018 to 6.16 days in 2019 to 5.53 days in 2020. There was a slight increase in 2021 to 5.79 days that the ASP team will continue to monitor. The primary ASP interventions that had the most impact were internal audits, mandatory consult for empiric use, prior authorization, and real-time usage review. The comparator hospitals’ prescription of meropenem increased from 5.84 days in 2018 to 6.72 days in 2021. In 2021 MMC faired better in mean days of resource use by approximately 1 day compared to the comparator hospitals. For the comparator hospitals, the percentage of the clinical population prescribed meropenem remained under 7.88% for all 4 years of the study. For ceftriaxone, even though mean days of resource use did not have much variation, the percentage of the clinical population prescribed ceftriaxone increased from 65.76% in 2018 to 77.22% in 2020. The ASP revised the guidelines, increased the review of positive blood cultures to every 2 hours (7 reports a day), changed to a narrower spectrum and oral when possible, and provided dosing assistance. This resulted in a reduction of the clinical population prescribed ceftriaxone to 65.23% in 2021. The ACP team will continue to review opportunities for improvement to further decrease this percentage. The comparator hospitals prescribe ceftriaxone to 45.38% of the population, a steady decline from 68.11% in 2018.

Comprehensive Care Teams Provide Communication Innovation During COVID-19

Janice Finder, MSN, RN, MD, Anderson Cancer Center; Michael Frumovitz, MD, MPH, MD, Anderson Cancer Center; Judy Overton, MLA, MD, Anderson Cancer Center; Ashlyn A. Proske, BS, The University of Texas MD Anderson Cancer Center

MD Anderson Cancer Center

Background. Transitions in care can be a complicated process for cancer patients. Comprehensive programs are needed to help these patients and their families engage at multiple levels. Patients and families may call with questions about visitation policies, which an operator can answer. The next call may be more serious in nature, such as a patient with a fever who needs help deciding whether they should go to the hospital. This abstract is our comprehensive call center’s single-site study of how best to address all facets of our patients’ needs to improve their experience, as well as decrease unnecessary readmissions. askMDA is a comprehensive call center that is available to patients and caregivers 24/7 to improve the patient experience. Patients played an integral part in the design of this program. The following is a brief description of the 3 levels of service within the program. Operators comprise the first level of service, providing the most updated facility information; COVID-19 restrictions, testing sites, and vaccine information; and the link for patient and caregiver needs and concerns. Health information specialists form the second level of service by sharing cancer protocol education, referring provider and self-referral intakes, electronic medical record assistance, and detailed COVID-19 testing and vaccine information. The third level of service is our clinical team, which offers immediate symptom management and care escalation, when necessary; has prescriptive authority; and provides education and resolution of clinical cares and concerns. During the COVID-19 pandemic, askMDA’s clinical team was available 24/7 to field calls regarding complexity of clinical presentation, course of illness, and limited provider-based resources. The narrative will also include measurable outcomes and data analysis and conclude with future needs for increased levels of care and patient satisfaction.

Intervention Detail. askMDA operators triage calls and promptly refer them to 1 of 3 levels of service. Operators in the first level handle over 1 million calls per year, which range from simple to complex, from providing directions to a clinic to responding to cardiac arrests. Operators stay informed of the latest information, so patients and caregivers feel confident during these interactions. Operators are known to stay on the line until a lost patient reaches their intended destination. In the next level of service, Health Information Team and Referring Provider Team members walk patients and referring physicians through a referral, discuss research protocols, and estimate the date of the first appointment. They provide a critical information link to patients and caregivers for all transitions in care, handling over 400 000 calls and over 10 000 referrals annually. Two new clinical programs, Discharge Phone Calls and After-Hours Clinical Support, comprise the third level of service. Discharge Phone Calls supports inquiries about inpatient, outpatient, and emergency center visits, while After Hours Clinical Support helps patients and on-call physicians after business hours. In addition to being skilled in telephone triage, oncology nurses need to know about disease processes and clinical patient care when the patient transitions from the inpatient to the outpatient setting. After Hours Clinical Support enables patients to speak with a nurse rather than page a provider and wait for a call back. Nurses manage patient issues that are within their scope; however, they will refer to the provider when necessary. Research shows these activities elevate the quality of care. They also lower the costs of care to an organization, protect the capacity of emergency centers, and allow hospitals to care for their most critical patients.

Outcomes and Impact. In a retrospective review, the call center received nearly 300 000 requests for assistance during a 90-day period in spring 2020. The clinical team received 3600 calls for symptom triage support. Health information specialists answered more than 41 000 requests for new treatment, status of treatment, electronic medical record and video visit assistance, and COVID-19 screening site appointments. The operators handled more than 250 000 questions about discharge times, waiting areas, locations for pick-up, and whether centers and entrances were opened or closed. Eighty percent of the patients responded to the call center nurses’ advice to go to the emergency center. Of those patients, 70% were admitted to the hospital. Call center nurses frequently review these statistics to ensure their recommendations are appropriate. Further study is necessary to determine what is needed for the 20% who did not present to the emergency center as advised or how to manage their symptoms using other modalities. Readmissions decreased from 20% to as low as 12.5%, remaining at 15%. In addition, Press Ganey patient satisfaction survey results improved 1% in the top box score and 2% in the peer ranking. The clinical team managed 50% of the calls without paging the physician on call. The decreased number of pages garnered positive comments from physicians on call and from patients who were happy to speak with a nurse. The COVID-19 pandemic set the stage for all hospitals to prepare for the evolving needs of the community. askMDA provides the right information at the right time in a cost-effective manner. The immediate responsiveness and accessibility of this service will continue to help patients adhere to care plans, which will drive down unnecessary visits to the emergency center; increase necessary visits to the emergency center; and improve the experience of patients, family members, and staff. The askMDAnderson project team: Elizabeth Garcia, Ashlyn Proske, Tamara Barnes, Veronica Smith, Albert Guerrero, Laura Moore, Wesley Amistad, Jessica Baulis, Tiffany Boutte, Kelly Brassil, Kristine Del Rosario, Shinu John, Liliana Larsson, Alexis Olivares, Vanezza Tolentino, Torie Thibodeaux, and Elizabeth Trejo.

Using Vizient Predictive Ranking to Develop an Improvement Project Portfolio

John D. Calhoun, MS; Bela R. Patel, MD, FCCP, CMQ, FCCM; Thanh Dao, MS

Memorial Hermann-Texas Medical Center

Background. The Vizient peer group ranking methodology continues to improve over time and details are shared openly with Vizient members. A benefit of this transparency is that members have an opportunity to replicate the ranking process as frequently as their locally available resources allow. Doing so enables the organizations to complete “what if” scenarios while being fairly accurate in estimating the impact on final rankings. The organization can determine where it has the most opportunity for improvement while also prioritizing projects in accordance with relative impact on year-end Vizient rankings. This approach can be used to generate both reasonable peer-to-peer ranking values and trends of organizational performance over time, month-to-month. The benefit is that peer-to-peer comparisons are effective in driving improvement results when there is high confidence in the ranking methodology. On the other hand, internal trending of performance over time is an easier sell to providers—particularly as complexity of the assessment approach increases. Complexity in the ranking methodology tends to create opportunities for stakeholders to dismiss poor performance as an effect of the assessment method rather than accepting poor performance as a true improvement opportunity. We found that the use of internal trending, even when done relative to overall peer group performance indicators, results in a higher organizational acceptance of project recommendations. With that in mind the Memorial Hermann-Texas Medical Center (MH TMC) team chose to trend the performance rankings of its service lines not only in comparison to the prior year’s peer group performance but also relative to each prior month’s service line performance. The team has used this approach the past several years and has shown it to be quite effective in producing end-of-year overall ranking predictions that are within plus or minus 2 points of the final Vizient rankings.

Intervention Detail. Completing the monthly update cycle from submission of new performance data to calculation of metric values, ranking results, and then opportunity identification is accomplished via a sequence of standard process steps. The team is required to: (1) replicate the Vizient Quality and Accountability calculator and Vizient standard report templates; (2) use the output created to develop an executive summary of service line performance by displaying prior year final performance rank next to the organization’s current year-to-date performance rank; (3) color code the results for ease of use in detecting changes in rank; (4) designate areas having a monthly rank increase as improvement opportunities; (5) review current projects in the opportunity area to ensure they are on target; (6) make adjustments as deemed appropriate; (7) if no project is in place, designate the area as a project opportunity; and (8) initiate a formal project effort to drive improvement. The MH TMC team continues to refine the overall process. When using predictive rankings, the team has noted that the overall process is well received by administrative and physician leadership because it is risk-adjusted and benchmarks against true peers—including peers at the service line-level and performance trending based on the organization’s own monthly performance.

Outcomes and Impact. Use of the predictive ranking approach over the last 5 years has strengthened MH TMC’s ability to align available resources with projects having the most impact on total care quality, where quality is correlated with Vizient ranking. In a typical year, the approach yields 3 or more project opportunities per domain. Taken together, these projects comprise our Vizient project portfolio for the year. On average, 40% of the projects in the portfolio drive improvements in administratively documenting the care and outcomes we deliver, while the remaining 60% drive either maintenance of current operational performance or improvements in areas of opportunity. An additional outcome is that the project portfolio is adjusted throughout the year in response to changes in predictive rank. Some projects are completed quickly and drop from the portfolio, while others are added in response to new opportunities. The positive impact of the approach on both our quality and our organization as a whole is best demonstrated by changes in our Vizient ranking over the years. MH TMC moved from a rank of 27 among academic medical center peers in 2006 to number 20 in 2010, number 12 in 2014, and then to a top 10 ranking in 2018 and 2019. In 2020, we reached the number 5 position—our highest ranking ever. The positive trend in quality and the correlated improvement in Vizient ranking is ultimately attributable to the commitments of our clinical and administrative teams to excellent patient care. In support of these commitments, these team members are guided in their ongoing performance improvement activities using the Vizient predictive ranking methodology.

We Don’t Miss a Beat: Reducing Care Variation in Acute Myocardial Infarction Patients

Douglas Obogo, MPH, MSc, Data Science & Analytics; Holly M. S. Daily, BHS, RRT, RRT-ACCS; Melissa Dowler, RN, MSN, CV-BC

MU Health Care

Background. The Vizient Quality and Accountability Study ranks academic medical centers (AMCs) based on performance in key domains of quality for select disease conditions. In 2019, acute myocardial infarction (AMI) was identified as one of the disease conditions with a great potential to impact MU Health Care’s Vizient ranking. At the beginning of calendar year 2019, MU Health Care (MUHC) was ranked 94 out of 99 in mortality, 89 out of 99 in readmissions, and 43 out of 99 in length of stay (LOS). This performance prompted the implementation of a comprehensive quality improvement program with the goal of achieving top quartile by the end of 2020. To guide improvement work, an evidence-based improvement framework was utilized. The AMI team opted to follow the Institute for Healthcare Improvement’s Model for Improvement, which includes the following steps:

• 1. Forming the team: Executive leadership formed a team tasked with improving outcomes that reported its progress biweekly to the executive steering committee. The core team included an executive sponsor, a medical director, an administrative leader, quality improvement professionals, a data analyst, and other ad hoc staff.
• 2. Setting the aim: The team aimed to achieve the top quartile in Vizient AMC rankings in the mortality, readmissions, and LOS domains by the end of 2020.
• 3. Establishing measures: To measure if a change led to an improvement, the team elected to use the Vizient Quality and Accountability Study to benchmark its performance against other AMCs in the Vizient Clinical Data Base.

Step 4 continues in the next section.

Intervention Detail. The fourth step was implementing changes(s), which began with a current state assessment that identified the following as key drivers: care variation, discharge planning, and miscoding of non-AMI cases into AMI diagnosis-related groups. The team decided to standardize the acute care episode, from inpatient admission to postdischarge. First, the team opted to standardize by adhering to evidence-based AMI treatment guidelines from the American Heart Association. Next, the team determined what resources (called “enablers”) were necessary to help providers translate AMI clinical guidelines into practice. The first enabler is the AMI care plan, which was standardized and had the ability to be embedded as a subplan within the admission order set. The second enabler is the inpatient nurse navigator, who meets with patients and ensures they are placed on the AMI care plan. Any patient not on the care plan is flagged for department leaders, who follow-up with the ordering provider. Nurse navigators also enable appropriate follow-up by ensuring patients have follow-up appointments scheduled prior to discharge. They call patients 1 to 3 days postdischarge to coordinate referrals and reinforce discharge education. Enablers were developed for the established nurse navigators. MUHC employs a readmission risk calculator to facilitate the work of nurse navigators. When an AMI patient is admitted, this tool generates custom readmission risk scores that are used to facilitate timely follow-up visits. To solve the issue of miscoding of patients into the AMI diagnosis-related group, cardiac unit coders worked with physicians to improve documentation and coding accuracy. Finally, to foster and maintain learning into practice, a multidisciplinary readmission and mortality case review team was established to identify improvement opportunities with each patient. The goal is to determine “what could be done differently if it could be done all over again?”

Outcomes and Impact. MUHC reached its goal of achieving top quartile by the end of 2020. Pre-intervention, the observed-to-expected mortality ratio was 1.4; post-intervention, this decreased to 0.62, representing a 57% reduction. For readmission and LOS, there was a 63% and a 19% reduction, respectively. Benchmarking against other AMCs in the Vizient Clinical Data Base, MUHC advanced 87 places and ranked 2 out of 99 in the 30-day unplanned readmission rate. For mortality, MUHC advanced 88 places and ranked 6 out of 99.

Using Teletriage in Emergency Medicine to Improve Patient Care and Provider Coverage

Jeffrey S. Kuo, MD, FACEP, FAAEM; Stephanie Anderson, MSN-HCSM, RN; Nicole Vegel, MPH; Lauren B. DeFrates, MS

Ochsner Health

Background. Emergency departments (EDs) nationwide experience delays in patient care due to throughput and capacity constraints (ie, space, resources, nursing, bed placement). Patients designated Emergency Severity Index level 2’s and 3’s frequently experience lengthy wait times before seeing a provider. These patients typically require labs and imaging studies ordered by the provider after initial evaluation—a delay that leads to increased length of stay (LOS) and door-to-provider time, as well as the potential of poor outcomes and decreased provider/patient experience. Ochsner Health is not immune to this. As a solution, we created the Ochsner Health ED TeleTriage program. Leadership accountability was established through 2 new roles, the TeleTriage program lead and ED telehealth medical director. Both function as program managers and understand the needs for both the physical and remote EDs. Accountability is accomplished through end-user feedback and tracking of real-time program metrics. Door-to-provider times, patient LOS, and TeleTriage utilization and provider ordering patterns are all captured and analyzed to ensure program success. Periodically, surveys are sent to nursing staff to better understand their perception and satisfaction of the program. The future of emergency medicine expands beyond the traditional brick-and-mortar facility. It requires innovative workflows and appropriate staffing to meet patient volume trends. Especially during the COVID-19 pandemic, TeleTriage proved helpful in providing a system-based coverage approach to EDs using a bunkered provider model. The TeleTriage program aligns with the purpose of the Vizient Connections Summit. We created an actionable solution that addresses current patient flow challenges in our ED. In addition, we improved door-to-provider times and drove innovations across the cost, quality, and market performance continuum.

Intervention Detail. Ochsner ED metrics identified challenges with increased capacity constraints, throughput, LOS, and anticipated winter surge rates. These challenges cause the provider/patient experience to suffer and put patients at risk for poor outcomes. Current state metrics helped determine where to focus efforts. Exploring new models of best practice led to the development and implementation of TeleTriage. Lessons were learned from both MedStar Washington Hospital Center and Jefferson Health. Electronic medical record dashboards were developed to optimize workflow and continually monitor the impact and efficiency of the TeleTriage program. The TeleTriage workflow was developed in collaboration with nursing, physicians, advanced practice providers, information systems, telemedicine, and ED operations. Ten ED sites were identified to pilot the program. A new TeleTriage platform and tracking board was developed in the electronic medical record. Three provider bunkers were installed across the system to allow for remote TeleTriage coverage, and telemedicine tablets were deployed to the pilot sites. Education and training were delivered to both the provider groups and clinical staff at each site prior to go-live. Long-term staffing and leadership models were created to support the program’s longevity and success. The Ochsner Health ED TeleTriage program initiates necessary patient workups sooner in the patient journey by implementing a provider-in-triage model via a virtual provider. The TeleTriage provider, normally a physician assistant or nurse practitioner, evaluates patients in triage and orders necessary labs, imaging studies, and medications from a remote bunker.

Outcomes and Impact. The program has decreased door-to-provider times, overall ED LOS, and left-without-being-seen rates during surge and peak volume times. Additionally, we realized approximately $528 000 in annual provider cost savings for Ochsner Health—the result of consolidated resource utilization, throughput improvement, and cumulative capacity improvements. Nursing staff has noted that the program has helped greatly during surge times and that patients appreciate seeing a provider much sooner in their ED journey.

Regional ECMO Consortium Development During COVID-19 Saved Lives

David Zonies, MD, MPH, MBA, Oregon Health & Science University; Jenelle Badulak, MD, University of Washington

Oregon Health & Science University

Background. Extracorporeal membrane oxygenation (ECMO) is labor intensive and a scarce critical care resource. Data from Asia and subsequently Europe documented that more than half of the patients with COVID-19 required critical care, 40% required invasive mechanical ventilation, and 11% required ECMO.¹ While ECMO is typically indicated in < 1% of critically ill patients,² hospital team directors at 2 major academic medical centers were concerned that uncoordinated referrals, regional surges, and lack of real-time access availability would endanger patient lives. Additionally, should crisis standards of care be required, ECMO would be one of the first resources to be depleted. Great concern existed about ethical allocation, the resources actually available in the region, and capability (both in terms of capacity and teams able to perform the treatment). Reports of required ECMO were 10-fold higher than anticipated during the pandemic, and compounding the limited ECMO bed availability was the fact that a major supply chain disruption occurred because major ECMO manufacturing was locked down in Asia and Europe.¹ ECMO directors wanted to ensure that any patient requiring treatment would have timely referral to the closest center possible. We also wanted to make certain that allocation would be equitable and ethically sound. This required vetting of inclusion and exclusion criteria for patients when ECMO became a resource-limited therapy. Accountability was established with each hospital’s chief medical officer and regional health authorities were updated in a regular cadence. As a learning health system, lessons during this crisis may serve as a model of cross collaboration for Vizient members and all hospitals in the future.

Intervention Detail. The ECMO directors from the 2 major academic medical centers in Oregon (Oregon Health & Science University) and Washington (University of Washington) initially held a regional conference call with other known ECMO centers in February 2020. This initiative rapidly grew over 6 weeks to include 11 ECMO centers treating both adults and children in Oregon, Washington, and Idaho. The stated goal was to collaborate on protocol development (inclusion/exclusion criteria) should we reach a crisis standard, to assure that any patient would be treated in a similar fashion in any of our centers. Additionally, a supply chain registry was established to facilitate equipment-sharing if required. This was an important feature, as it involved 9 additional health systems, both public and private. Virtual meetings were held biweekly and then transitioned to a monthly check-in once steady state was reached. Major collaborative products included a literature review and best practice document on an ethical approach for inclusion and exclusion criteria of a scarce resource. Through a modified Delphi method, all centers agreed on the principles of inclusion and management of this unique patient population. A web-based platform was established to create a continuously updated dashboard for center capacity. This, in turn, allowed for the rapid referral of patients in need of a life-saving intervention when 1 region became overwhelmed. In some cases, this data was then fed into a local resource hospital system, allowing regional operation centers to maintain visibility.

Outcomes and Impact. A grassroots effort across multiple health systems allowed the Pacific Northwest region to develop a robust ECMO network. Between March 2020 and February 2021, 250 ECMO patients were managed across the network. One-hundred twenty-six (50%) of those patients were treated specifically for severe COVID-19 pneumonia (center variance 14% to 74%). A combined total of 29 adults and 11 children could be simultaneously managed at any given moment. The program was so successful that the rate-limiting step was ultimately the ability to have appropriately trained critical care ECMO specialist nurses to staff the beds. The average number of ECMO devices per center was 4. As COVID-19 patient surges occurred at a local level, the network facilitated rapid referral to the next-nearest available center. The result was improved opportunity for survival for the most critically ill patients in our collective region. When patients were too critically ill to be transferred to a center, 2 systems were able to forward deploy teams to evaluate and place patients on ECMO support at peripheral hospitals. Two additional teams were deployed to retrieve patients already on support but who required an available bed. This increased the catchment beyond the 3-state collaborative to include regional capability of 5 states, including Alaska. During the pandemic, many systems worked toward regional medical operations centers, with the goal of resource allocation to optimize patient care delivery. This super-specialized and scarce resource was delivered to more patients because of innovative collaboration that crossed multiple health systems in multiple states. This will be an enduring relationship after the pandemic abates.

1. Grasselli G, Pesenti A, Cecconi M. Critical care utilization for the COVID-19 outbreak in Lombardy, Italy: early experience and forecast during an emergency response. JAMA. 2020;323:1545–1546.

2. Combes A, Brodie D, Bartlett R, et al. Position paper for the organization of extracorporeal membrane oxygenation programs for acute respiratory failure in adult patients. Am J Respir Crit Care Med. 2014;190:488–496.

Listening in Medicine: Storytelling and the Future of the Patient Experience

Stephanie Kindt, MA, JD, Penn Medicine; Aaron Levy, PhD, University of Pennsylvania

Penn Medicine

Background. In October 2019, the Penn Medicine Listening Lab was launched under the sponsorship of the Penn Medicine Experience Leadership Team. The Listening Lab was designed to gather and share stories about the importance of being present, nurturing compassionate relationships, and empowering communities—standards at the heart of the Penn Medicine experience. Leadership support during the launch and subsequent communication of the project across the system has been critical. Project goals are set by a systemwide project team and led by the senior consultant for the patient experience, who has a systemwide role and is accountable to the associate vice president for patient experience. The voices of patients, providers, staff, and family members were critical to design and execution of the Listening Lab. Patient and family advisory councils from each Penn Medicine entity were consulted and many of the initial seed stories came from this group of key stakeholders. These are incredibly challenging and difficult times—especially for health care and essential workers; those raising children while quarantined at home; those caring for parents or loved ones; those managing illness alone; and anyone dealing with housing instability, food insecurity, unemployment, and/or the many other social, economic, and political vulnerabilities affecting our communities. The Listening Lab was launched just as the COVID-19 pandemic began, and the lab has become a resilience-building tool for our staff, patients, and the community. How we care for and listen to each other has never been more important. The aim of the project is to act as a catalyst for conversations that lead to healing. It is for these reasons that it is important to include work like this in the Vizient Connections Summit.

Intervention Detail. A 2018 study found that clinicians elicit the patient’s agenda in only 36% of the encounters studied. In 67% of those encounters, the clinician interrupted the patient after a median of 11 seconds.¹ And in a 2011 study in the journal Health Affairs, Lown, Rosen, and Marttila found that it only requires 40 seconds for a health care provider to communicate with kindness and empathy, yet a majority of the patients studied reported having interactions with providers that lacked compassion.² A 2011 article by Seth Holmes and Maya Ponte charts 1 possible cause of this breakdown in listening and compassion. Medical students are trained to transform the human subject patient into a recognizable, generic clinical case. As they are taught to translate the world of patient suffering into structured medical records, medical trainees are also being socialized to listen to the patient narrative in particular kinds of ways that typically exclude that which is chaotic, unruly, and unique.³ What is gained and lost in this translation process? The Listening Lab is influencing providers, nurses, staff, patients, and family members in their approach to listening as they both give and receive care. One provider shared that after listening to a story describing a colleague who changed their behaviors when asking required questions on depression and abuse, they too have altered their workflow to collect the same information in a very different way. What began as a single moment in an uncomfortable patient visit became an opportunity to build deeper connections and trust between patients and providers systemwide—simply by telling that story.

Outcomes and Impact. Early in the process, a patient and family advisory council member contributed a powerful story about a difficult experience she had while receiving care and the changes that occurred when she spoke up and provided feedback—all of which gave her hope that change is possible. Toward the end of her story, she said, “The healing we can do as human beings with each other is greater than just the healing of stitching up the body.” Her story beautifully suggests the importance of helping people heal—not just physiologically, but also psychosocially. Healing is not an event but a journey, and it rarely returns us to the place or person we were before. The project demonstrates how stories can help socialize behavioral standards and competencies across Penn Medicine, and how they are an effective vehicle for building social cohesion among patients, caregivers, staff, and providers. The project features a diversity of voices, including physicians, nurses, caregivers, and patients, as well as chaplains, call center staff, social workers, home care workers, and ambassadors (also known as “Red Coats”). All story types are welcome, and the archive features first-person narratives of lived experiences and issues, including disability, sexual assault, COVID-19, trauma, and racism. The Listening Lab has had ripple effects across the health system, capturing the attention of staff from senior leadership to front desk Red Coat staff. The archived stories have now been listened to more than 25 000 times and are incorporated into new employee orientation and recruitment efforts, medical school and nursing education/training, staff training, leadership events, meetings, and staff huddles. The stories are also included in COVID-19 CEO updates, and in the recently launched PennMedicineTogether platform of self-care resources. The Listening Lab will also be featured on iPads in patient rooms in the recently opened Penn Medicine Pavilion at the Hospital at the University of Pennsylvania, located in University City. The pavilion has state-of-the-art single-patient rooms for those with the most serious conditions. The pavilion is the newest addition to Penn Medicine, completed and opened during the pandemic.

1. Ospina NS, Phillips KA, Rodriguez-Gutierrez R, et al. Eliciting the patient’s agenda-secondary analysis of recorded clinical encounters. J Gen Intern Med. 2019;34:36–40.

2. Lown BA, Rosen J, Marttila J. An agenda for improving compassionate care: a survey shows about half of patients say such care is missing. Health Aff (Millwood). 2011;30:1772–1778.

3. Holmes SM, Ponte M. En-case-ing the patient: disciplining uncertainty in medical student patient presentations. Cult Med Psychiatry. 2011;35:163–182.

Designing an Innovative Program for the Novice Nurse Workforce

Larecia M. Gill, PhD, MSN/Ed, RN; Tracy W. Suber, EdD, MSN/Ed, RN

Phoebe Putney Memorial Hospital

Background. New nurse graduates (NNGs) are faced with many unique challenges when transitioning from nursing student to registered nurse, including inexperience in clinical decision-making and lack of clinical skills—in addition to learning the nuances of the work environment (eg, documenting using an electronic medical record, unit/organization policies and protocols). With the introduction of computerized testing for the National Council Licensure Examination, NNGs began entering the workforce as registered nurses much quicker than before. This earlier entry into the nursing workforce requires NNGs to assume duties for which they are often unprepared. During spring semester 2020, COVID-19 erupted into a global pandemic and disrupted nursing programs’ classroom and clinical experiences. As a result, senior nursing students were unable to complete hands-on clinical training during their last semester of nursing school. As the pandemic persisted, virtual learning affected the academic experiences of nursing students. Clinical skills were learned using video recordings and hands-on practice was nonexistent—further perpetuating the chasm between classroom and clinical practice. As a result, NNGs are unprepared to enter into the workforce.

Intervention Detail. To address the clinical deficiencies that resulted from lack of clinical experiences in our nursing program, the Phoebe Simulation and Innovation Center at Phoebe Putney Memorial Hospital was designated as a dedicated education unit. In addition, dedicated education unit staff developed the Nursing Simulation and Training Education Program (NSTEP), an 8-week intensive program facilitated by experienced nurse educators. NSTEP includes hands-on simulation, didactic training, and skills assessment using high-fidelity mannequins. During the last 4 weeks of the program, NNGs also engage in observation and progressive floor/unit assignments through preceptorship training. All nurses who graduated in spring, summer, and fall 2020 were required to participate in NSTEP, in addition to completing hospitalwide orientation. The first cohort of NSTEP included 35 NNGs; cohort 2 began in July 2020 and included 5 NNGs. The third cohort included 19 NNGs and began in September, while cohort 4 began in February 2021 and included 16 NNGs. Additional cohorts will be offered for future NNGs whose clinical experiences are impacted by COVID-19.

Outcomes and Impact. Participants of each NSTEP cohort (through February 2021) were asked to complete pre- and post-test evaluations to determine the impact of the program. Eighty percent of respondents (N = 39) were female with a mean age of 29 years. Using a Likert scale from 1 (strongly disagree) to 5 (strongly agree), respondents reported feeling: (1) better prepared to respond to changes in their patient’s condition (87%; M = 4.23); (2) more confident in their nursing assessment skills (95%; M = 4.31); (3) empowered to make clinical decisions (90%; M = 4.29); (4) confident in their clinical decision-making skills (87%; M = 4.20); (5) confident in their ability to prioritize care and interventions (92%; M = 4.35); (6) confident in communicating with their patients (92%; M = 4.43); (7) confident in their ability to report information to the health care team (95%; M = 4.43); (8) confident in providing interventions that foster patient safety (95%; M = 4.35); (9) confident in using evidence-based practice to provide nursing care (87%; M = 4.27); and (10) confident in their ability to teach patients about their illness and interventions (87%; M = 4.42). In addition, findings from repeated measures using analysis of variance indicate that participants’ confidence in each of these elements increased significantly from week 1 to week 8 of NSTEP (P < 0.05).

Ambulatory Quality in the Time of COVID-19

Catherine Chang, MD, MMM; Katie Lawrence, MHA, CMPE

Prisma Health

Background. As value-based ambulatory care contracts have matured over the last 5 years, high performance in quality of care has transitioned from the main key that unlocks success to one among many challenging and complicating factors that must be addressed. The cost of care conundrum—often reflected in terms of overall payor spend per patient and impacted by ambulatory admissions/readmissions, emergency department utilization, imaging, specialist procedures, and medication—is now critical to success in Medicare, Medicaid, and commercial, value-based arrangements. Prisma Health’s Ambulatory Team developed a multilevel and multifaceted approach using physician practice accountability, transparency in performance assessment, and pathways to success with in-practice support. The goal was to implement a digestible plan of action that would allow physician practices in the ambulatory setting to lower costs of care and manage a larger patient population by opening additional access portals—without sacrificing a continued commitment to the highest level of patient care quality. With COVID-19 presenting additional challenges, the Ambulatory Team pivoted its approach, incorporating digital health options and challenging physicians at the point of care to follow evidence-based guidelines even more closely when patients present infrequently. As our region progresses rapidly toward full capitation, the future of health care in our area will only be as secure as our commitment to understanding how best to help our physician practices and our patients navigate the challenges of cost, quality, and access. This presentation is included in the 2021 Vizient Connections Summit because most of the country is facing similar ambulatory practice challenges while simultaneously under the same time crunch; we have to learn how to do this together quickly if we are to maintain our market share in the midst of significant outside competition. And in light of the challenges COVID-19 presents to ambulatory care, it is critical to think deeply about how best to connect with patients who may not want to present in person.

Intervention Detail. The approach we used included performance improvement and operational processes cascaded via a curriculum calendar using leadership structures created for scale and physician contractual alignment, with transparent data reporting on progress. This multifaceted approach included: (1) follow-up from previous quality improvement work, with a more intensive focus on the Plan-Do-Study-Act (PDSA) cycle to improve patient care on quality measures that are even more challenging during COVID-19; (2) partnership with tech companies to enhance our ability to connect with patients who need services related to their age/condition/gender; and (3) increased efforts to engage practices and physicians in the culture of commitment to quality. To share progress and motivate improvement, reporting included: (1) a proprietary, web-based scorecard using electronic medical record data for clinical quality performance of individual physicians, with transparent comparisons to their practice colleagues, specialty peers, and the entire clinician network; (2) weekly reviews related to PDSA work via chart audits; and (3) consistent gathering of data from third-party partners related to our technology work.

Outcomes and Impact. PDSAs and interventions at the point of care for our primary care and specialist targets resulted in measured improvement in following evidence-based guidelines for the care of hypertensive and diabetic populations. The ambulatory quality performance of primary care physicians and specialists remained stable during COVID-19, staying in our near top decile or top quartile performance in clinical quality measures tracked by our organization. Our work in connecting patients with needed screenings resulted in increased breast and colorectal cancer screenings during COVID-19, despite the challenges posed by having to temporarily close screening locations. Our focused work on flu resulted in almost 90% of patients being vaccinated/or offered, resulting in close to zero flu cases across our system.

Innovative Pharmacy Strategies to Increase Financial Sustainability Amidst the COVID-19 Pandemic

Sophia Humphreys, PharmD, MHA; Elie Bahou, PharmD, MBA

Providence

Background. Prior to the COVID-19 epidemic, American not-for-profit health care organizations had an average operating margin of only 2.1%.¹ Unfavorable factors, including the ever-rising cost of medications, threatened sustainability. The onset of COVID-19 further threatened financial viability. It is estimated that the pandemic caused an economic loss of more than $90 trillion internationally.² In the United States, 46 million individuals have filed for unemployment benefits since mid-March 2020.³ The American Hospital Association estimated a financial loss of $323 billion for health systems in 2020.⁴ These and other COVID-19-related factors may further reduce reimbursement to health care entities, diminish patient access to care, and negatively impact the financial sustainability of the health care industry. The cost of medications accounts for 17% of the expenses incurred by the American health care industry.⁵ Thus, any reduction in medication expenditure will increase health system sustainability. Providence is the third-largest nonprofit health care system in the United States, with an annual medication spend in the billions. Providence’s pharmacy division implemented an innovative and comprehensive medication utilization program that simplified the formulary, reduced overall drug cost, improved collaboration with contracting and the Vizient network, increased contract compliance, optimized 340B program benefits, and improved medication shortage management. This program took a multidisciplinary approach that was systematic and integrated. Enhanced collaborations took place among pharmacists, physicians, finance experts, 340B program experts, pharmacy buyers, Vizient network colleagues, and health care informatics professionals. Such innovative medication utilization achieved more than $40 million in drug purchase savings in 2020 and improved overall organization financial sustainability

Intervention Detail. (1) Innovative medication utilization management (UM): Pharmacoeconomic analysis revealed that the top 20 medications by spend at Providence were all biologic medications that constituted 40% of the entire drug budget. Thus, efforts were focused on finding biosimilars that are highly similar in efficacy, safety, and immunogenicity to these biologics, but cost 40% to 50% less. Seven UM initiatives were implemented to promote biosimilar adoption throughout the health system. These initiatives saved approximately $16 million in 2020. (2) Streamline COVID-19-related protocol approval process: A COVID-19 pharmacy work group (work group) was formed that included several mature multidisciplinary clinical practice committees. The work group evaluated emerging clinical evidence and drafted multiple COVID-19 medication utilization protocols to guide frontline providers. The work group aligned its efforts with the UM program to standardize practice and reduce cost. A special focus was placed on COVID-19 essential medication therapeutic subclasses. As a result, a COVID-19 treatment essential medication list was developed. A joint clinical council was formed to quickly and thoroughly review the emergency use authorization for COVID-19 vaccines. All vaccines were approved the day after the Centers for Disease Control and Prevention’s Advisory Committee on Immunization Practices recommendations were published. (3) Leverage electronic health record (EHR) tools: All UM initiatives, as well as COVID-19-relevant medication protocols, were built into the EHR system to guide physician prescribing in order to optimize the benefit of Providence-preferred products, such as biosimilars. (4) Maximize procurement power: A pharmacy network was established by partnering with Vizient, Providence’s supply chain partner. The network streamlined contracting efforts and enhanced biosimilar program savings. Providence also collaborated with pharmacy distributors and buyers to maximize rebate opportunities, bulk buy possibilities, and contract compliance. (5) Minimize drug shortages: A multidisciplinary team established the COVID-19 essential medications list; identified possible shortages; and collaborated with regular distributors, individual manufacturers, and Civica Rx (a nonprofit generic drug company) to ensure essential medication supply.

Outcomes and Impact:

• A. Utilization program savings:

• All 7 biologic medications with biosimilar products were identified and their respective biosimilars were added to the system formulary.
• One preferred biosimilar was selected for each high-cost biologic, and the preferred biosimilars for all 7 biologics were built into our EHR as the default products.
• Biosimilar-specific UM initiatives were implemented for all 7 preferred biosimilars.
• Total annual drug purchase savings for these biosimilar initiatives exceeded $16 million in 2020.

• B. COVID-19-relevant protocols and approvals:

• “Considerations for COVID-19 Patients’ Ventilation-Related Medications,” which consisted of protocols for neuromuscular blockers, sedation agents, and analgesic medications, was approved and implemented within Providence in April 2020.
• “Providence St. Joseph Health Guidance for SARS-CoV-2 (COVID-19) positive in Non-pregnant, Hospitalized Adult Patients” was approved by the system Pharmacy and Therapeutics Committee and other experts. This guideline outlined the utilization of remdesivir, dexamethasone, convalescent plasma, and many other COVID-19 treatment options. This guided Providence frontline providers in their efforts to treat and manage COVID-19 patients. It is updated regularly with new evidence as it becomes available.
• All 3 COVID-19 vaccines were approved via expedited process by the Joint Clinical Council and made available for administration as soon as Advisory Committee on Immunization Practices recommendations were finalized.

• C. Purchase optimization:

• Total Providence pharmacy purchase optimization program savings exceeded $11 million in 2020.

• D. Vizient pharmacy network:

• The network implemented both contract and UM initiatives, yielding accumulated savings exceeding $7 million in 2020. This comprehensive pharmacy program reduced Providence’s overall drug spend more than $40 million in 2020, while facing unprecedented financial challenges due to COVID-19.
• The program also enabled us to expedite the execution of high-impact initiatives.

1. Paavola A. Fitch: Nonprofit Hospital Margins Improve for First Time Since 2016. Becker’s Healthcare: Hospital CFO Report. 2019. Accessed February 6, 2021.

https://www.beckershospitalreview.com/finance/fitch-nonprofit-hospital-margins-improve-for-first-time-since-2016.html.

2. Jackson JK, Nelson RM, Weiss MA, et al. Global Economic Effects of COVID-19. Congressional Research Service (CRS). 2021. Accessed January 30, 2021. https://sgp.fas.org/crs/row/R46270.pdf.

3. U.S. Department of Labor. Unemployment Insurance Weekly Claims [press release]. 2020. Accessed February 5. 2021. https://oui.doleta.gov/press/2020/061820.pdf.

4. American Hospital Association. New AHA Report: Losses Deepen for Hospitals & Health Systems — Catastrophic Financial Impact of COVID-19 Expected to top $323 Billion in 2020 [press release]. 2020. Accessed March 5, 2021. https://www.aha.org/press-releases/2020-06-30-new-aha-report-losses-deepen-hospitals-health-systems.

5. Hartman M, Martin AB, Espinosa N, et al. The National Health Expenditure Accounts Team. National health care spending in 2016: spending and enrollment growth slow after initial coverage expansions. Health Affairs (Millwood). 2018;37:150–160.

Rush’s Anchor Mission: Addressing Zip Code Impact on Life Expectancy

Shweta Ubhayakar, MBBS, MS-HSM; John Andrews, MS-HSM

Rush University Medical Center

Background. Chicagoans who live a few blocks from each another have up to a 16-year difference in life expectancy on average.¹ For nearly 150 years, Rush University Medical Center (RUMC) has helped make people healthier by providing excellent health care. However, research increasingly suggests that factors outside of health care have a much larger impact on overall health. Pervasive social issues such as lack of access to quality education, healthy food, affordable housing, systemic racism, and unemployment contribute to many illnesses that shorten people’s lives. Historically, there have not been significant investments in these neighborhoods to improve health and economic vitality. As an academic medical center with deep roots in the community, RUMC leadership recognized that the organization was well positioned to address these social and structural determinants of health. In 2017, RUMC launched an anchor mission (AM) strategy to align resources across the health care delivery system, university, and business units to improve the health and economic vitality of 10 West Side neighborhoods. The evidence of RUMC’s strong commitment, including that of senior leadership and the RUMC Board of Trustees, is demonstrated by the community strategy being publicly identified as 1 of the organization’s 4 primary strategic priorities. The overall goal is to reduce the gap in life expectancy between Chicago’s Loop and West Side communities by 50% by 2030. RUMC leadership recognizes that this audacious goal cannot be achieved alone and requires strategic partnerships across organizational partners and community residents. This mission has led to the formation of the West Side United collaborative. Sharing RUMC’s AM efforts helps reinforce the idea that hospitals and health institutions have a critical role to play in addressing health disparities. It will allow other institutions to learn and implement from our experience while also enhancing our efforts to identify better ways to achieving health equity.

Intervention Detail. RUMC decided to take a holistic approach and developed an all-in AM strategy to address the gap in life expectancy. The AM strategy includes 5 key initiatives that align RUMC’s business units as an economic engine to improve the vitality of West Side communities. The 5 initiatives include: (1) hire locally and develop talent; (2) buy and source locally; (3) hire local labor; (4) invest locally; and (5) volunteer locally. Target and stretch goals are set annually for each initiative and reported to senior leadership and the board of trustees. RUMC partnered with local, community-based organizations to ensure that the community voice and recommendations are incorporated within initiatives and to help build community trust. RUMC works with community sourcing and training organizations to advance its local hiring initiative. A community-based organization, tiered-approach model was developed to refer qualified candidates for entry-level positions and provide wraparound support services. RUMC develops career pathway programs for employees to advance their careers. Two incumbent worker pathway programs were launched, including the Medical Assistant Pathway Program and Patient Care Technician. The Rush Capital Projects Team includes goals for AM local hiring and spend in contract language. We have committed to investing in social impact projects that address community needs, such as affordable housing, workforce development, and food access, through our impact investment initiative. RUMC is also committed to the health and vitality of its own employees, particularly low-wage employees, and to helping them create wealth and financial stability. We launched several pilot programs on financial education and offered student loan opportunities. We collaborate with several West Side organizations to structure volunteer opportunities for our employees. Lastly, our Anchor Procurement Team identifies procurement needs and develops a detailed work plan to track spend with local vendors. We have also developed partnerships to move tier 2 spend to West Side vendors.

Outcomes and Impact. Since 2017, RUMC has increased local hiring by 7% compared to baseline and opened 16 application hubs in the AM communities—specifically to enhance connectivity that can provide additional resources to candidates during their job search and after job attainment. The Medical Assistant Pathway Program is a 2-year program for Rush employees that fills high-demand medical assistant jobs. Through June 2020, we have launched 3 cohorts with 28 participants. Patient Care Technician pathway is a 9-month program that seeks to develop local residents with little to no experience into future health care professionals. We have launched 4 cohorts with 37 participants, with a 65% program completion rate and a 95% employment rate. Local procurement initiative outcomes include an increased 7% shift in spend to West Side vendors over the past 2 years. Through our strategic community partnerships with Concordance Healthcare Solutions and Foods, we have successfully shifted more than $2 million in spend to West Side communities. Our partnership with Concordance Healthcare Solutions includes a contractual commitment to hire from AM communities. Since the start of this partnership, approximately 40% of their warehouse staff has come from AM communities. RUMC invested $6 million over 3 years in West Side social impact projects for affordable housing and workforce development. The number of employees who are saving 6% or more of their earnings has increased by 8% in the last 2 years, and we provided employee financial counseling through Working Credit, a not-for-profit organization that can help you build a strong credit score. We have pooled investments as part of West Side United collaborative and awarded approximately $1 million in grants to small businesses. As part of our employee volunteer program, we have provided food to more than 8000 employees and residents on the West Side and raised $80 000 in donations for food distribution. RUMC published an AM playbook to summarize our journey in this important work.

1. Rush University Medical Center. Community Health Needs Assessment Report. 2016. Accessed March 18, 2021. https://www.rush.edu/sites/default/files/2020-09/rush-chna-august-2016%284%29.PDF.

Purchased Services Be Nimble, Be Quick: From Zero to $6.6 Million in Savings

John Andrews, BA, MS; Hawanya Blakely, BS, MBA

Rush University Medical Center

Background. The COVID-19 pandemic has affected many industries nationwide, particularly health care. Cost increases for personal protective equipment and additional support services, combined with lost revenue from the dramatic drop in surgeries and patient census, has significantly impacted the bottom line at Rush University Medical Center (RUMC). The pandemic is not yet behind us, and so the question remains, how do we stop the bleeding? Like many health care organizations, RUMC took critical and immediate measures to ensure its financial and operational position. One of the biggest challenges was devising a process for reviewing overall costs in a quick and comprehensive manner. The ability to be nimble and quick was paramount, as we had to shift cost savings priorities and focus on speed-to-value savings opportunities. In addition, we needed to devise a way to be collaborative with departments that had traditionally been able to negotiate outside the organizational strategic sourcing plan. Purchased services is 1 of many areas across RUMC in which Strategic Sourcing was tasked with identifying savings opportunities. Purchased services encompasses any purchased, contracted, or outsourced third-party services. This was previously an area of limited involvement from the Strategic Sourcing team, which typically focused on traditional purchased services categories (eg, copier repair, shredding services, record storage). The financial impact of the pandemic forced us to review this category more broadly. After receiving senior leadership buy-in, which was imperative, the Purchased Services Initiative (PSI) project was launched. Project goals were to: (1) identify potential savings opportunities; (2) identify contracts currently negotiated outside of Strategic Sourcing; (3) eliminate unnecessary or underutilized services; (4) standardize services; and (5) renegotiate contracts. This abstract details how we addressed the purchased services category and why Strategic Sourcing should always have a seat at the table.

Intervention Detail. First, the Strategic Sourcing team reviewed fiscal year 2020 general ledger spend, which took 2 weeks to complete. Upon initial review, it was clear that the data needed to be scrubbed, mainly due to miscoded general ledger accounts. Another contributing factor of inaccurate data coding was the Strategic Sourcing team’s lack of involvement in all purchased services contract engagements—making it difficult to delineate between purchased services and supply spend. Once the data scrub was complete and the purchased services spend identified, individual templates were developed for each department—including their fiscal year 2020 spend by vendor. The template also included a section for the departments to complete. Each department reviewed their spend and populated the template with the requested contract information, that is, service description, effective dates, purchase order number, if applicable. Departments were given a specific due date to complete the templates and return them back to Strategic Sourcing. Stakeholder meetings were scheduled, during which the completed templates were presented to department leads. The meetings were intended to review and gain insight into each department’s purchased services spend, challenge them to identify cost savings opportunities, and eliminate unnecessary or underutilized services. In addition, they were able to identify contracts that were targets for renegotiation. In total, 35 departments participated in the spend review process. In the past, outside consultants were hired to lead cost reduction initiatives. Insourcing the work allowed RUMC to expedite the process and forgo any additional consulting costs.

Outcomes and Impact. One of the chief outcomes was the realization of $6.6 million in savings in fiscal year 2021. This was achieved through a combination of renegotiating pricing, eliminating nonrepeat expenditures, and eliminating underutilized services—a direct result of our departmental meetings. To ensure that savings targets were achieved we developed a purchased services dashboard for each department, which will be used to monitor spend and hold departments accountable to their savings targets. Overall, departments across the organization were willing to work in a collaborative way with Strategic Sourcing to make the PSI successful. Despite having senior leadership’s approval, we initially encountered pushback from some stakeholders. Many were accustomed to managing their sourcing activities without including Strategic Sourcing in the process. The PSI project provided Strategic Sourcing an opportunity to lend its expertise to the purchasing process and demonstrate why we should be involved in the early stages of any sourcing initiative. Furthermore, this culture shift will free up department leads to focus on their core work. The PSI is an ongoing project. Leading this project enabled Strategic Sourcing to identify other areas of potential improvement, including how future budgets are established, how departments manage their spend, and how to hold departments accountable to savings targets. Other areas of focus as we continue our PSI journey to help bring value to the organization through continued process improvement include: (1) eliminating or minimizing misclassification of expenses to general ledger accounts; (2) smoothing out the impact on the general ledger through coordinated accruals (reduce variance explanations); (3) create a robust contract management plan that Strategic Sourcing can use to support departments moving forward; and (4) reduce the amount of spend without appropriate purchase orders being created.

Documentation and Billing Accuracy for Implanted Devices: A Hyper-Rapid Performance Improvement Model

Thomas A. Deutsch, MD, Rush University System for Health; Anisa J. Jivani, MHSA, Rush University System for Health; Paul Casey, MD, MBA, Rush University Medical Center

Rush University Medical Center

Background. The Decision Support team at Rush University Medical Center determined that 22% of the charges and 6% of the actual number of implants used in the operating room in fiscal year 2020 fell into open or incomplete, service-specific implant billing codes. These codes are typically used by nurses to manually document the implant/supply log when implants are not already in the institution’s item master (also referred to as “on-the-fly” implants). Open codes must then be reviewed by the billing team on the back end to ensure the implant is under contract and that the appropriate charge and charge code are assigned to the item before posting the surgical log for billing. Thus, open codes lead to manual interventions by both the nursing and billing teams in order to ensure accurate billing. On the back end, decision support has no insight into what these items actually are, which can create an inaccurate representation of what items are being charged and the reason for the charge.

Intervention Detail. A multidisciplinary team was assembled and participated in 2 rapid improvement events aimed at finding and implementing solutions with immediate results. The first event focused on identifying the drivers of on-the-fly implants and developing interventions to reduce their occurrence. The second event examined nursing documentation workflow and how we could leverage our inventory management software and the scanning documentation function to search for implants before deciding to manually document.

Outcomes and Impact. After implementing the identified solutions, we immediately reduced the weekly number of on-the-fly implants by 44%—which led to a decrease in monthly open code spend/count by 50% across 4 months. By doing so, we were able to reduce both the nursing team’s documentation burden and the time required to post a surgical log, while also improving our data accuracy for decision-making purposes. A secondary and successful goal was to demonstrate that the hyper-rapid improvement approach could assess a problematic area and quickly test and implement solutions that have a significant (positive) effect on administrative processes in the perioperative environment.

Expanding Primary Care Access to Behavioral Health: A Redesigned Psychiatric Chronic Care Management Model

Neha V. Gupta, MD; Michael Hanak, MD, FAAFP; Jonathan Kaplan, MD; Erika D. Owen, MS, LCSW; Aatif Hashmi, MS

Rush University Medical Center

Background. Depression is a leading reason for primary care visits and often the initial opportunity to address this potentially disabling health condition. Primary care providers (PCPs) face a high burden in arranging specialty care for behavioral health conditions, as demand is outpaced by limited resources and historically poor access. This is especially true in underserved communities. The consequences of this misalignment are untreated mental health conditions and higher downstream health care costs, resulting in a growing economic toll in the United States. Universal depression screening and intervention aim to solve this gap while addressing disparities around access to behavioral health services. Historically, equal access to high-quality mental health treatment has been limited by socioeconomic factors, including insurance coverage, provider scarcity, lack of awareness about mental illness in many communities, and cultural views and stigma surrounding mental illness. Universal depression screening and intervention aims to solve this gap while addressing disparities around access to behavioral health services. Despite implementing a systemwide screening and intervention program over the course of 4 years, we identified significant barriers in addressing the disparities within our community, including the need to simplify access into our program by PCPs. In addition, we were not following an evidence-based model of care proven to improve depression outcomes, nor were we utilizing our highly qualified clinical team to the fullest potential. Further, to support program stewardship and allow for concise reporting of program utilization, we needed to organize program throughput. Lastly, in the age of health care reform, we were missing key revenue opportunities for time spent on chronic care management. Because this work can be successfully scaled and creates quality improvement opportunities alongside new revenue streams, this work is included as part of the 2021 Vizient Connections Summit to help peer organizations achieve similar goals.

Intervention Detail. The Collaborative Care Program (CCP) consists of 3 licensed clinical social workers known as behavioral health care managers (BHCMs) a psychiatric consultant, and a patient registry that BHCMs utilize to identify new patients, track patient participation, and monitor clinical progress. Within 7 days of a PCP referral, the patient is added to the registry and receives an outreach telephone call from the BHCM. During this call, the patient is either connected to mental health resources (which closes the intervention), or the patient proceeds to a comprehensive evaluation to enter the CCP. For patients who enter the CCP, the BHCM then liaisons with the psychiatric consultant through weekly caseload reviews, and care plan feedback is subsequently presented back to the PCP to support their management of the patient. The BHCM continues to follow the patient on a monthly basis, tracking progress until goals are met. To address existing barriers, we made several changes to the CCP program to more closely follow the Collaborative Care Model, a clinically proven type of integrated care developed by the Advancing Integrated Mental Health Solutions Center at the University of Washington. First, from a clinical workflow standpoint, it was important to emphasize the BHCM’s role in providing evidence-based psychotherapeutic intervention rather than solely navigating external referral resources. This required BHCM training, documentation optimizations, and organizational alignment to support dual BHCM reporting to psychiatry and care management leadership. Second, we created several electronic health record (EHR) and process optimizations aimed at improving outreach efficiency. To expand our scope in addressing a higher complexity of mental health conditions, a direct referral order (without a formal depression screening requirement) was added as an option for primary providers. Furthermore, by defining inclusion and exclusion criteria for the program, we were able to provide targeted care and reduce duplicative effort. Finally, our EHR-based registry was reorganized to provide a variety of daily caseload tracking tools and abilities, including monitoring screening tests, alerting care managers when an intervention was due, and identifying patients who were not responding to treatment. The most impactful change was the implementation of incident-to-billing practices for services. This provided additional revenue to support the program, while encouraging a patient-centered, collaborative approach to addressing mental health in the primary care setting.

Outcomes and Impact. Quantitative: The program launched in November 2016 at a single pilot clinic. At that time, depression screening across primary sites averaged 17% of visits, with less than 40 program referrals in the first months of the launch. During the first phase of program expansion from 2017 to 2018 (which included the delivery of individualized clinical trainings to each added practice site by the psychiatric consultant, the BHCMs, and the EHR team), screening rates increased to 74%, peaking at 87% in 2019. Screening continued to exceed expectations at 84% in 2021, with adjusted goals during the COVID-19 pandemic. As a result of increased screening efforts and valuable clinical services being offered, program referrals grew substantially, now exceeding 1000 referrals/year across a significantly expanded clinical footprint of 69 primary care and specialty clinical practices. The redesign of this program in October 2020 included the launch of billing for services (which required more intensive documentation and follow-up). This implementation resulted in charges for 481 procedure codes from January 1, 2021 to December 31, 2021, an average of 40 charges per month. These charges reflect services rendered to a smaller portion of patients with eligible insurance products who had not already been referred outside of the program. Payer reimbursement specific to these behavioral health integration codes averaged 30% of charges, although payer-sponsored quality incentives for depression screening functionally subsidized program costs. Continued program development and assessment of downstream utilization lends support to payer contract renegotiation. The CCP is thus an efficient model for delivering high-quality behavioral health services in settings customarily constrained by access. Qualitative: Through this program we have seen a dramatic shift in PCP comfort level with screening and treating behavioral health conditions. Patients who formerly waited weeks or months for a psychiatry consultation are now evaluated within 7 days. Treatment plans are managed by PCPs with close support from behavioral health specialists throughout the process. Further efficiencies have been achieved by training BHCMs as psychotherapists and psychiatric liaisons, reducing the need for escalation to specialty care.

Collaboration: Together We Can

Heidi Allbee, RN, BSN; Wanda Lanz, RN, BSN

Sanford Health

Background. The Vizient Quality and Accountability (Q&A) Study is our benchmark for performance, while also helping us determine our greatest opportunities for improvement. We started our Vizient journey in 2017 with subpar performance. The following year showed little improvement and demonstrated that a change was needed to improve our quality of care. Vizient identifies 5 key attributes of high-performing organizations, including: (1) a shared sense of purpose; (2) leadership style; (3) an accountability system; (4) a focus on results; and (5) a culture of collaboration. Our focus on a culture of collaboration has brought about a sense of teamwork and a shared sense of purpose. We would like to reveal how our mantra of “together we can” has improved our rank in the Vizient Q&A Study. Sanford has a strategic goal of reaching the study’s top 10—a goal that is talked about regularly among leadership and frontline staff and has created a shared vision with clear objectives.

Intervention Detail. The 2019 Vizient Q&A Study identified Sanford Bismarck’s largest areas of opportunity as mortality, effectiveness, and safety. While each of these domains are unique, lack of collaboration was a common denominator.

Mortality

• Partnered with clinical documentation/coding to ensure accurate documentation and understanding of coding rules.
• Identified a physician champion.
• Ensured appropriate use of hospice/palliative care.
• Established hospital informatics team for appropriate admit source/type.
• Results: Our highest mortality index was 1.77 in January 2017, while our lowest was 0.42 in September 2020.

Effectiveness

Readmissions

• Established a multidisciplinary work group, including providers, frontline nurses, navigators, pharmacy, and ambulatory staff.
• Reviewed all readmissions for learning opportunities, including home health and nursing home, follow-up appointments, heart failure clinic referral, follow-up phone calls, and use of community paramedics on high-risk patients.
• Worked with clinical documentation improvement, quality, coding, and providers for documentation opportunities.
• Acute myocardial infarction readmissions have dropped significantly, from a high of 13.33% in January 2019 to 4.76% in November 2020.
• Results: Heart failure readmissions dropped from a high of 28% in February 2018 to 9.7% in September 2020.

Safety

Catheter-associated urinary tract infections (CAUTIs)

• Defect tool review completed on all CAUTIs.
• Daily Foley catheter necessity form completed by nursing units and sent to infection prevention (IP) department for review.
• Prevention review by IP staff to remove Foleys and/or replace with external urinary catheter.
• Enterprise-based Foley removal/intermittent catheter protocol for urine retention implemented.
• Two-person Foley insertion recommendation for emergency room and nursing units.
• Work with anesthesia to remove Foley order from epidural orders as appropriate.
• Results: CAUTI standard infection ratio (SIR) rate decreased from 2.20 in the 3rd quarter of 2018 to below 1.00 in 5 of the last 6 quarters.

Central line-associated bloodstream infections (CLABSIs)

• Curos caps launched for all central line patients.
• Midline trial completed and implemented.
• Value analysis team and IP collaborated on central line indications and care needs.
• Defect tool review completed on all CLABSIs.
• Prevention review by IP staff for line necessity or maintenance needs.
• Results: CLABSI SIR rate was 2.32 in the 4th quarter of 2017, while 11 out of the last 12 quarters had a SIR rate below 1.00.

Outcomes and Impact. Our overall Vizient Q&A Study rank has improved greatly over the past several years. From 2018 to 2021, period 1:

• Mortality ranking improved from the 95th percentile to the 30th percentile.
• Effectiveness ranking improved from the 43rd percentile to the 9th percentile.
• Safety ranking improved from the 81st percentile to the 27th percentile.
• Overall ranking improved from 77th percentile to the 12th percentile.

Our successes are due to our ability to collaborate with individuals who embrace the true meaning of teamwork: working for the good of a goal, having a common aim, and working together to achieve that aim. We included members of quality, nursing, clinical documentation improvement, physicians, health information management, executive directors, and our chief medical officer.

Electronic Provider-to-Provider Consults: Improving Access to Quality Care

Sheena Reddy, MPH, MSc, Stanford Health Care; Olivia Jee, MD, Stanford University School of Medicine; Lena Giang, MPH, Stanford Health Care

Stanford Health Care

Background. Stanford eConsults, asynchronous, provider-to-provider consults in Epic with a turnaround time of 3 business days, play an important role in the organization’s patient access strategy. The workflow consists of the ordering provider seeing the patient and sending a clinical question with the patient’s concern to the consulting specialty. After the consulting specialist (eConsultant) replies with their recommendation, the ordering provider implements the recommendation. If the patient’s concern is complex, the eConsultant may recommend that the patient instead have a face-to-face visit in their specialty. eConsults accrue many benefits to stakeholders, including patients, providers, and the health system, and facilitate the right care, right time, right place approach to patient care. Patients can access timely specialty care at a lower cost. Ordering providers are educated on how to manage low-acuity specialty concerns for their patients. eConsults formalize curbside consults and free up specialty appointment slots for higher-acuity patients. eConsults are an effective triage tool for the health system, leading to efficient use of resources. eConsults also have telehealth advantages, such as saving valuable hospital space. The Stanford eConsult program launched primary care-to-medical specialty eConsults in October 2019, with a goal of reducing specialty wait times, and expanded during COVID-19 to reduce patient and provider exposure. The program was a cross-departmental collaboration, managed by the Stanford digital health team under chief medical information officer oversight. We have since launched surgical, interspecialty, intraspecialty, and inpatient eConsults, with over 3000 eConsults ordered at Stanford for nearly 20 specialties. Given our robust eConsult experience, this presentation provides an eConsult introduction to summit participants who are unfamiliar with this digital health tool, as well as practical information for those whose organizations are interested in implementing eConsults. This discussion is also timely given the essential role of telehealth during the COVID-19 pandemic.

Intervention Detail. Our business team began by scoping the business and operational use case for eConsults, as well as conducting an analysis to determine which specialties were well suited to eConsults by using a spectrum of criteria: specialty wait time, referral volume, specialty complexity, financial impact, clinician support, and common eConsult specialties at other academic medical centers. We then identified key stakeholders and departments, including billing, legal, compliance, primary care and specialty leadership, reporting, and information technology. Once a list of specialties was identified, the digital health team met with specialty leadership to designate eConsult champions and eConsultants to participate. These eConsultants underwent Epic training and helped determine which diagnoses within their specialty were suitable for the eConsult order. Primary care providers (PCPs) select a diagnosis when placing an eConsult order, which populates a template with helpful guidance on that diagnosis, as well as recommended labs to include. Partnering with billing and compliance, we developed a reimbursement model for providers based on Centers for Medicare & Medicaid Services recommendations. These billing workflows were presented to providers during their eConsult training. Our information technology team completed the initial Epic build in about 2 months starting with 3 specialties, and thereafter implemented additional specialties on a rolling basis (~1 d per specialty). As new specialties are launched, we send an announcement to PCPs in our monthly newsletter. When the program initially rolled out, we also did a “roadshow” to different PCP clinics to introduce eConsults and provide an Epic demo. We continue to do outreach and have introduced incentives, such as a bonus for PCPs.

Outcomes and Impact. Our resolution rate averages 80% (the percentage of eConsults that are resolved without requiring a specialty visit), and ranges from 40% in orthopedics to 95% in psychiatry. Although eConsultants have 3 business days to reply to an eConsult, on average, PCPs receive a response within 1 business day. As patients typically wait for 30 or more days for a specialty appointment, this dramatically sped up access to specialty care. Patients appreciate the eConsult service; in a survey conducted after our initial eConsult pilot, 96% of patients had a favorable experience and 95% felt this service saved them time.¹ We have also seen decreased wait times for eConsult patients whom the eConsultant recommends for a face-to-face visit, because the eConsultant has already triaged these patients and they can be fast-tracked into the specialty if the concern is urgent. In dermatology, the average wait time for a new patient visit is 50 days, while the average wait time for a visit was 20 days for patients who had their dermatology eConsults converted to a face-to-face visit. This population also demonstrated lower no-show and canceled rates, possibly due to the decreased wait times. In dermatology, the average no-show and canceled rates for all face-to-face visits were higher, at 5.6% and 27%, respectively, compared to the subset of eConsults that was recommended for a face-to-face visit, which were, respectively, 2% and 19%. Our eConsult volume per 10 000 primary care patients at the end of program year 1 was more than double the volume of other academic medical centers that also launched eConsult programs within a similar timeline. We attribute the increase of eConsults to consistent provider outreach, rapid roll-out of specialties, and primary care incentive programs.

1. Kim GE, Afanasiev OK, O’Dell C, et al. Implementation and evaluation of Stanford Health Care store-and-forward teledermatology consultation workflow built within an existing electronic health record system. J Telemed Telecare. 2020;26:125–131.

The Evolution of Frontline Problem-Solving

Jake Mickelsen, MBA, MBB, Stanford Health Care; Amelia Sattler, MD, Stanford Family Medicine

Stanford Health Care

Background. Health care conferences throughout the country continue discussions on the importance of supporting frontline problem-solving. Improvement activities can help systems achieve the Quintuple Aim: better patient health, better patient experience, lower cost of care, improved work life for providers and staff, and health equity. Upside-down pyramids showcase the idea of improvement starting from the front line, with impactful projects being supported and spread by leadership. Although the principle makes sense conceptually, it is difficult to execute effectively, or even respectfully. Frontline team members are busy managing their day-to-day challenges and have limited time to focus on improvement. Teams are most accountable and responsible for how they manage patient care. There is far less emphasis and time dedicated to improving on existing processes and procedures. Finally, the knowledge and tools needed to improve a complex health care system are not easily accessible. Common approaches to engaging the front line include daily huddles, visibility boards, and idea cards—yet the struggle continues. What actually works in supporting the front line to do improvement work? Does the current paradigm need to evolve? In February 2018, a survey was performed across a sampling of clinics assessing improvement capability (n = 90) of physicians and staff. The results showed that 33% of employees did not feel supported to do improvement work or feel comfortable applying improvement methods. Looking specifically at physician engagement, only 5% of physicians were actively involved in improvement work each year.

Intervention Detail. Forty improvement professionals and physicians were asked to identify the main drivers for effectively supporting frontline improvement. Four key drivers were identified. First, use an appropriate accountability structure. Improvement programs must provide accountability in a way that is fair and consistent in order to achieve results. For example, for 13 primary care clinics, a 30-minute weekly virtual meeting was created where leaders engaged with frontline team members. Frontline project leaders were accountable for reporting on project progress in front of their peers. Second, problems must be triaged. Just like patients coming to the emergency department are triaged based on level of clinical acuity, frontline problems must be triaged based on level of complexity, feasibility, and relevance to institutional priorities. Some projects are appropriate for discussion in daily huddles, while others require a different context. A variety of project report-out opportunities must exist to support the varied complexity of frontline problems. Third, there must be practical tools and designated resources to support improvement work. For example, physicians and staff with experience in improvement methods were given protected time to coach frontline project leaders. Easy-to-use templates and helpful YouTube videos were developed and available for all participants to access. Fourth, ensure people are recognized for their work, no matter how small or large the impact of the results. At the conclusion of each project, deliberate efforts should be made to calculate the annualized impact, which can be related to financial, quality, patient experience, or engagement outcomes. The effectiveness of these 4 drivers was measured by number of participants and projects completed, engagement surveys, and benefit realized. Employee engagement surveys measured if the structure was helpful and if knowledge growth occurred. Benefit realization for each project included cost, patient experience, quality and safety, and employee/physician engagement.

Outcomes and Impact. One clinical department (approximately 300 providers and staff) implemented the frontline improvement structure, taking into consideration the 4 proposed drivers. The department completed 52 projects in 52 weeks, thus naming their structure “52in52.” Annualized outcomes included $10 million in revenue and the reduction of more than 50 near-miss safety events. The approach has continued to yield similar results for more than 3 years. Another department implemented a similar structure in a 13-clinic outpatient environment. The percentage of physicians involved in leading improvement projects increased from 5% to 33% (n = 100) annually. After the first year, over 8000 hours of wasteful labor tasks were removed from outpatient processes (eg, time spent entering orders, answering patient messages, managing medical records). Over 12 000 patients experienced improved care and access to services as a result of the project work (eg, appropriate tobacco screening and counseling, decreased wait times, improved blood pressure control). Some project results were so significant that they spread throughout the clinical care system and led to regional and national conference presentations. Participation in improvement projects is now part of the medical assistant promotion ladder at our institution. The improvement structure continues today as a key employee engagement mechanism and a vehicle for completing local projects supported by physicians, medical assistants, and nurses. After the program was implemented, staff and physicians saw significant increases (> 10%) in feeling supported in performing improvement work and feeling comfortable in applying improvement methods and tools independently. The entire medical institution has recognized that all units and departments should adopt and implement a structure for frontline improvement based on these 4 drivers to support their local teams in solving frontline problems in pursuit of the Quintuple Aim.

Transforming Health Equity Through Data Visualization and Quality Improvement

Amy C. Lu, MD, MPH, Stanford School of Medicine; Angela Graf, MS, Stanford Health Care

Stanford School of Medicine

Stanford Health Care

Background. Of the 6 aims of providing quality care espoused by the National Academy of Medicine (formerly the Institute of Medicine), progress in providing equitable care has significantly lagged behind the others. The COVID-19 pandemic and the Black Lives Matter movement created the burning platform to reexamine longstanding health disparities and structural inequities in patient care outcomes. Our health system rapidly responded by creating an inclusive, organizational health equity committee with representation from multidisciplinary, interprofessional community and patient members. Using an improvement framework that includes clinical and quality governance, we created strategic goals of building a culture of health equity, as well as improving community engagement and outreach, clinical and quality outcomes, and patient recruitment and access. One of our health system’s initial priorities was to create a meaningful health equity dashboard stratified by patient demographic characteristics, medical conditions, and social determinants of health in our patient population. The dashboard presented unique challenges in data abstraction and validation that have important implications for future efforts in research and the care delivery system. The dashboard was co-created with a human-centered approach through numerous iterations with multidisciplinary stakeholders, including patients, caregivers, and clinicians from traditionally underrepresented and vulnerable populations. Given the events of the pandemic, it is imperative that health systems address health disparities. We recommend that Vizient support critical clinical and quality initiatives by highlighting health equity efforts among its member institutions. Our proposal supports this recommendation and is a valuable part of the 2021 Vizient Connections Summit as an example of our continued innovations in national health care imperatives.

Intervention Detail. Methods and materials: (1) Data preparation: Prior to data visualization, Vizient and Epic data sets were analyzed and validated for various demographic attributes, including race, ethnicity, language, age, insurance, zip code, and sexual orientation. Where applicable, data for race and ethnicity were updated to match data collection standards. Opportunities for data collection improvement were documented and presented to the Stanford Medicine Health Equity Committee for formal project development. Sources for various benchmarking and geographic-specific attributes are obtained from the American Community Survey, data.census.gov, and in some cases, individual county data sets. A key component of analyzing health equity data was accomplished by leveraging the Vizient Quality and Accountability Study methodology and Vizient Clinical Data Base Resource Manager. The patient population methodologies for sepsis, non-ST segment elevation myocardial infarction, and heart failure were integrated into the analysis, as well as the process measures for those cohorts. Vizient-specific outcome measures were also analyzed, including risk-adjusted mortality, length of stay, and direct cost indices; however, these outcomes were assessed beyond the Vizient-defined equity cohorts. Finally, certain Vizient demographic taxonomy definitions from the equity domain methodology were applied, including socioeconomic status and race. (2) Data visualization: The team utilized Tableau as a platform for visualizing the Stanford Health Care Inpatient Health Equity Dashboard. This allows for dynamic selection of filters and views, as well as leveraging data from our enterprise data warehouse, which includes many data sources such as Epic, Vizient, and various subject area marts. Two dashboard views are available for interaction by the end user, including a high-level, demographic-only view that summarizes the inpatient population and provides an overview of the patient landscape and a detailed measure view that explores several different outcome and process measures through various demographic attributes. The latter view provides the opportunity to identify differences between patient populations or demographics and further explore if potential disparities may exist.

Outcomes and Impact. Using Vizient and Epic data, we analyzed medical conditions according to race, ethnicity, language, socioeconomic status, age, and sex. Initially, we measured high-impact outcome and process measures over several years. Some differences, even if not statistically significant, are continuously monitored because of clinical importance and patient volume for those conditions. Additionally, clinical judgment (rather than statistical significance) is also a factor in the prioritization of certain outcomes and demographic pairings. Initial analyses revealed several areas with statistically significant differences, including inpatients from a lower socioeconomic status (as determined by public insurance) who had a higher rate of readmission (14.8%) than those admitted with private insurance (8.6%; P < 0.001). Patients whose primary language is not English were also more likely to be readmitted (11.9%) compared to patients whose primary language is English (7.9%; P < 0.001). Similar to national COVID-19 trends, inpatients with COVID-19 were more likely to be from a lower socioeconomic status level than inpatients without COVID-19 (P < 0.001) and were more likely to be from Hispanic/Latino ethnicities than non-Hispanic/non-Latino ethnicities (P < 0.001). Other common conditions where differences were seen that were not statistically significant included patients admitted for heart failure, non-ST segment elevation myocardial infarction, and sepsis. We are continuing to analyze our data according to less commonly measured demographic characteristics, including preferred language, sexual orientation, gender identity, and social determinants of health (eg, food insecurity, housing instability). We intend to implement improvement interventions to address the health disparities using data from our health equity dashboard. This includes examining the highest impact social determinants of health for our vulnerable patient populations; improving our community partnerships and outreach efforts, as well as the accuracy of our demographic data collection; and designing staff, faculty, and trainee education on antiracism, implicit bias, and cultural humility.

Analyzing the Effects of Penicillin Allergies on Antibiotic Ordering Patterns

Kush Gupta, BS, Biomedical Engineering, Stanford University School of Medicine; Lisa Shieh, MD, PhD, Stanford Health Care; Ron C. Li, MD, Stanford Health Care

Stanford University School of Medicine

Stanford Health Care

Background. Penicillin-class antibiotics are the most frequently reported drug allergy in the electronic health record (EHR),^1,2 with approximately 10% of the United States population reporting allergies to penicillin.^2,3 However, literature has consistently found that over 95% of patients documented to have a penicillin allergy can tolerate penicillin-class antibiotics.^1,4 The reasons for this discrepancy are well studied and include improper allergy documentation, poor education regarding types of allergic hypersensitivity reactions, and a plethora of other factors. Nevertheless, EHR clinical decision support software produces prolific drug-allergy alerts based on documented penicillin allergies that ultimately affect physician prescribing behavior. It has previously been demonstrated that patients reporting penicillin allergy receive suboptimal therapy compared to those without penicillin allergy.⁵ However, there is a lack of discussion in literature regarding the specific and immediate deleterious effects of electronic drug allergy warning alerts and their effects on antibiotic prescription behavior. We describe an approach using EHR audit trail data to characterize how drug allergy warning alerts impact physician antibiotic orders for patients with documented penicillin allergies. We then describe the impacts of this analysis within the context of treatment for methicillin-sensitive Staphylococcus aureus (MSSA) infection.

Intervention Detail. We analyzed EHR data containing inpatient medication orders and medication administration data from Stanford Health Care between 2008 and 2018. Chi-square tests were used for significance. We first identified all orders for beta-lactam antibiotics (penicillin, cephalosporin, carbapenem, and monobactam-class medications) among patients with documented penicillin allergies. We next cross-referenced drug allergy alert data to discriminate between alerts that were ignored and alerts that resulted in cancelation of the order request. Medication administration records were used to confirm administration of ordered medications to the patient; these data further allowed identification of orders for which administration was stopped for reasons including allergic reaction. Finally, a retrospective cohort analysis of patients with culture-confirmed MSSA was performed. Bacterial antibiotic susceptibility laboratory data were analyzed and antibiotic orders placed immediately thereafter were collected. These data were then stratified by patient allergy phenotype and pooled for analysis.

Outcomes and Impact. Among the 133 795 patients with documented allergies to penicillins and/or cephalosporins, 149 631 drug allergy warning alerts fired between 2008 and 2018 in response to prescription orders for beta-lactams. Approximately one-third of these orders (49 997, or 33%) were canceled following the drug allergy alert. Seventy percent (34 759) of these cancelations occurred despite the patient having previously safely tolerated a beta-lactam. When comparing the antibiotics initially requested (responsible for triggering the drug allergy alert) to those ultimately ordered, in the majority of instances (20.8%), the medication was changed to vancomycin. This change usually reflected a broadening of antibiotic spectrum of activity. Finally, when comparing treatment for MSSA infection among patients with and without penicillin allergy, a statistically significant difference (P < 0.0001) was noted. Patients with penicillin allergy were more likely to receive vancomycin (24.3% versus 18.4%) and less likely to receive nafcillin (7.7% versus 10.5%) or cefazolin (17.6% versus 19.3%) than their nonallergic counterparts. We found that the vast majority of antibiotic cancelations and changes resulting from electronic drug allergy warning alerts are preventable; nearly 70% of these patients previously demonstrated tolerance to beta-lactams and could conceivably tolerate them again. In the majority of instances (18% of all antibiotic orders), these alerts result in a broadening of antibiotic selection to vancomycin. These differences in antibiotic selection abound even in scenarios such as treatment for MSSA infection, where antibiotic therapy is highly standardized and regimented; penicillin-allergic patients are significantly (P < 0.0001) more likely to receive vancomycin (24.3% versus 18.4%) and less likely to receive nafcillin (7.7% versus 10.5%) than their nonallergic counterparts. Our work has significant implications within the fields of quality improvement, antimicrobial stewardship, and hospital operations. We are soon implementing a clinical decision support solution that will display historical penicillin antibiotic administration data within the medication ordering workflow.

1. Blumenthal KG, Park MA, Macy EM. Redesigning the allergy module of the electronic health record. Ann Allergy Asthma Immunol. 2016;117:126–131.

2. Macy E, Ngor EW. Safely diagnosing clinically significant penicillin allergy using only penicilloyl-poly-lysine, penicillin, and oral amoxicillin. J Allergy Clin Immunol Pract. 2013;1:258–263.

3. Joint Task Force on Practice Parameters; American Academy of Allergy, Asthma and Immunology; American College of Allergy, Asthma and Immunology; Joint Council of Allergy, Asthma and Immunology. Drug allergy: an updated practice parameter. Ann Allergy Asthma Immunol. 2010;105:259–273.

4. Blumenthal KG, Shenoy ES, Huang M, et al. The impact of reporting a prior penicillin allergy on the treatment of methicillin-sensitive Staphylococcus aureus bacteremia. PLoS One. 2016;11:e0159406.

5. Shenoy ES, Macy E, Rowe T, et al. Evaluation and management of penicillin allergy: a review. JAMA. 2019;321:188–199.

Mobile Stroke Unit Program

David Fiorella, MD, (Neurosurgery); Eileen Conlon, RN

Stony Brook Medicine

Background. Time is brain. This expression is a reminder that when a patient has a stroke and the brain cannot get the blood and oxygen it needs, brain cells die. Stony Brook University Hospital identified the need to provide the most comprehensive and efficient care to our stroke patients before they even get to the hospital. It is estimated that when a blood vessel supplying the brain is blocked, up to 2 million brain cells are lost for each minute that passes, making stroke 1 of the most time-sensitive diagnoses in medicine.¹ The faster that blood flow can be restored to the brain, the more likely a person will have a full recovery. The Mobile Stroke Unit Program allows Stony Brook physicians to more rapidly identify if someone is a candidate for intravenous (IV) tissue plasminogen activator (tPA) or a mechanical thrombectomy—a procedure to remove clots that block large vessels. The American Heart Association and American Stroke Association now recommend for selected acute ischemic stroke patients to receive mechanical thrombectomy as the standard of care.² However, an individual must get to a hospital with the physician experts and technology to perform a mechanical thrombectomy early enough to get the most benefit from the treatment. The Mobile Stroke Unit Program will improve chances of good outcomes by providing earlier treatment to a greater number of patients.

Intervention Detail. The Mobile Stroke Unit program is community-based and structured to optimally serve our geography and patient demographics. The 2 units are positioned in 2 exits along the Long Island Expressway to provide service to the greatest number of patients in Suffolk County. The Mobile Stroke Unit Program currently serves 929 734 people (63% of the county population) distributed in 3 Suffolk County townships. Once a third unit is in service on Long Island’s East End, an additional 135 985 people (9% of the county population) will have access to our Mobile Stroke Unit services. We believe that this system of deployment will allow us to achieve the lowest rates of stroke-related morbidity and mortality in the United States. The Mobile Stroke Units bring Stony Brook University Hospital closer to our diverse populations. By deploying these strategically located, state-of-the-art Mobile Stroke Units, Stony Brook University Hospital brings proper care to more people faster. We have also launched public relations initiatives in Long Island to make the public aware of the program’s existence and the impact it has on people’s lives. Two videos are available for viewing that showcase the reception the Mobile Stroke Unit has received and the initial impact observed.^3,4

Outcomes and Impact. The Mobile Stroke Unit Program is a 2-fold initiative. IV tPA works best when given as soon as possible after stroke onset. By providing IV tPA at the scene where the patient is located, we are saving time and allowing the drug to work at its best potential. In addition, given the window of time in which the drug must be administered, more people may be eligible to receive it. Early administration of IV tPA can dissolve a dangerous blood clot, restore lost blood flow, prevent neuronal loss, and even completely reverse stroke-related deficits. Prior to the Mobile Stroke Unit Program, patients were transported directly to the closest hospital. However, those who suffered a stroke caused by a large vessel occlusion need to be treated at a comprehensive stroke center. Average door-in/door-out times from local emergency medical service (EMS) drop off to pick up for transfer to a comprehensive stroke center is 2 to 2.5 hours—sometimes as much as 6 hours. The Mobile Stroke Unit brings patients directly to the comprehensive stroke center, which saves hours of brain cells dying. This has resulted in a greater percentage of patients with 90-day modified Rankin scale improvement.

Calls from April 8, 2019 to December 31, 2020—2272 calls:

• Transported calls to hospitals: 750 calls (33%)
• Transport to Stony Brook: 533 (71%); to outside hospital: 217 (29%)
• Age range: 21–101, mean 72; gender: 53% female, 47% male
• Canceled/disregard/no transport/care transfer: 1476 (65%)
• Mobile Stroke Unit Standard EMS: 46 (2%) (nonstroke calls)
• IV tPA administered: 53 (7% transported calls); 43 to Stony Brook; 10 to outside hospital (51/53 diagnosis ischemic stroke)
• Discharged home: 72% (IV tPA only-26/36)
• Thrombectomy/thrombolysis cases to Stony Brook: 85 (39% of MSU Stony Brook ischemic stroke patients); 3 to outside hospital
• 730 feedback loop documents sent to 45 EMS agencies

1. Saver JL. Time is brain—quantified. Stroke. 2006;37:263–266.

2. Power WJ, Rabinstein AA, Ackerson T, et al. Guidelines for the early management of patients with acute ischemic stroke: 2019 update to the 2018 guidelines for the early management of acute ischemic stroke: a guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke. 2019;50:e344-e418.

3. PIX 11 News. The Mobile Stroke Unit’s First Patient: Stony Brook Medicine. 2019. https://www.youtube.com/watch?v=DqxnW0fAZu8.

4. CBS 2. Just Weeks After Deployment, Mobile Stroke Unit Savings Lives on Long Island. 2019. https://newyork.cbslocal.com/2019/05/20/stony-brook-university-hospital-mobile-stroke-unit/.

Keep Your Friends Close But Your Enemies Closer With Benchmarking

Laura Haubner, MD; Jennifer McCasland, MHA, LSSBB, CPHQ

Tampa General Hospital

Background. One of our top priorities is to become the safest academic health system in America by achieving world-class performance in all measures of quality. The Vizient Quality and Accountability (Q&A) Study scoring and ranking model is perhaps one of the best determinants to achieve this goal. We knew that using the Q&A ranking, we would be able to compare our performance to that of similar hospitals. Tampa General Hospital (TGH) felt the need to further incorporate quality into our strategic plans through a deployment strategy, allowing our work to be closely connected to the hospital’s mission and values. We needed a strategic and focused approach to our quality efforts that was thoughtful, would anticipate errors, would engage customers in the service design, and would create value-driven work. A domain oversight committee was formed to align quality and performance improvement with the Vizient domain structure. This governing body evolved into a governance structure that would set direction and strategy for improvement within each of the 6 Vizient quality domains (mortality, safety, effectiveness, efficiency, equity, and patient centeredness). Each domain consisted of a work group that was designed to report directly to the executive domain committee, allowing for direct leadership engagement and transparency. Through this initiative, we created a governance structure and strategic process that we used to design organizational systems to produce services and meet customer needs. While it may seem easy to build a governance structure, culture can be deeply rooted and hard to change. Through this process, we learned how to increase hospital transparency and leadership engagement. Data-driven decisions were also key to our success. Identifying process gaps and opportunities to improve quality outcomes using internal and peer benchmarking has been embedded into our culture.

Intervention Detail. Simultaneously, we put into place the foundations of the domain oversight committee and our 6 domain work groups (mortality, safety, effectiveness, efficiency, equity, and patient centeredness). The first step was to gather the appropriate team members for each group. The domain oversight committee consisted of hospital leadership and executives. Their primary functions were to help remove barriers, engage teams, and understand and support our quality system. Along with subject matter experts, each domain work group included a performance improvement specialist. The specialists assisted with data, identified areas of opportunity for committee decisions, presented opportunities to the oversight committee for prioritization, coached improvement work groups, and kept the domain oversight committee informed of progress and status of active work group improvement work. The next step included brainstorming, prioritizing, and selecting performance improvement initiatives. An inventory of all current work that aligned with each domain was compared against ideas from each driver diagram. A gap analysis between driver diagram and project inventory was conducted. All potential improvement ideas and projects were scored using an assessment scoring model following a thorough analysis of Vizient data. We began by testing any hypotheses that were thought to be potential root causes for unfavorable outcomes within each domain. Numerous statistical analyses were conducted using a year’s worth of patient-level and benchmarking data from Vizient. After determining which drivers had the biggest impact on each domain, final project selection was completed. Any new projects were commissioned, while those leading existing projects that were determined to have little to no impact were urged to continue efforts outside of this formal structure. Moving forward, these projects would have ongoing reviews with their respective domain committee to ensure progress.

Outcomes and Impact. At the start of the initiative, TGH was ranked at the 43rd percentile overall on the Vizient fiscal year 2019 Q&A scorecard. On the Vizient fiscal year 2020 Q&A scorecard, TGH had improved to the 18th percentile. Along with creating a goal for the overall Q&A score, the executive domain committee chose both annual goals and stretch goals for each domain work group. These goals were incorporated into TGH’s strategic plan and are reviewed annually at a minimum. From fiscal year 2019 to fiscal year 2020, the mortality domain improved from the 68th percentile to the 22nd percentile, efficiency improved from the 30th percentile to the 22nd percentile, and effectiveness improved from the 80th percentile to the 62nd percentile. While these domains exceeded fiscal 2020 year goals early, it was important for us to ensure that we had hardwired all reliable processes that needed to be achieved to maintain this performance. Although the equity and patient centeredness domains did not improve from 2019 to 2020, both exceeded established goals. This process allowed us to deeply engage in and understand our processes and outcomes.

ASPs Are the MVP: Leading the Way in Pandemic Response

Amy Brotherton, PharmD, AAHIVP, BCIDP, The Miriam Hospital, Lifespan; Diane Parente, PharmD, BCIDP, The Miriam Hospital and Rhode Island Hospital, Lifespan

The Miriam Hospital, Lifespan

Background. During an emerging outbreak with limited proven treatment options, there is an increased risk for inappropriate use of medications. Because antimicrobial stewardship programs (ASPs) are designed to optimize medication prescribing across a health system, Lifespan leadership looked to our ASP to rapidly provide guidance on appropriate allocation of scarce medication resources in a continuously evolving environment. Health care systems nationwide encountered multiple barriers when operationalizing therapeutic options for COVID-19 under emergency use authorization (EUA), due to limited data, strict requirements for use, and mandatory monitoring and reporting. Many institutions were unable to administer therapies, such as monoclonal antibody (MAb) infusions, due to significant logistical challenges in implementation. Lifespan’s antimicrobial stewardship (AMS) pharmacists played a vital role in the safe and appropriate prescribing of multiple therapeutic options for COVID-19 and developed innovative pathways to meet institutional goals. AMS pharmacists established new workflows for MAb administration, developed solutions for challenges, and monitored for continuous improvement in established processes. Our experiences provided us with a blueprint that may be reproducible in other health care systems and across multiple disciplines. We hope to demonstrate that ASPs are a valuable resource during a pandemic and can overcome challenges to meet the needs of both health care systems and the patients they serve. Strengthening ASPs is essential to maintaining core ASP practices while simultaneously tackling the challenges presented by a pandemic.

Intervention Detail. Early in the COVID-19 pandemic, Lifespan’s ASP was recognized by hospital leadership as the centralized hub for identifying and evaluating novel therapeutic options for COVID-19, as well as developing strategic plans for implementation, monitoring, reporting, and continuous quality improvement. For each novel therapeutic option, our ASP relied on core stewardship principles and a systematic approach for evaluation and implementation, including: (1) critical review of primary literature to perform a risks versus benefits analysis; (2) development of treatment guidelines, criteria for use, unacceptable uses, and preauthorization processes; (3) medication use operationalization through collaboration with key stakeholders; (4) systemwide education and rapid dissemination of new information; (5) prospective audit of medication use with real-time intervention and feedback; and (6) data collection, analysis, and reporting to identify trends and opportunities for process improvement. Through this strategic approach, our ASP was able to implement multiple initiatives to improve medication use throughout the health system.

Outcomes and Impact. From March 2020 to September 2021, our ASP developed and operationalized 11 treatment guidelines and order panels for therapeutics related to COVID-19. For inpatient therapeutics, AMS pharmacists reviewed 733 courses of remdesivir under Food and Drug Administration EUA for appropriate use in real time and reported these courses to the Department of Health weekly. As a result, our institution had a 99.5% adherence rate to the guideline criteria. For outpatient therapeutics, Lifespan became the first infusion site in the state to offer outpatient MAb therapy to Rhode Islanders at risk for hospitalization from COVID-19. Although preexisting infrastructure for administration of outpatient parenteral therapy was not in place, our ASP promptly developed a novel workflow to provide this service safely and effectively within 1 week of MAb EUA release. The Rhode Island National Guard and other Rhode Island health systems looked to Lifespan’s practice model for guidance on implementation and trends to determine opportunities for expansion. To date, over 2500 referrals for MAb therapy have been received and reviewed by our ASP, and 1732 infusions have been administered to qualifying patients. Hospital admission rates for COVID-19 for individuals who received MAb at Lifespan were 1.8%, compared to 10% reported in clinical trials for those who did not receive Mab. This highlights the significant impact of this ASP-driven initiative in reducing hospitalization rates and reducing the financial burden of COVID-19 on the health system.

Transforming Treatment of Opioid Use Disorder in the Emergency Department

James Chenoweth, MD, MAS; Daniel Colby, MD; Aimee Moulin, MD

UC Davis Health

Background. Opioid use disorder (OUD) is an increasing burden on US health care systems, with an annual cost of $15 billion each year.¹ These patients have frequent emergency department (ED) visits, long lengths of stay, and high rates of leaving against medical advice. OUD-related complaints are common in the ED, but treatment in this setting is often limited. OUD patients are also often the subjects of social stigma, which can bias ED care. Screening, brief intervention, and referral to treatment is a simple intervention that can be performed in the ED that has shown to increase outpatient treatment utilization and decrease hospital readmission rates. Medications for OUD, such as buprenorphine and methadone, have shown to increase the number of patients in treatment 30 days after discharge, while decreasing 1-year mortality in patients presenting to the ED after an opioid overdose.^2,3 Despite this, barriers to ED-based treatment for OUD remain. There are multiple reasons for this, including lack of experience with these medications, restrictions on these medications, continued stigmatization of substance use disorder (SUD), and concerns about obtaining adequate follow-up once a patient leaves the ED.

Intervention Detail. Our health system, led by our interdisciplinary SUD Committee, endeavored to increase access to treatment for patients presenting to the ED with OUD. A group intervention was performed in 2018 where all ED faculty received the in-person portion of the X-waiver training as part of an all-faculty educational retreat. We also designed a consult service called the Substance Use Intervention Team (SUIT) to aid with treatment recommendations and outpatient follow-up for SUD patients. This consult service consists primarily of substance use navigators (SUNs) and physicians who are board certified or board eligible in addiction medicine or medical toxicology. SUNs are drug and alcohol counselors who have developed expertise in the local treatment ecosystem and are a key innovation of our consult team. This service can be consulted by the ED treatment team at any point in a patient’s care. Most patients are initially evaluated by the SUN, who performs all aspects of the screening, brief intervention, and referral to treatment, including a warm hand-off to outpatient treatment providers. If the ED provider identifies the patient as being eligible for buprenorphine, they can start treatment in the ED if they are comfortable with this process or they can consult the physician on the SUIT service for assistance. This consultation involves in-person evaluation during which the risks and benefits of different treatments are discussed, a thorough substance use history is performed, and recommendations regarding timing of treatment and outpatient prescription recommendations are made. Patients are followed by the SUN team and contacted at 1 week and 1 month to determine if they are still in treatment and to help them work through barriers.

Outcomes and Impact. As a result of providing a dedicated setting for group X-waiver training, 73% of our physician group completed the process and received their X-waiver. This intervention, combined with the SUIT service (in particular the SUN’s ability to connect patients to treatment), has increased the number of patients prescribed buprenorphine at ED discharge from 123 in 2017 to 787 in 2020. Through the SUNs’ activities, 57% of patients started on buprenorphine in the ED have successfully obtained outpatient follow-up. This occurred despite the ongoing SARS-CoV-2 pandemic. Prior to the declaration of the pandemic in January and February, the follow-up rates were 71% and 70%, respectively. We are still assessing the impact on patient return visits, morbidity, and mortality. An additional benefit of this process was a significant decrease in stigmatization of OUD patients. This culture change is felt on both the physician and patient side, and we anticipate this will have added benefits to our providers and patients, while also leading to more equitable treatment of people with OUD. We plan to continue educational interventions for trainees, attending physicians, and all other treatment team members so that we can improve the ED treatment of patients with OUD.

1. Ronan MV, Herzig SJ. Hospitalizations related to opioid abuse/dependence and associated serious infections increased sharply, 2002-12. Health Aff (Millwood). 2016;35:832–837.

2. Mattick RP, Breen C, Kimber J, et al. Buprenorphine maintenance versus placebo or methadone maintenance for opioid dependence. Cochrane Database Syst Rev. 2014:CD002207.

3. Larochelle MR, Bernson D, Land T, et al. Medication for opioid use disorder after nonfatal opioid overdose and association with mortality: a cohort study. Ann Intern Med. 2018;169:137–145.

Change the Hospital’s Pain Prescribing Practices…No Problem, Go Team!

Cathy Lammers, MD; Eddie Eabisa, CDE; Christine A. Williams, RN; Myan Pham, MHA

UC Davis Health

Background. Root cause analyses of opioid-associated patient safety events revealed patterns of prescribing to be addressed and opportunities to drive change in nursing practices. We identified recurring systemic issues related to range orders, drug choice, dose stacking, lack of supportive medications, and variation in rescue parameters and responses. Prescribing trends had evolved such that intravenous (IV) hydromorphone became the first-line opioid, acetaminophen and nonsteroidals were generally ordered as needed for mild pain, and nonpharmacologic therapy was sporadically utilized. New Centers for Disease Control and Prevention recommendations, the opioid epidemic, and opioid shortages demanded a change in prescribing practices, but each department had its own order sets/preferences. We lacked a systematic way to evoke change. Concepts such as multimodal analgesia, “start low and go slow,” and oral route preferred over IV needed to be disseminated and utilized throughout the hospital. Nursing practices surrounding assessment/reassessment and communications with patients and prescribers also required updating. Current policies and patient care standards did not link reassessment to the route of the drug administered or the expected time of peak onset. Reassessment was limited to a pain score and did not address sedation/respiratory status. The sedation scale we used to assess for opioid side effects was better suited to assess intentional sedation, delirium, and agitation. We engaged our nurse leaders to drive change. Order sets accumulate in the hospital’s electronic health record (EHR). As new physicians and ideas emerge, new order sets are added. Many were written over a decade ago. Encouragingly, order set functionality is greatly improved, providing additional safety features such as clinical support tools to guide prescribers (eg, latest renal labs) and a reduced need to rewrite orders in common clinical transitions (eg, advancing diet). We invited our pharmacists and information technology experts to join our multidisciplinary team to educate us and build our vision.

Intervention Detail. With UC Davis Health executive sponsorship, we secured additional resources required for success. The opioid stewardship physician champion and executive director of patient care services established the inpatient pain task force. Members included opioid steward pharmacists, the EHR director, EHR pharmacists, the quality and safety manager, and physician/nurse/pharmacy representatives. Reports were created to understand how opioid orders originated (order set versus preference list) and what methods of opioid prescribing existed (range doses, routes, frequency). While analyzing the current state of ordering, we discovered that 75% of opioids were ordered outside of order sets as “one time only” orders. To secure engagement from ordering providers, we targeted our admission order sets for impact and ease of workflow. Each admission order set was modified to include a standard multimodal order panel that included opioids. The task force created standalone pain order sets (infant, pediatric, and adult) and also embedded them as panels into individual admission order sets. We leveraged existing EHR data to drive prescriber selection of the appropriate order set branch for renal function. Range doses were eliminated, and opioid doses instead appear in grouped regimens containing an as-needed (prn) moderate dose, a prn severe dose, and up to 3 rescue doses in 24 hours. To engage prescribers, we recruited physician champions by service line to help disseminate information and act as liaisons. The opioid stewardship physician champion conducted more than 60 meetings. In-person meetings were critical for our success and provided opportunity for prescribers to share day-to-day challenges and barriers. We addressed concerns in real time, demonstrating new and streamlined EHR functionality in a focused session. We conducted user acceptability testing sessions, presented at grand rounds, and engaged the Medical Staff Executive Committee. Nurse leaders and educators worked closely with us to align policy and patient care standards to: (1) include multidimensional pain assessments; (2) adopt the Pasero Opioid-Induced Sedation Scale at peak onset (by mouth versus IV); (3) expand respiratory assessments of sleeping patients; (4) adopt a rescue dose strategy to replace range orders; and (5) standardize patient education.

Outcomes and Impact. In total, we reviewed more than 160 order sets and retired more than 30 outdated order sets. We embedded or modified 114 admission order sets. A new pain policy was created, and we authored a pain guideline document to serve as a multidisciplinary resource on our intranet. Nursing teams on 2 units participated in a pilot and provided feedback for the final changes. We launched our pain policy and new pain prescribing panels/standalone order sets hospitalwide in July 2020. We received approval to launch the pain EHR changes outside of the EHR quarterly updates to allow staff to focus on these important updates. A multidisciplinary education module was assigned 2 weeks before go-live to socialize shared terminology, demonstrate EHR functionality, and provide each profession/discipline within the interprofessional team a glimpse of the other’s workflows. By August 2020, the module had been completed by 960 physicians, 1915 nurses (90%), and 68 pharmacists (85%). Post-launch, the percentage of opioids administered as range orders was reduced from almost 50% to less than 10% in just 6 months and oral opioid ranges were completely eliminated. The hospitalwide average number of IV opioid doses dispensed per month decreased by 21.3% and the average number of oral opioid doses decreased by 1.7%, despite hospital census increasing by 5.1%. The majority of all opioid orders are now written in the context of a multimodal pain order set or panel that starts with nonpharmacologic modalities, nonopioid pharmacologic medications and adds opioids only if needed. Our primary aim was to decrease physician/nurse/pharmacist burnout by fostering collaborative teamwork and meaningful verbal contact through more efficient EHR workflows, common language, and an understanding of each other’s intentions and challenges to achieve the common goal of safe, effective pain management for our patients.

Improving Sepsis Mortality With Dedicated Nurse Practitioners

Deborah D. Miller, DNP, FNP-C; Eric Gross, MD, MMM, FACEP; Stacy Hevener, RN, MSN, CSSGB, CPHQ

UC Davis Medical Center

Background. In keeping with national trends, the majority of patients diagnosed with sepsis at UC Davis Medical Center have sepsis present on admission. Although patients who develop new or worsening sepsis in the hospital make up a small share of our sepsis population, these patients account for a disproportionate number of our sepsis deaths. Early identification of sepsis is paramount in terms of increasing survival rates, but identification can be challenging in patients who are admitted for other complex concerns and for those with long lengths of stay. In addition, there is variability among provider teams regarding adherence to evidence-based sepsis treatment recommendations and willingness to initiate a higher level of care for at-risk patients outside of the intensive care unit. To decrease sepsis mortality on our adult acute care units, we designed and implemented a new nurse practitioner (NP) role dedicated to the care of patients at risk for severe sepsis and septic shock. The focus of the sepsis NPs was to identify these patients as early as possible through risk-stratified surveillance and to facilitate timely delivery of evidence-based sepsis care using a standardized workflow. The sepsis NPs have been active at UC Davis Medical Center for over a year, during which time a robust process was developed with stepwise expansion to include all adult acute care units.

Intervention Detail. Two sepsis NPs began evaluating patients on 4 acute care units in November 2019. This gradually increased by 2 to 4 units at a time, for a total of 15 units. Sepsis identification was initially based on elevated lactic acid results, but later expanded to include an electronic medical record-based predictive sepsis analytic model score. Automated pager alerts and an auto-populated list were built for patients with a sepsis analytic model score > 8 or lactic acid > 2. The NPs automatically consulted on patients meeting these criteria to evaluate them for potential severe sepsis/septic shock. A collaborative approach between the NPs and the primary teams was utilized for timely patient management. Two time periods were compared to evaluate the efficacy of the NP program: the 12-month baseline period (November 2018 through October 2019) and the 12-month implementation period (December 2019 through November 2020). Data were collected on patients with diagnoses of severe sepsis or septic shock. The key outcome measure was in-hospital mortality, and process measures included SEP-1 core measure bundle compliance and utilization of a dedicated sepsis order set.

Outcomes and Impact. During the baseline period, 145 patients met severe sepsis/septic shock criteria on adult acute care units, and 150 patients met the criteria the year following implementation. Fifty-two patients in the post-implementation group had an NP consult. Compared to patients with severe sepsis/septic shock during the baseline period, patients seen by the NPs had lower mortality (17% versus 21%), higher severe sepsis order set utilization (38% versus 25%), and higher compliance with the SEP-1 bundle (40% versus 31%). Compliance in that group was also higher for 5 of 6 bundle elements, including initial and repeat lactic acid (96% versus 81% and 92% versus 86%, respectively), blood cultures (65% versus 59%), fluid resuscitation (43% versus 36%), and vasopressor use (38% versus 25%). Antibiotic compliance did not change (71% pre/post). We often fail to act on sepsis early, especially for acute care patients hospitalized for other reasons. The sepsis NPs were effective care team extenders, and their consults were associated with decreased mortality and increased evidence-based treatment. There was no change in timely antibiotics, possibly because providers often defer to rescue care rather than preemptive care and are concerned about potential harmful effects of antibiotics such as Clostridioides difficile colitis and antibiotic resistance. Opportunities remain to educate providers about the benefits of early recognition and treatment of sepsis.

Saving Lives and Resources With a Sepsis Predictive Model

Georgia McGlynn, RN, MSN-CNL, CPHQ; Eric Gross, MD, MMM, FACEP; Taufa Lee, RN

UC Davis Medical Center

Background. Because half of the patients who die at UC Davis Medical Center have a sepsis diagnosis, reducing sepsis mortality has been the primary goal of the UC Davis Medical Center Sepsis Improvement Collaborative since its inception over a decade ago. The collaborative is a multidisciplinary group sponsored by the chief quality officer and nursing executive director with broad representation from nursing and medical staff, pharmacy, information technology, and health information management. The collaborative has 3 clinical co-leads from emergency medicine, hospital medicine, and nursing. Since 2011, automated screening tools based on systemic inflammatory response syndrome (SIRS) criteria formed the basis of our response to sepsis. However, the lack of specificity and low positive predictive value (PPV) of SIRS led to alert fatigue, wasted resources, and difficulty identifying patients who were truly at risk versus those experiencing other inflammatory processes. To address this, the collaborative explored the functionality of an electronic medical record (EMR)-based sepsis predictive analytic model (SAM). Other interventions, including the addition of 2 nurse practitioners dedicated to sepsis care, standardized order sets, widespread education, and other EMR enhancements were already in place during the baseline period (February 2020 through June 2020) prior to initiation of the SAM pilot, which ran from August 2020 to December 2020 on 2 acute care units.

Intervention Detail. The intervention consisted of 2 parts: First, assessment of the predictive value of the SAM score relative to SIRS, and second, development and implementation of a sepsis identification workflow based on the SAM. The analytic model automatically generates a risk score between 1 and 100 based on > 80 variables in the EMR, and the score updates every 15 minutes. We ran the model alongside the SIRS-based alerts for 6 months, silently calculating and filing scores for all patients > 18 in the emergency department and inpatient units. Using this data, we analyzed the predictive value of the SAM score at various thresholds compared to SIRS for 50 000 encounters. While the SIRS alert had 10% PPV and 82% specificity, a SAM score of > 4 had 16% PPV and 92% specificity, and a score of > 8 had 34% PPV and 98% specificity. Although SIRS had been used as a binary predictor solely to drive screening by the nurse, the use of a continuous predictor allowed us to develop a tiered response to patients at different levels of risk. We selected a threshold of > 4 to drive screening by the nurse and > 8 to alert the provider and trigger a bedside evaluation. We designed a pilot workflow with automated alerts that fire when these criteria are met and that suggest sepsis screening and treatment orders to patient care teams. Existing SIRS-based alerts were turned off and the new alerts were implemented on 2 acute care units between July 2020 and December 2020. We studied encounters on these units with discharge diagnoses of sepsis and clinical evidence of severe sepsis or shock. The primary outcome measure was in-hospital mortality, and secondary measures included sepsis order set utilization, compliance with the Centers for Medicare & Medicaid Services SEP-1 core measure, and the number of unique patients identified.

Outcomes and Impact. During the 5-month pilot period, 28 patient encounters with severe sepsis or septic shock were identified on the pilot units. The mortality rate for these patients was 18%, compared to 23% mortality for 16 encounters with severe sepsis and septic shock on the pilot units during the baseline period. Additionally, during the pilot period we observed higher utilization of the sepsis treatment order set (54% versus 31%) and increased SEP-1 core measure compliance (60% versus 44%). During the pilot, alert volume decreased by 15% and the number of patients identified with severe sepsis/shock increased by 9%. The PPV of the SAM > 4 alert was 21%, and 34% for the SAM > 8 alert. The pilot phase allowed for iterative improvements to the workflow driven by feedback from the direct care clinicians involved, and development of nursing and provider education and EMR reports that were leveraged to support implementation. SAM score implementation to drive sepsis assessment resulted in lower sepsis mortality and higher SEP-1 compliance on the 2 pilot units. Based on the success of the pilot, the SAM score was implemented in the remainder of the institution in January 2021. Comparing 6-month periods immediately prior to and after implementation, SEP-1 compliance improved from 35.5% to 39.1%. Severe sepsis and septic shock mortality decreased from 16.7% in the pre-implementation period to 14.1% post-implementation. Future analysis will focus on the SAM score’s performance in the intensive care unit population.

Using Lean to Improve Oral Chemotherapy Safety and Increase Revenue

Marc S. Hoffmann, MD; Debbie Fernandez, LMLP, MHSA, CPHQ; Marshall Johnson, PharmD; Jeff Wright, RT; Terry Tsue, MD; Leigh Anne Scott, PharmD

University of Kansas Cancer Center

Background. The University of Kansas Cancer Center provides oral chemotherapy treatment to more than 2000 unique patients across 8 metropolitan locations. Pharmacists are located in every oncology, hematology, and bone marrow transplant clinic, and on average, 1600 prescriptions are reviewed by our pharmacists monthly. In addition, our specialty pharmacy provides benefits investigation, as well as financial, medication, and grant and foundation assistance applications for more than 1000 oral chemotherapy patients per month. Historically at our institution, the management of oral chemotherapy was the sole responsibility of ordering physicians and advance practice providers—who were accountable for ordering, refilling, monitoring, education, and ensuring patient adherence. No formalized workflows existed to check for medication interactions; to verify appropriate dosing; to confirm if patients received or needed their medication; to verify if patients understood storage, administration, and disposal instructions; or to ensure that patients knew who to contact with any questions. In addition, there was no formalized clinical pharmacist or nurse involvement in the oral chemotherapy process. However, after a serious patient safety event involving oral chemotherapy in late 2016, the safe management of oral chemotherapy became a top priority. Patient safety, therapeutic outcomes, and financial opportunity are compromised without appropriate interdisciplinary processes to educate and monitor patients utilizing oral chemotherapy. The safety event prompted a comprehensive analysis of our oral chemotherapy prescription and delivery system with the goal of restructuring it to be safer and more efficient. This analysis led to initiatives that not only standardized practices to align with the same safety standards applied to parenteral chemotherapy, but also resulted in decreased prescribing errors, elimination of serious patient events, more efficient oral chemotherapy delivery, and growth of prescription capture and revenue for our specialty pharmacy. This presentation is an important part of the 2021 Vizient Connections Summit, as it provides a concrete illustration of how to coordinate strategic priorities, leadership support, and operational improvements that lead to successful patient outcomes.

Intervention Detail. Lean methodology, including 3 Kaizen workshops and iterative Plan-Do-Check-Act cycles, restructured our oral chemotherapy prescription and delivery system to be safer and more efficient, as evidenced by both process metrics and statistical analysis. Several specific interventions occurred. First—and perhaps most important, given that it was the foundation for all subsequent interventions—was that we mandated that all oral chemotherapy prescriptions be processed through our electronic medical record (EMR), and more specifically, the treatment plan module Beacon. This required not only a fair amount of provider education on how to operate the Beacon system to ensure that prescriptions were appropriately released, but also mistake-proofing by ensuring that oral chemotherapy orders generated by other means were automatically routed through the oncology pharmacy for review and entry into Beacon. Once all prescriptions were being processed in a similar manner and routed to the same pharmacy pool, it was possible to have pharmacy double-checks identical to those of parenteral chemotherapy. In addition to this double-check process, when a clinic nurse received a refill request, a safety check work standard was developed (to mimic the nursing double-check) and recorded in the EMR. We tracked utilization of the safety check and correlated it with defects in the prescribing process. We closely track oral chemotherapy orders outside of the EMR, nursing work standard utilization, prescribing errors, and serious events that may cause harm. When any of these processes show a concerning signal, as they have at various points since implementation, targeted strategies are employed to correct these defects. Specific data measures used to assess the performance gap include order processing productivity gains, quality defects related to consent signed, nursing safety check work standard adherence, and ongoing weekly monitoring of defects that reach oncology pharmacy.

Outcomes and Impact. Orders entered outside of the EMR decreased from approximately 20% of all prescriptions to zero within a few months. Over the last several years, a couple of prescriptions were processed outside of the EMR. When this happens, a comprehensive analysis is undertaken, the root cause is identified and corrected, and we return to zero orders processed outside of the EMR. Checklist utilization resulted in a remarkable decline in prescribing errors and events that could lead to patient harm. Among the 70% of patients who had a prescription refilled when a checklist was utilized, only 2 errors were observed. However, among the remaining 30% with whom the checklist was not utilized, 29 errors were observed (X² = 34.12; P < 0.001). Recognizing the value of the checklist in decreasing defects, our nursing education team led a comprehensive training to achieve 100% checklist utilization. This education effort led to near eradication of defects, and consequently, we now track defects using rare event analysis rather than control charts. Substantial improvement in pharmacy processing efficiency occurred, as lead time to verify all oral chemotherapy prescriptions decreased from 300 minutes to less than 60 minutes. Routing prescriptions universally through the EMR improved specialty pharmacy capture of oral chemotherapy prescriptions, increasing from approximately 15% prior to intervention to 55% now. Additionally, when prescriptions were mandated out by insurance, they were routed to 340(b)-contracted pharmacies to improve revenue by identifying gaps in contract pharmacy utilization. After interventions were put into place, utilization increased from 65% to more than 90% for all prescriptions eligible to be sent to a contracted location. The value of this work for oncology demonstrates a process for oral chemotherapy that is as safe as the traditional mode of intravenous chemotherapy. This work also demonstrates the applicability of Lean production and management systems within the health care sphere.

One and Done: Utilizing Dalbavancin to Create Value and Capacity

Abby Bailey, PharmD; Jeffrey Reynolds, MBA; Daniel Moore, MD, FACEP; Kathryn Ruf, PharmD

University of Kentucky HealthCare

Background. The trend of hospital and emergency department (ED) closures in parallel with increasing demand on such services has created severe and worsening overcrowding in EDs, particularly in referral centers. The ED serves as the front door for most hospital systems through which the majority of admitted patients and transfers enter, including our system. University of Kentucky (UK) HealthCare serves as a tertiary care center for much of central and eastern Kentucky and accepts the most transfers of any health care system in the nation. As a result, our hospital and ED suffer from severe inpatient boarding problems (annually > 360 000 h of ED boarding). However, we still turn down thousands of transfer requests annually, due to lack of capacity. Thus, as value-based delivery systems progress in the era of overcrowded hospitals, it is imperative that institutions find interventions that create value for all stakeholders while also eliminating waste and creating capacity from length of stay gains. Acute bacterial skin and skin structure infections (ABSSSI) are a common reason patients present to the ED. Historically, severe ABSSSI have required admission for intravenous antibiotics. Dalbavancin is a long-acting lipoglycopeptide that provides effective treatment for ABSSSI and helps avoid inpatient admission, due to its noninferiority compared to intravenous antibiotics and hospital admission. Disease-specific interventions such as this, where clinical outcomes are maintained while value and capacity are created, are worthy of future study and innovation.

Intervention Detail. Dalbavancin was approved for formulary addition at UK HealthCare in 2016. Patients eligible to receive dalbavancin include those who would otherwise be admitted for intravenous antibiotics for treatment due to infection severity, were ≥18 years old, and had ABSSSI lesion size ≥75 cm². Care coordination is achieved through multidisciplinary collaboration with ED providers, ED pharmacists, and representatives from antimicrobial stewardship. A financial analysis was performed to evaluate whether there were opportunities to improve capacity by expanding use of dalbavancin. Patients included in the dalbavancin group were identified based on pharmacy charges over the study period. Patients in the inpatient comparator group were identified based on diagnosis code (ICD10) and receipt of vancomycin for the treatment of nonnecrotizing ABSSSI. Patients were excluded from the inpatient comparator group if they required surgical intervention, admission to the intensive care unit, or had a length of stay > 30 days. In extracting data for the backfill analysis, patients were included if they were ≥ 18 years old and admitted to UK HealthCare as a transfer from an outside facility. Given an open ED bed could be filled by either medical or surgical patients, both were included in the backfill analysis. Additionally, patients were included in the backfill analysis regardless of whether they required intensive care unit level of care. It was assumed that backfill patients would be in the form of transfers because our ED maintains a left-without-being-seen rate of less than 0.5%—thus, new patient volume would be attributable to captured transfers.

Outcomes and Impact. From calendar year 2018 to calendar year 2020, UK HealthCare treated an average of 357 patients with a primary diagnosis of ABSSSI requiring admission; thus, possibly eligible to receive dalbavancin. Of those patients, 319 were admitted for treatment with an average length of stay of 4.24 days, or the equivalent of 1351 patient days. Only 10.6% (N = 38) of the cohort was treated with dalbavancin in the ED, consequently avoiding an inpatient admission. We identified multiple quantifiable benefits by treating patients with dalbavancin over inpatient admission. Financial analysis demonstrated all stakeholders could benefit from increasing use of dalbavancin. From the patient perspective, out-of-pocket expenses were similar between groups. However, the primary advantage is time saved outside the hospital. For those treated with dalbavancin who had subsequent readmission, they also experienced 1 less hospital day relative to the comparator group. Payors also benefited, as the cost for care was less in the intervention group. Avoiding admission resulted in an average reduced cost to the payor of $5800 per case. The ED and UK HealthCare, as a whole, benefited from capacity creation from inpatient length of stay gains, revenue generation from drug administration, and in the backfill of additional transfers secondary to inpatient capacity generation. Treating patients with dalbavancin created bed capacity by freeing up 4.24 patient days per patient treated. There was a direct cost savings of $1525 per case and the possibility of $981 967 in incremental margin if the opportunities are maximized in a future state. Because of new length of stay gains, inpatient capacity generation could allow for the backfill of additional transfers with a total maximum increase of 174 new patients to the system annually, and an increased inpatient capacity of 1214 patient days.

Microadjustments: Michigan Medicine’s Patient-Centered Approach to Balancing COVID-19 and Non-COVID-19 Care

Vikas Parekh, MD; Jennifer Pardo, MHSA; Max Garifullin, MS

University of Michigan Health—Michigan Medicine

Background. In October 2020, Michigan’s second wave of increased COVID-19 cases began. Across the state, hospital beds increasingly were filled with COVID-19-positive patients. At Michigan Medicine, we used lessons learned from the first wave on the risks of delaying care for our non-COVID-19 patients to establish principles for balancing access for both COVID-19 and non-COVID-19 patients. Core to our ability to do this was an understanding of the controllable and noncontrollable sources of patient volume, as well as effective tools to manage what is controllable through a series of informed microadjustments to controllable volumes. Michigan Medicine is a 1050-bed, 3-hospital academic medical center. Because COVID-19 impacted adult patients much more significantly than pediatric patients, we focused on deploying the management system in our adult hospitals. The volume of patients that enters the hospital through elective surgical admission is considered both controllable and predictable, since surgical cases are generally scheduled well in advance and with a known date of admission. Managing all controllable volume via microadjustments using the power of prediction allowed us to avoid having to make daily decisions on how many cases we needed to cancel, given our COVID-19 patient bed demand. This work is important and novel because it provides a real tool and management structure to manage emergencies that require reducing the bed capacity in which an institution operates while maintaining operational effectiveness and patient centeredness—and it is applicable even outside of the COVID-19 pandemic scenario.

Intervention Detail. To control elective surgical volumes, we used an innovative management tool called the Surgical Smart Scheduling Tool, which provides accurate predictions of downstream inpatient census for elective surgical patients. The tool is based on a “bed-base” concept, which describes the typical number of beds required to care for a patient population. Elective surgical patients are managed to fit within their bed base, representing the 85th percentile of historical census. The Surgical Smart Scheduling Tool outputs a predicted census for an upcoming day, given the scheduled admit volume on that day, as well as previously admitted patients who are not yet expected to be discharged. If predicted bed need exceeds the bed base for elective surgical patients, operating room (OR) leadership places a banner on the surgical scheduling application that indicates, “NO elective admit cases unless authorized by [OR director].” This banner is visible to any surgical scheduling staff and indicates the OR is closed for additional admit cases. Active management of surgical elective volume began when COVID-19-positive patient volume in intensive care units exceeded 25 beds. Using a run rate of the growth of COVID-19-positive patient census, we estimated expected growth of that census, finding that COVID-19-positive patients would need 15 more beds by the end of the 2-week period. We then dropped the bed base thresholds by that magnitude and placed banners on the days with expected census higher than the new bed base, which meant that no new admit cases could be scheduled on those days. COVID-19 census continued to rise, and after 3 weeks of managing under the revised bed base we lowered it again, this time by a magnitude of 10 beds. This was a nearly 20% decrease in the elective surgical admission bed base relative to pre-interventions.

Outcomes and Impact. We measured the impact of our interventions during the first 11 weeks after beginning active management of surgical elective volume under the lowered bed thresholds, representing November 1, 2020 through January 16, 2021. During this time period, the census of COVID-19-positive patients more than doubled, rising from 35 patients to more than 80 within 4 weeks. Our system of managing through microadjustments allowed us to absorb this volume without significantly impacting care delivery for non-COVID-19 patients. Excluding the 3 holiday weeks, average adult hospital occupancy was 88%, with no week during the date range having an average occupancy lower than 86% or higher than 90%. Average weekly emergency room boarding census remained between 10 and 20 over the duration of the measurement period. Although we placed strong controls on the number of surgical admit cases that could be done in 1 day, decreasing the number of inpatient beds available to elective surgical admits by 20%, OR utilization and the number of cases completed each week were also not significantly impacted by the interventions. Once a day was closed to surgical admit cases, surgeons were able to continue filling their block time with either outpatient cases or cases for patients who were already admitted. This intervention also allowed for minimal cancelations of scheduled cases because we were able to intervene, blocking a day to add-on cases before cases were placed on the schedule. This assessment is specific to our management of COVID-19 and non-COVID-19 care, but the predictive tools and management systems used can and have been applied to other situations when controls on elective volumes were necessary—specifically during major renovation of a cardiovascular intensive care unit and mitigation planning for a potential nursing staff stoppage of work.

Staff Up: A Flexible Nurse Staffing Model

Karen G. (Davis) Keady, PhD, RN, NEA-BC, University of Rochester; Mary Carey, PhD, RN, FAHA, FAAN, Strong Memorial Hospital

University of Rochester

Strong Memorial Hospital

Background. During the COVID-19 pandemic hospital surge, novel nurse staffing models were developed in response to higher hospital occupancy and acuity levels. The purpose of this abstract is to describe a nurse staffing program to increase bedside staffing using existing permanent nursing staff. This innovative program was developed to respond to short-term critical vacancy needs in times of consistent and higher-than-normal hospital volumes. During the first phase of the COVID-19 pandemic, nonurgent surgeries and procedures were placed on hold and those nurses were utilized to staff inpatient areas experiencing a surge in volume. To be able to continue offering surgeries and ambulatory procedures during the second surge, those same nurses were not available to bolster inpatient staffing. During the height of the pandemic, many organizations looked to external staffing agencies or travel nurses to meet staffing demands. This approach can be extremely costly and travel nurses may not be familiar with the organization’s policies and procedures. The disparity in compensation between permanent staff nurses and agency nurses can negatively impact morale and degrade teamwork. Many nurses picked up overtime shifts, but this was sporadic and voluntary, not allowing for labor projections across a scheduling cycle. We needed a process to recognize and reward existing nurses who were willing to commit to working additional hours over a defined period to provide adequate staffing and avoid bringing in additional travel nurses.

Intervention Detail. “Staff Up” was a time-limited, temporary program in which nurses were invited to commit to an extra 12-hour shift each week and were assigned to units with critical staffing needs. Importantly, the nurse’s skills and competencies were evaluated to ensure they matched requirements on the units in need of additional staff. For example, pediatric nurses were assessed to determine if they had the skills and competencies to take an assignment on an adult unit. Full- and part-time nurses were eligible to participate and were required to work an additional 12 hours above their standard hours, for a total of 72 hours over 6 consecutive weeks. Nurses received overtime for hours over 40 hours per week. Nurses were paid a premium in 4-hour increments of $65 for the day shift and $85 for the night shift for additional hours worked above their standard hours. Nurses who participated in this program worked their standard hours on their home unit as currently scheduled and were reassigned to other units for the additional shifts. At the end of the program period, the nurse received a completion payment of up to $1700 for registered nurses and up to $1200 for licensed practical nurses, prorated based on full-time equivalent status.

Outcomes and Impact. With the initial launch of the 6-week program, 167 nurses signed commitments (11 licensed practical nurse [7%] and 156 registered nurse [93%]); most (75%, n = 126) were 36-hour employees. The overall cost of the program was approximately $700 000, which was made up of overtime ($450 000) and completion payout ($250 000). The extra hours worked were equivalent to 50 incremental, full-time nurses. Had these hours been worked by contracted travel nurses the cost would have been approximately $2 million at current pandemic travel contract rates. Staff Up optimized existing permanent nursing staff, which ensured safe patient care because the nurses were already familiar with policy and guidelines without having to orient and train agency nurses. Nurses reported that they felt valued and appreciated because the organization was investing in them rather than looking to outside resources to meet staffing needs. Staff members were asked to commit to the program for 6 weeks so that schedules could be adjusted and staffing projections calculated. This novel nurse staffing model is administratively easy to implement and end as needed to meet the rapid needs of the ebb and flow of occupancy. It is also a fiscally responsible alternative to purchasing outside resources, which is especially important during hospital surges and nursing shortages.

Hospital and Home Care Collaboration for Recovering Critically Ill COVID-19 Patients

Kristi Filmore, MS, RN, ACNP-BC, University of Rochester Medical Center; Allison Panzer, MD, MPH, University of Rochester Medical Center; Anthony Pietropaoli, MD, MPH, UR Medicine; Christine Groth, Pharm D, FCCM, BCCCP, UR Medicine; Thomas Caprio, MD, MPH, MS, UR Medicine Home Care; Jane Shukitis, RN, BSN, MPA, UR Medicine Home Care; Stephanie Chalupa, MS, RN, UR Medicine Home Care; Bethany Rague, MS, RN, UR Medicine Home Care; E. Kate Valcin, DNP, RN, CCRN-K, NEA-BC, CNL, UR Medicine

University of Rochester Medical Center

Background. Post-intensive care syndrome (PICS) refers to new or worsening physical, cognitive, or mental health impairments that arise after critical illness and persist beyond acute care hospitalization.¹ Up to 80% of patients surviving acute respiratory failure in an intensive care unit (ICU) experience PICS.² When the COVID-19 pandemic began in spring 2020, there was an unprecedented rise in acute respiratory failure and critical illness,² creating a large patient population at risk for PICS. Collaborators at an academic medical center in upstate New York sought to meet the needs of COVID-19 patients at risk for PICS through the development of the Critical Illness Recovery Program, a collaborative pilot project focused on transitions of care. The program goal is to eliminate barriers to care by bringing interventions to patients in their home and via telemedicine, including social work, mental health screening, therapy services, pharmacist pharmacotherapy consultation, and evaluation by a critical care physician. The model merges traditional home care services with virtual visits to provide timely screening and collaborative interventions in the patient’s home, removing logistical barriers and thereby optimizing both patient engagement and social distancing. It addresses access barriers to an existing, outpatient PICS clinic while also adapting to the unique constraints of the COVID-19 pandemic. This program was quickly brought to fruition by leveraging a preexisting, strong working relationship between the hospital and its affiliated, certified, home health agency. A multidisciplinary team was led by the PICS clinic medical director with support from a performance improvement coach. Medical center and home care leaders provided upfront commitment and leeway for creative problem-solving.

Intervention Detail. A Lean Six Sigma framework was used to guide the project. An interdisciplinary work group convened to evaluate existing resources and developed innovative ways to meet patient needs during the pandemic. Protocols were adapted from the preexisting PICS clinic (closed as a result of COVID-19) and new workflows were developed to safely and effectively deliver care via a hybrid remote and home-based care model. Patient screening for PICS occurs in the ICU. Standardized communication templates, pathways, and data reports were created to identify and track eligible patients. Home care referrals are initiated in the ICU and patients are followed by the home care agency throughout hospitalization and discharge. Social work, nursing care, mental health screening, and therapy services (including physical, occupational, and speech) are provided in the home within the first 2 weeks after discharge. A 14-day telemedicine visit is held with the patient, home care nurse, critical care pharmacist, social worker, and critical care physician. The patient is referred to additional services as appropriate and a final telemedicine visit is planned for 90 days postdischarge. The program was piloted with critically ill COVID-19 patients in June 2020, just 5 weeks after the first planning meeting. Plan-Do-Study-Act cycles have been used to promote continuous improvement of the model. In May 2021, the pilot was expanded to screen all medical ICU patients (COVID-19 and non-COVID-19) for participation in the program. Primary outcome measures include patient participation and 30-day hospital readmissions.

Outcomes and Impact. As of June 2021, 86 patients met inclusion criteria upon discharge home from the hospital or rehabilitation facility. Of those, 71 patients (83%) accepted some or all home care services and completed the initial telemedicine visit. This is a considerable increase in engagement compared to the in-person clinic, which had an estimated patient participation rate of 10%. Approximately 40% of patients were referred for specialty care follow-up during the initial telehealth visit, most frequently to behavioral health and pulmonology. Almost all patients benefited from at least 1 pharmacy intervention, such as medication counseling or identification of vaccine omissions. Of the 71 patients who completed the initial telehealth visit, 5 (7%) had a readmission within 30 days of discharge home. This is lower than observed 30-day readmission rates of 15% after discharge from an ICU³ and 12% for patients discharged with support from home health after COVID-related illness.⁴ A 3-month telehealth visit was recommended for 64 patients (90%); 10% of patients were well enough that a second visit was deemed unnecessary. Thus far, 81% of scheduled patients have completed the follow-up visit. The increase in participation for both initial and follow-up visits is attributed to the elimination of logistical barriers inherent to a center-based model. Beyond these quantitative outcomes, there has been improved communication and responsiveness among team members, patients, and families. Use of telemedicine in conjunction with the in-person home care visit allows the team to conduct “house calls” with real-time data and assessments, placing the patient at the center of care. This approach has been fulfilling for team members, and in the words of 1 physician—“transformative.”

1. Needham DM, Davidson J, Cohen H, et al. Improving long-term outcomes after discharge from intensive care unit: report from a stakeholders’ conference. Crit Care Med. 2012;40:502–509.

2. Hosey MM, Needham DM. Survivorship after COVID-19 ICU stay. Nat Rev Dis Primers. 2020;6:60.

3. Hirshberg EL, Wilson EL, Stanfield V, et al. Impact of critical illness on resource utilization: a comparison of use in the year before and after ICU admission. Crit Care Med. 2019;47:1497–1504.

4. Lavery AM, Preston LE, Ko JY, et al. Characteristics of hospitalized COVID-19 patients discharged and experiencing same-hospital readmission — United States, March-August 2020. Morb Mortal Wkly Rep (MMWR). 2020;69:1695–1699.

The Pursuing Excellence Initiative: Getting Patients Better Faster

Michael Apostolakos, MD; Diane Hartmann, MD; Stephanie Northwood, MPH

University of Rochester Medical Center

Background. In 2016, the University of Rochester Medical Center (URMC) was selected by the Accreditation Council for Graduate Medical Education as 1 of 8 physician-training sites in the country to take the lead in improving the way residents and fellows learn quality and safety practices. The national initiative was fueled by Accreditation Council for Graduate Medical Education studies showing that most physicians nationwide are not acquiring the quality improvement skills they need during their training to practice and lead the team-based, outcomes-driven care crucial to the success of our health care system. Studies also showed that residents who do learn quality improvement skills in their training carry those with them throughout their careers, which is necessary to improve patient safety and quality of care and essential to sustaining the future of health care.¹ University of Rochester Pursuing Excellence, launched in fall 2017, leveraged the existing quality improvement infrastructure of the Unit-Based Performance Program, in which teams focus on improvement projects at the local unit level as a method to catalyze institutionwide improvement. Leaders from various disciplines came together to design and implement University of Rochester Pursuing Excellence, including: the chief medical officer; physician leadership; the operations excellence performance improvement department; quality and safety; graduate medical education; workplace learning; nursing; and the Institute for Innovative Education, with support from executive leadership (URMC and Strong Memorial Hospital chief executive officers, chief operating officer, and chief nursing officer). We believe our work can contribute to the 2021 Vizient Connections Summit by encouraging other institutions to integrate interprofessional teamwork with addressing significant clinical issues.

Intervention Detail. The intervention focused on the goals of creating a cadre of highly skilled clinical leaders (medical directors and nurse managers) working in functional teams to achieve the program mission of “better teams, better care.” The approach is trifold and includes: (1) building leadership skills to change culture; (2) developing high-functioning teams; and (3) incorporating foundational quality improvement skills into daily work within the clinical environment. The first step was to implement an interdisciplinary leadership development program for clinical leaders to develop their skills in leading teams and change initiatives necessary in quality improvement, while also playing an active role in transforming the health care learning environment. The leadership development program is deployed via an interactive workshop, which includes offline work, a 2-day retreat followed by 3 additional 3-hour retreats, implementation of team-specific initiatives, and ongoing coaching over the course of a year. Cohorts start every 15 months and consist of 12 to 15 teams. Team development builds upon our existing Unit-Based Performance Program, led by clinicians trained in the leadership development program. The goal is to create high-performing, interprofessional teams, incorporating quality improvement principles into the daily work of delivering the highest-quality care. In addition to incorporating these principles and skills into their daily work, team members are trained in DiSC, Work of Leaders, and Patrick Lencioni’s The Five Dysfunctions of a Team during the 2-day retreat, and themes from this tool kit are carried through at quarterly follow-up sessions where teams are provided with additional education on quality improvement, leadership, and teambuilding skills. Foundational quality improvement learning integrates existing URMC curricula into a cohesive learning framework. Curriculum content centers on available, evidence-based content developed by the Institute for Healthcare Improvement, as well as the Rochester Improvement Science Education curriculum, an original framework developed by URMC faculty.

Outcomes and Impact. Each team identifies a specific quality improvement project with an overarching focus of patient flow, employing a variety of process improvement tools such as Plan-Do-Study-Act cycles, root cause analysis, data analysis, aim statements, outcome measures, and graphical presentation. At each follow-up session, in addition to receiving educational material, teams present the progress of their projects and receive feedback from hospital leaders representing various functional areas. At 2 years post-program completion, the teams involved in the first Pursuing Excellence cohort collectively reduced their average length of stay (ALOS) by 6 to 7 hours, meaning that 1 in 4 patients goes home a day early—yielding an average of 4.3 additional daily available beds. This generates capacity for URMC to care for 293 more patients annually. Furthering the mission to get patients better quicker, as ALOS decreases, so do readmissions. Cohort I reduced readmissions by 1.6%. Cohort II has had similar success in its first-year post-program completion. ALOS decreased by 3 hours, meaning 1 in 8 patients goes home 1 day early, and 2.5 additional beds are available daily, on average. Annually, these numbers allow for 151 more patients to receive care, and readmissions have decreased by 1.6%. These results demonstrate that the methods behind this program are sustainable and replicable.

1. Wagner R, Weiss KB, Passiment ML, et al. Pursuing excellence in clinical learning environments. J Grad Med Educ. 2016;8:124–127.

Hungry for Change in Malnutrition Documentation Practices

Salsawit Shifarraw, MS, University of Texas Health Science Center at Houston; Jolyn S. Taylor, MD, MPH, MD, Anderson Cancer Center; Swaroop Gantela, MD, University of Texas Health Science Center at Houston; Robert E. Murphy, MD, University of Texas Health Science Center at Houston; Ankita Podichetty, BS, MS, University of Texas Health Science Center at Houston

University of Texas Health Science Center at Houston

MD Anderson Cancer Center

Background. Vizient identifies a case as having malnutrition if at least 1 of the ICD-10 diagnosis codes defined within the Vizient malnutrition risk variable appears with presence on admission indicated. During a University of Texas Health Intelligence Platform analysis of ICD-10 codes using encounter-level data from the Vizient Clinical Data Base, we identified several malnutrition ICD-10 codes as being documented at a significantly lower rate at MD Anderson Cancer Center (MDA), compared to other Alliance of Dedicated Cancer Centers (ADCC) members. As malnutrition is associated with decreased radiation and chemotherapy tolerance, decreased recovery, and lower overall chances of survival among oncology patients, this prompted a deep dive into malnutrition documentation at MDA, where we found that initiation and criteria for malnutrition referrals was inconsistent across departments. Targeted interventions to improve malnutrition coding through education and analysis of patient outcomes data from the Vizient Clinical Data Base began. This multidisciplinary project involved establishing relationships between physicians and nutritionists and improving malnutrition screening and documentation for malnutrition in the patient plan of care. The interventions implemented were in accordance with best practices determined by the American College of Surgeons. Project progress was tracked through monthly reports focusing on the percentage of patients with documented comorbidities. This project is an important part of the Vizient Connections Summit because it successfully demonstrated that Vizient data can be used to drive meaningful change in documentation and coding practices, change in culture, and improvement in patient outcomes. Nutrition status is a major determinant of outcomes for any type of surgery, especially for high-risk oncology patients. This initiative led to improved recognition of malnutrition and implementation of practices that increased the number of patients identified to have malnutrition prior to surgery, and allowed providers and nutritionists to intervene in a timely manner. We are glad to share our journey of improving malnutrition coding for the benefit of oncology patients.

Intervention Detail. The Vizient risk models reflect the importance of malnutrition by including it as a significant variable in almost every risk model for mortality, length of stay (LOS), and cost. We analyzed the rate of documentation for malnutrition ICD-10 codes for each model group level to ensure that similar patient groups across hospitals were being compared using analysis of proportions. Low outliers were identified using patient-level data from 2018 and 2019, which was downloaded using the Vizient Clinical Data Base. We found that the malnutrition ICD-10 codes were low outliers for a majority of the model groups. Malnutrition ICD-10 codes were generally documented at a significantly lower rate at MDA, compared to other ADCCs. We established an interdisciplinary, physician-led team of nutritionists, physicians, and analysts to address the low rate of malnutrition documentation and coding identified by University of Texas Health Intelligence Platform. During outpatient pre-surgery visits, a nutritionist would screen patients for malnutrition and place a note in the patient’s problem list if malnutrition was identified. This was then reviewed by the surgeon, who would document the diagnosis and plan of care. To track progress of this model, a monthly patient outcomes report focusing on the percentage of patients with comorbidities was shared with each department within the division of surgery. These reports provided data for the percentage of patients documented to have malnutrition each month at MDA for the division of surgery and within each Vizient service line. These reports also provided data on the percentage of cases with comorbidities with trends for LOS index, mortality index, mean number of diagnoses per case, and percentage of cases with no comorbidities. Through education, collaboration, monthly reporting, and transparent data review, the team was able to improve documentation and coding for the malnutrition diagnosis codes.

Outcomes and Impact. The percentage of cases with a documented malnutrition ICD-10 code for the division of surgery increased from an average of 5.3% in 2019 to 17.8% in November 2020, which implies that malnutrition identification, coding, and documentation improved as a result of the intervention. In year 1, according to our analysis of proportions, MDA had 26 ICD10-model group combinations that were coded and documented significantly below the ADCC average. In year 2, there were 19 codes below average. The most improvement was seen with malnutrition ICD-10 code E46 (9 model groups in year 1 to 2 model groups in year 2). Some key patient outcomes metrics also showed improvements for the oncology subservice line, as the LOS index decreased by 12%, the mortality index decreased by 37%, and the percentage of cases with no comorbidities decreased by 31%. Criteria for moderate and severe malnutrition were defined and providers were trained on how to add them to outpatient problem lists prior to surgery. Currently, a best practice advisory panel is being built into Epic to further improve malnutrition documentation. The MDA team continues to review monthly malnutrition rates to see if more patients are being identified. Overall, this initiative led to improved recognition of malnutrition in MDA cancer patients planning to undergo a surgical procedure. Discussing and screening for malnutrition is now incorporated into the preoperative assessment, allowing more at-risk patients to be identified. Each identified patient then receives a nutrition consultation, and MDA’s nutritionists create for them a plan of care with the goal to improve their overall nutrition and postoperative recovery.

Pathway to Success: Optimizing Employee Prescription Benefits for Sustainable Pharmacy Growth

Breanna Meinzer, PharmD, BCACP; Holly Smith, RPh, MBA

The University of Toledo Medical Center

Background. The University of Toledo Medical Center (UTMC), Toledo, Ohio, has established a robust and integrative approach to provide pharmacy services to employees and patients over the years. The 2006 merger of The University of Toledo and the Medical University of Ohio created the third-largest public operating budget in the state of Ohio, while also establishing a land of opportunity for learners, employees, and growth in patient-centered services. Despite being the only publicly owned, single-entity hospital/health system in Toledo, UTMC’s pharmacy enterprise has become a major player, providing key service lines in the northwest Ohio region. At the heart of UTMC’s success has been the UT-sponsored employee prescription benefit that has laid the way for the foundation’s success and expansion of pharmacy services. The UToledo prescription benefit covers approximately 10 000 lives and steers members to 3 on-site UT pharmacies leveraging cost savings as a pass-through model. With its 2007 inception, the cost savings model has seen substantial growth over time, including: (1) an additional on-site hybrid outpatient/specialty pharmacy; (2) a Ryan White 340B pharmacy; (3) additional staffing, including a managed care clinical pharmacist; and (4) optimized 340B savings in conjunction with employee prescription capture. In addition to growth of pharmacy services, UTMC created a strategic partnership with its pharmacy benefit manager (PBM) to optimize benefits and provide additional cost-savings opportunities. This innovative approach serves to capture all savings through formulary management, employee copay incentives, plan benefit design strategies, and manufacturer rebates. Future plans to enhance and streamline the prescription benefit include partnering with The University of Toledo Rocket Wellness team, as well as the expansion of pharmacy-driven, employee clinical disease state management to further strengthen existing relationships with UTMC providers.

Intervention Detail. In 2007, drug prices were predicted to rise over time, with a large emphasis on trends pertaining to specialty medications. The 2006 merger of The University of Toledo and Medical University of Ohio created an academic-affiliated medical center with a large, combined employee university/health science population. In 2007, a cost savings model through employee prescription benefit services was implemented. Over time, the prescription benefit has grown to cover nearly 100 000 prescriptions annually, undergoing many plan design optimization tools and growth strategies. As the number of prescriptions filled has more than doubled over time, the outpatient and ambulatory pharmacy enterprise at UTMC has experienced unprecedented growth in numbers and services. Three new pharmacies encompassing outpatient and specialty pharmacy have been constructed in an effort to increase capture rate, enhance patient care, and provide clinical learning opportunities. In 2016, UTMC started a specialty pharmacy after trends identified a 15% to 20% increase in specialty prescription spend from 2014 to 2016. In 2018, a new PBM provided additional cost savings, including lower administrative and clinical fees and increased rebate offerings. This contract also included a managed care clinical pharmacist position that facilitated close strategic oversight of the prescription benefit and partnership with the PBM to analyze and maximize cost savings drivers. As a result, the managed care clinical pharmacist position provides proactive outreach to members by troubleshooting formulary changes; offering diabetes prevention programming; and serving as the liaison between human resources, the PBM, and employees. On a quarterly basis, trends in prescription drug spend are analyzed to identify opportunities for intervention.

Outcomes and Impact. Prior to 2021, the capture rate of employee prescriptions remained steady at 79%, and has since increased to a record 92% by utilizing plan optimization tools with day supply limits. Copay incentives, as well as distributing prescriptions by mail and via curbside, also contributed to the rise in employee prescription capture. With a cost-savings model, the prescription benefit has saved around $1.7 million annually by utilizing in-house pharmacies compared to contracted network pharmacies. With an increase in employee prescriptions filled at UTMC pharmacies, the prescription benefit is projected to reach $2 million in savings. The creation of the specialty pharmacy proved to be a beneficial decision, as specialty spend has skyrocketed even though the UToledo employee prescription specialty spend has remained below average projections. By leveraging the prescription benefit to steer members to the UT Access specialty pharmacy, the benefit was able to prepare the foundation for the expansion of all specialty prescriptions and in doing so, generated record-breaking revenues and exceeding budget. This pharmacy is well-equipped and staffed to serve all specialty patients, including employees. It maintains an average prior authorization turnaround time of less than 17 hours and has established a comprehensive patient assistance program with an average annual savings of $6 million. In addition to the success of the specialty pharmacy, the prescription benefit program furthered savings through its partnership with the new PBM. The 2018 contract generated an additional $2.5 million in cost savings and the early contract renewal in 2020 generated over $300 000 in savings within the first year. As our clinical pharmacists continue collaborating with UTMC providers, consulting agreements have aided in the pharmacist management of patients, including employees. Looking toward the future, an emphasis will be placed on chronic disease management and preventative medicine to provide simultaneous savings in prescription and medical benefit costs.

Frontline Nurse Leader Development Program: Keys to Success

Juan M. Hernandez, RN, MSN; Michael D. Danielson, MBA; Michael L. Sanchez, MBA, BSN, RN; Scott Christensen, DNP, MBA, APRN

University of Utah Health

Background. Nurse managers perform an essential role in health care organizations by overseeing clinical operations, instituting leadership initiatives, and taking accountability for organizational performance metrics. The problem for nurse managers is that they are both asked to field top-down requests from executives, supervisors, coworkers, and clinical leaders, while also maintaining bottom-up operations in overseeing multiple employees. This demand includes adapting to emerging needs of the nursing workforce—including higher patient volumes and acuity with nursing teams that are becoming younger and less experienced. An important approach for alleviating the workload of nurse managers is to shift some responsibilities to frontline nurse leaders, including nurse coordinators, charge nurses, and nurse educators. Growing, developing, and supporting these leaders has shown to improve everything from nursing turnover to staff engagement and clinical outcomes. The challenge at University of Utah Health was that most system training and development resources targeted nurse managers. In contrast, development efforts for frontline leaders were disparate and highly dependent on the individual manager. If we truly wanted to offload responsibilities from the nurse manager’s workload, we needed to improve the way we designed, equipped, and grew frontline leaders’ duties and skillsets. The purpose of the Nursing Leadership Principles program was to serve as a comprehensive formal system for developing the leadership skills of our frontline leaders. The significance of nurse manager leadership has been emphasized by health organization leaders and there is broad interest in supporting nurse managers to be successful. We are proud to share this abstract’s findings as part of the 2021 Vizient Connections Summit to promote alleviating nurse manager workload through empowered frontline nurse leaders.

Intervention Detail. We developed program content by conducting a literature review to identify evidence-based practice for promoting frontline nursing leadership skills while also seeking feedback from organizational leaders and stakeholders. This review helped us form a training curriculum for the Nursing Leadership Principles program, with leadership topics that included communication, coaching, inspiring others, adaptability, decision-making, conflict resolution, and an assessment of individual strengths through the StrengthsFinder assessment. We engaged organizational stakeholders as the last step in finalizing course content, including training sessions for nurse managers and a pilot group of frontline leaders, as well as a focus group session in which we gathered their feedback. We designed the training with a formalized classroom component and corresponding online modules. This hybrid learning approach called for participants to learn basic concepts at their own pace through online courses before attending a subsequent in-person session to practice and translate knowledge to their day-to-day work. We developed content so nursing scenarios or situations connected directly to a frontline leader’s workplace experiences. The course outline included an initial online module and 1-day training, followed by monthly modules/in-person sessions over 8 months. We broadly identified those who could participate, offering several class sessions and cohorts to include as many participants as possible. We implemented the program in 3 cohorts, spanning 2017 through 2019. We used a pre- and post-questionnaire design to measure participant perception using 5-point Likert scale questionnaire items based on the Kouzes and Posner Leadership Participation Inventory Model of Transformational Leadership. We administered the survey 2 weeks before and 2 weeks after initial training. We also gathered course evaluation feedback after participants completed each module. As a final measure, we monitored the participant employment turnover rate 12 months after their respective completion of the course. The analysis included descriptive statistics, the Mann-Whitney U test, and Cliff’s Delta test for effect size.

Outcomes and Impact. The pre-post questionnaire measured the leadership skills and capabilities of participants (n = 526). Frontline leaders reported increased leadership, proficiency, and ability, with the vast majority of participants reporting increased usage of leadership behaviors in the post-survey. Some differences between pre- and post-responses were noteworthy. Post-implementation survey results suggested participants used novel approaches to lead and connect with employees (P = 2.543 × 10^–16; d = 0.28), had a clear understanding of their role as a nursing leader (P < 2.2 × 10^–16; d = 0.272), and created a shared vision with those they taught (P = 5.653 × 10^–15; d = 0.260). The impact of these findings was echoed in course feedback from participants. When asked at the conclusion of the program to rank their agreement, 86.1% of participants agreed with, “this program helped prepare me to better supervise and/or lead staff,” 80.1% agreed with “this program helped better prepare me to lead day-to-day patient care issues,” and 78.1% agreed with “I am a better decision-maker after participating in the program.” Finally, we evaluated the turnover rate for participants 12 months after their respective program completion. We found that only 2.5% of program participants turned over from their job positions during the 12 months after implementation, while the typical annual turnover rate for the nursing job family at the project facility was 10.5%.

Pathway-Driven Care in an APP-Led Oncology Acute Care Clinic

Sadaf Charania, MPAS, PA-C, UT Southwestern Simmons Comprehensive Cancer Center; John W. Sweetenham, MD, FRCP, FACP, FASCO, UT Southwestern Simmons Comprehensive Cancer Center; Angela Bazzell, DNP, APRN, FNP-BC, AOCNP, UT Southwestern Simmons Comprehensive Cancer Center; Edie Brucker, MSN, MPH, APRN, AGPCNP-BC, AONCP, UT Southwestern Medical Center; Judith Devlin, MS, ANP-C, AOCNP, UT Southwestern Medical Center; Shayna Simon, PharmD, BCOP, UT Southwestern Medical Center

UT Southwestern Simmons Comprehensive Cancer Center

Background. Emergency department (ED) visits for cancer-related complications lead to delays in cancer treatment, compromise quality of care, negatively impact patient experience, and increase costs. Immediate care clinics have been implemented by cancer centers to provide an alternative model of care while reducing ED visits. Overcrowded EDs not only expose immunocompromised patients to community infections, but oncology patients often require specialized care—a challenge for providers unfamiliar with cancer and its treatment. Coordination of care with the primary oncology team can also require time that is not readily or easily available in a fast-paced ED environment. According to UT Southwestern Simmons Comprehensive Cancer Center 2016-2019 Medicare claims data, the rate of ED utilization for our patients was high at the 90th percentile compared to other cancer practices nationwide. Common presenting symptoms in the UT Southwestern ED were gastrointestinal complaints, fever, and dehydration, which can all be managed in an ambulatory setting. To avoid unnecessary ED use, we implemented an advanced practice provider (APP)-led oncology acute care clinic. Standardized evidence-based care pathways were developed to ensure high-quality care.

Intervention Detail. A multidisciplinary team of APPs, nurses, pharmacists, physicians, and cancer center leaders were engaged to decrease unnecessary ED use by our cancer patients. Factors contributing to ED visits were identified using a causal analysis. A literature review and 9 conditions identified by the Centers for Medicare & Medicaid Services provided the foundation for development of clinical pathways, based on National Comprehensive Cancer Network and American Society of Clinical Oncology guidelines, as well as UT Southwestern protocols. The pathways were approved by a multidisciplinary group before finalization, which included pharmacists and disease teams. A review of ED use by our patients led us to establish Monday through Friday from 7 am to 7 pm as the clinic hours likely to be most impactful. Patients were referred by primary oncology teams and an electronic medical record process was established to capture overnight, on-call referrals. Order sets reflecting clinical pathways were included in the electronic medical record to ensure evidence-based and efficient care. A patient discharge follow-up process was established and a data collection plan implemented to measure outcomes. Collaboration with multiple departments, including imaging, laboratory services, and bed control, was necessary to successfully facilitate patient-centric workflows. Feedback from the patient family advisory council was integrated into patient communications. For patients presenting with symptoms similar to COVID-19, a rapid COVID-19 test was obtained prior to scheduling same-day appointments.

Outcomes and Impact. From August 2020 to July 2021, 659 clinic patient visits were completed. Of the visits, 524 patients were discharged home, 82 patients were directly admitted to the hospital, and 53 patients were transferred to the ED for a higher level of care. Approximately 80% of the patients managed were discharged home. Evaluating oncology patients in the ambulatory setting reduced admissions and unnecessary ED visits. A quarterly audit of clinical pathway use from August 2020 to June 2021 found that of 600 patients seen, 351 (59%) had a clinical pathway correlating to their chief complaint. When a clinical pathway was available, 95% pathway compliance was achieved. Gastrointestinal manifestations (28%), hypovolemia/weakness (19%), and infections (12%) were the most frequently encountered symptoms. The use of clinical pathways standardized diagnostic work-up and treatment, while enhancing physician acceptance of the new service. The pathways also allowed APPs to practice at the top of their license. A multidisciplinary team reviews and updates the clinical pathways quarterly to ensure current evidence-based practices. Audits were implemented to evaluate pathway use and optimize patient outcomes, as well as determine areas of improvement and growth. Opportunities identified included creating additional pathways for cardiac symptoms, mouth disorders, and infections unrelated to neutropenia. Supporting oncology patients in the ambulatory setting reduced admissions and unnecessary ED visits, leading to cost savings/avoidance for patients and the health system. Patient experience data from August 2021 showed greater than 95% overall patient satisfaction. The highest-rated scores related to timely access to care and evaluation in a familiar environment. Additional studies are underway to assess costs savings, as well as the economic impact on health care resources.

Human-Centered Design Approach to COVID-19 PPE Challenges

Sarah E. Henrickson Parker, PhD, Virginia Tech Carilion School of Medicine, Carilion Clinic; Patrice M. Weiss, MD, FACOG,, Carilion Clinic, Virginia Tech Carilion School of Medicine

Carilion Clinic

Virginia Tech Carilion School of Medicine

Background. Throughout the COVID-19 pandemic, personal protective equipment (PPE) supply chains have been inconsistent at best, and at times, highly risky and unreliable. Simultaneously, health care organizations have had to update their PPE-related processes and policies sometimes on an hourly basis, given the evolving recommendations. While practicing based on the best evidence available is critical, frequent changes combined with PPE supply chain issues can contribute to frontline caregiver exhaustion and burnout. Many frontline caregivers were (and are) eager to help their organization respond to the pandemic by doing more than providing critical patient care. In our organization, many providers become innovators, creating new ideas for PPE and developing inventions to help with many of the issues facing COVID-19 patients. Our organization and its frontline staff partnered with Virginia Tech to filter and then develop innovative PPE solutions. For example, aerosol-generating procedures were identified early in the pandemic as high-risk scenarios. Inventors developed many different ideas to address this challenge. Face shields were recommended by the Centers for Disease Control and Prevention; however, many shields were so uncomfortable that frontline caregivers removed them, despite the risk. Our multidisciplinary team heard these challenges, and rather than institute a policy of punishing frontline providers for not wearing appropriate PPE, the organization invested in developing better PPE that matched provider needs by listening to our frontline staff. We believe the work presented here represents a trend toward human-centered design in health care—not just for patients, but also for providers. Better-designed PPE may result in a higher rate of usage, adherence to guidelines, and enhanced patient/provider safety. This information adds value to the 2021 Vizient Connections Summit, as our strategy was to hear the voices of our clinical staff.

Intervention Detail. Aerosol-generating procedures have been extremely concerning throughout the response to COVID-19, given how the disease spreads. The first issue the multidisciplinary team addressed was to develop some type of respirator valve that would filter exhaled air from half-face respirators. The team went through multiple iterations, benchmarking with other innovators and health systems across the United States. The team 3D-printed multiple valves and tested them first in simulation, and then with frontline caregivers by developing prototypes. The second issue was the development of a face shield. After conducting a literature review and benchmarking with other fields and institutions, we conducted a 4-phase, mixed-method study. We administered a self-report survey study at a single health care institution asking about challenges and facilitators for face shield usage using both quantitative and qualitative analysis. The top 5 responses from the analysis were then used to develop design criteria. The prototype was designed based on these criteria and tested in a randomized trial compared with current shields. An updated prototype II was developed and iteratively tested in 2 additional study phases. In phase 1, 1648 surveys were returned. The design criteria included: (1) the ability to adjust tension; (2) load-bearing, to shift the weight of the face shield from the temples to the top of the head; (3) anti-fogging, improved ventilation; (4) increased freedom of movement while wearing the shield; and (5) improved durability. In phase 2, 26 participants (experimental group: n = 14, control group: n = 12) wore the shield for 1 week and provided daily feedback via text and qualitative feedback via focus group. In phase 3, 10 participants wore an updated version for 1 shift and provided verbal feedback. Phase 4 mirrored phase 3 methodology, with 16 participants and the updated prototype III.

Outcomes and Impact. To date, our multidisciplinary team’s work has resulted in the development of a respirator valve, which was ultimately replaced with a GE product. This work has impacted frontline caregivers directly. For the face shields, our multiphase process yielded a face shield that meets frontline caregiver needs. Based on our quantitative and qualitative findings, issues were identified that informed the development of design criteria, as articulated by a multidisciplinary team. All iterations of face shield designs are available for viewing and download.¹ The criteria were then used to update the shield, with issues reported by phase. During phase 4, fewer issues were cited, and importantly, our team found that individuals self-reported that they took the shield off less, resulting in fewer potential contaminations.

1. The Parker Lab. Face Shield Project. Accessed January 20, 2022. https://labs.vtc.vt.edu/parker/research/face-shields/.

More Bang, Less Bother: Efficiently Maximize Quality by Decreasing LOSI

Brittany Nicosia, MSN, WHNP-BC; Michael Buckingham, MHA

Wake Forest Baptist Health

Background. Length of stay index (LOSI), the actual length of stay (LOS) divided by the expected LOS, is used as a benchmark for efficiency and value-based, quality care. Hospital systems monitor LOSI to drive quality improvement initiatives that lower costs without compromising patient care. Measuring and tracking LOS can be difficult, and without insightful data the information is objectively null. The Vizient Clinical Data Base provides data for like institutions to gain insights on their metrics for success. Historically, our academic institution’s LOSI on the gynecologic oncology service line, as it compares to other like organizations in the Vizient Clinical Data Base, has been unexpectedly high. We believed that our actual LOS exceeded our expected LOS because our providers were not accurately capturing the acuity of our patients in their documentation. We recognized that accurate acuity capture would not only reduce our LOSI but would also produce more meaningful data for our comparisons in the Vizient Clinical Data Base. Thus, we set out to develop a strategy to improve our data capture to precisely measure our LOSI so that we could produce rich data that enables the accurate measure on the impact of our interventions. Our goal was to not just improve our LOSI, but to impact our ability to assess, monitor, and compare our quality improvement initiatives. Our documentation strategy was developed to maximize our opportunity for acuity capture, while minimizing disruption to provider workflow. We designed this strategy to implement sustained change—enhancing our providers’ ability to improve patient acuity capture with minimal effort, while producing more bang with less bother.

Intervention Detail. The opportunity for improvement in the inpatient gynecologic oncology service line LOSI was identified using the Vizient Clinical Data Base. A gap in our LOSI was recognized in comparison to other academic medical centers within the Vizient Clinical Data Base, so we set out to develop a strategic model to improve performance without compromising quality care. Using the 80/20 rule, we identified 20% of the diagnosis-related groups (DRGs) that make up 80% of our desired population. We adapted the algorithm provided by Vizient to apply weighted measurements to expected LOS that were DRG-specific. This process allowed us to identify the key admission diagnoses that we needed to capture. Through our electronic medical record, a smart list was embedded with the top DRGs and their most heavily weighted coefficients, along with their associated billing codes that were found to be most impactful to the gynecologic oncology service line quality outcomes. This smart list was then saved to the service line’s user smart phrases under a dot phrase. Subsequently, the dot phrase was incorporated into a standardized discharge summary that was shared with all service line providers. And finally, health care providers were educated on the importance of using the standardized discharge summary to improve quality care and the necessity to address each diagnosis selected in the smart list within the body of the discharge summary and the hospital course progress notes.

Outcomes and Impact. Through utilization of our documentation strategy, we effectively decreased the gynecologic oncology service line LOSI from an average of 1.15 in November 2018 to an average of 0.91 in December 2019. Our efforts have improved the capture of true acuity in the gynecologic oncology patient population at our academic institution through sustainable change in provider workflow. The accurate reflection of our LOSI has provided a reliable denominator for index models to monitor change while increasing the value and impact of our quality-focused projects. We now have an accurate benchmark to evaluate if our interventions focused on improving care are truly impactful. The meaningful data produced by this documentation initiative has contributed to databases like the Vizient Clinical Data Base to drive comparative data analysis. We improved provider satisfaction through the collaborative design of an efficient tool that streamlined documentation, providing an efficient mechanism for capturing patient acuity. The ease and efficiency of our documentation initiative has been the largest contributory factor to our success in sustaining this change. By optimizing capture of all relevant diagnoses we increased our likelihood of billing at a higher DRG, improving quality metrics and increasing our revenue in both fee-for-service and quality-based reimbursement models. In summary, our documentation strategy has successfully decreased our LOSI, provided meaningful data to accurately measure quality initiatives, increased the value of comparative insights offered by operational databases, improved provider satisfaction, and increased revenue through reimbursement.

Assessing the SEP-1 Impact on Antibiotic Use and Patient Outcomes

Joy Peterson, PharmD, BCPS, BCIDP; Jessica Pearson, DNP, AGACNP-BC; Tanea Womack, PharmD, MBA, BCIDP

Wellstar Health System

Background. The SEP-1 core measure represents the Centers for Medicare & Medicaid Services’ attempt to address sepsis management through policy. Based on recommendations from the Surviving Sepsis Campaign, the measure includes administration of intravenous, broad-spectrum antibiotics within the first 3 hours of severe sepsis presentation. To comply with the measure, a multidisciplinary, systemwide group of leaders and frontline staff at Wellstar Health System worked together to develop and implement needed changes. Actions taken included implementing an education campaign, modifying the current sepsis order set to ensure antibiotics met the measure and local antibiogram, reviewing the antibiotics stocked in automated dispensing cabinets, and developing a table for frontline staff that lists appropriate order and intravenous compatibility of antibiotics. As the changes were made, data were regularly reviewed to assess compliance with the measure and the impact on mortality rates. Changes in the volume of antibiotic use was not routinely part of this data monitoring, as it was measured by a different committee. The SEP-1 core measure, with the rapid decision-making required, necessary use of broad-spectrum agents, and significant consequences of not meeting the measure, has a propensity to drive unnecessary broad-spectrum antibiotic use through overdiagnosis of sepsis and overuse in septic patients if proper de-escalation practices are not utilized.¹ The Infectious Diseases Society of America has voiced concerns to this effect and the appropriateness of the measure is an area of debate.² Since implementing the previously listed changes, an increase in antibiotic use has been seen throughout the health system. This unique approach represents a merging of Sepsis Committee work and antimicrobial stewardship programs (ASPs), rather than solely focusing on volume of antibiotic use. As a result of this more comprehensive view, it was uncovered that in this instance, increased broad-spectrum antibiotic use led to improved patient outcomes.

Intervention Detail. The primary goal of this project was to quantify the impact of the SEP-1 core measure on broad-spectrum antibiotic use within the health system when comparing pre-SEP-1 practice to post-SEP-1 practice. Secondary endpoints were designed to evaluate the impact of this change in antibiotic use on patient outcomes, and included assessment of the appropriateness of initial antibiotic therapy and comparisons of mortality, intensive care unit length of stay (LOS), hospital LOS, and readmissions between the 2 groups. As one of the underlying concerns driving this project was the possibility of overdiagnosis of sepsis, the target patient population included all patients with suspected sepsis—regardless of whether this was later found to be accurate. Using diagnosis codes, laboratory data, or medication use data would bias the population toward those who were more likely to have a correct diagnosis of sepsis. It was determined that the best way to identify this population was through use of the sepsis order set, which was reliably utilized at the first suspicion of sepsis. Patients with orders from the sepsis order set from January 2015 to June 2015 were compared to those from January 2018 to June 2018. Patients 18 and older with suspected sepsis were included. Those with life expectancy of less than 48 hours were excluded.

Outcomes and Impact. The number of patients prescribed antibiotics for sepsis more than doubled between 2015 and 2018, even when controlled for changes in hospital admission volume. This increase was also reflected in the number of antibiotics prescribed. Not surprisingly, ASP activity increased in a similar fashion to almost double the number of interventions between the 2 time periods. However, the incidence of correct identification of sepsis increased and, most importantly, mortality decreased by over one-third, a reduction that was also demonstrated in the metrics measured by the Sepsis Committee. Median antibiotic days of therapy, hospital LOS, and intensive care unit LOS also decreased in 2018 compared to 2015. In summary, antibiotic use and the demand on ASP resources have increased since implementation of the SEP-1 core measure, but may be contributing to a decrease in mortality, hospital LOS, and intensive care unit LOS. The relatively high accuracy of sepsis diagnosis (78%) is above the historical cohort and indicates that, while there is likely still room for some improvement in de-escalation and discontinuation, the antibiotics are not being used inappropriately in a widespread manner. This information provides valuable insight to help pharmacy and hospital leadership understand both the increased utilization and effect on the health system as a whole.

1. Centers for Medicare & Medicaid Services. QualityNet: Inpatient Hospitals Specifications Manual, Version 5.2. 2016. Accessed October 12, 2018. https://www.qualitynet.org.

2. IDSA Sepsis Task Force. Infectious Diseases Society of America (IDSA) POSITION STATEMENT: why IDSA did not endorse the surviving sepsis campaign guidelines. Clin Infect Dis. 2018;66:1631.

Design of an Inpatient Predictive Model to Optimize Patient Experience

Starr Watson, MSN, RN, CNL; Michaela Harner, MSN, RN, CNL; Nadia Prince-Williams, MSN, RN, CNL

Wellstar Paulding Hospital

Background. A major goal of our organization is to provide a positive journey for all patients who are admitted to our acute care units. A positive patient experience is associated with higher quality and safety outcomes, and can also help drive overall system transformation. Previous work showed great improvements in leader rounding structure and use of a positive coaching model that increased our patient experience metrics. An area of opportunity was to optimize our current rounding tool by adding a question that allowed the nursing leader to predict (on a zero to 10 scale) how the patient would likely recommend the hospital on the patient experience survey. This strategy was tested and measured on actual survey results versus nursing leader predictions. Nursing leaders received a weekly predictive accuracy report to gain insight on opportunities missed. They then called patients to discuss opportunities, trending missed interventions using a Pareto chart. The nursing leaders identified countermeasures to improve their prediction accuracy, which resulted in immediate and improved patient experience metrics on all units. With our continuous cycle of learning, nursing leaders were able to make predictions on trends that impacted the risk for a negative patient experience. They utilized A3 thinking with support from the executive team to aid process change on the unit level, but also needed facility improvements that mitigated negative patient experiences. Examples of facility improvements include an improved food delivery process to patient rooms, a new parking lot to address demand, and A3 problem-solving visual management implementation in all support service areas. Nursing leader predictions were added to our daily voice of the customer huddle, where learning could be shared hospitalwide. Executives and senior leaders also became involved in developing immediate action plans for patients who were not satisfied with their experience. This abstract on how to implement a successful patient-centric prediction model is important to include in the 2021 Vizient Connections Summit because key learnings can be facilitated from all levels in the organization to support improvement of the patient experience.

Intervention Detail. The design of our inpatient predictive model comprised several key processes and steps. Our first task was to add a question to our current leader rounding tool that allows our leaders to predict on a zero to 10 scale how a patient would likely recommend the hospital on the patient experience survey. After this question was added, we would be able to collect quantitative data on prediction accuracy for each unit and leader. The unit nursing managers could not accurately predict if a patient was dissatisfied or satisfied with the hospital care they received. If the unit manager cannot recognize when a patient is unhappy, then we were not able to perform appropriate service recovery interventions. We added this learning to our inpatient A3 and worked through process improvement and learnings with leaders and bedside team. In evaluation of the negative patient experiences, several opportunities were identified, such as improved rounding content, service recovery approaches, and shared learnings. With our learnings, improvements in prediction accuracy increased up to 30% on all units. The impact of this prediction led to visual management lanes being added on all inpatient units. These are visual problem-solving tools that provide at-a-glance information about current process performance to help unit team members guide their daily work and monitor ongoing improvements. These lanes allowed open discussions on the patient experience with the bedside team and encouraged unit-level problem-solving. Unit leaders discuss the patients who are dissatisfied with their care and predicted to rate us at 8 or below in their daily team huddle. The team talks about each patient’s concerns and develops immediate, individualized action plans to improve the patient’s experience. These stories of service recovery and negative patient experiences were added to the voice of the customer daily huddles, where they could be shared with all leaders across the organization. Negative patient survey responses were evaluated and trended on the unit Pareto chart to gain better insight on unit trends. The National Research Corporation (NRC) Health On-Demand Research Study Consumer Sentiment Data survey question on how likely a patient would be to recommend the hospital was reviewed to evaluate performance gaps on all units. After 6 months, all inpatient units showed an average 15% improvement in the “likely to recommend” patient survey question.

Outcomes and Impact. The predictive model was built to accurately predict the patient response on a zero to 10 scale for “likely to recommend the hospital” in leader rounding and on the patient experience survey. The leader rounding prediction was extracted from our rounding tool and then compared to the patient survey after discharge. The prediction accuracy improvement for 3 trialed units from July 2019 to September 2019 includes: 77% to 90.9% (+14%), 60% to 91.7% (+32%), and 17% to 50% (+33%). Through our cycle of learning, we determined the importance of also using our Lean management system on the unit level to gain insight from the bedside team on current patient experiences, as well as discussing as an organization shared learnings and service recovery efforts. Our pre-intervention percentile rank on all inpatient units from our patient experience NRC Health survey was in the 60% range. The patient experience scores increased 15% on all units by December 2019. With visual management problem-solving, we were able to collect qualitative data via Pareto charts on trends identified on the unit. Multiple interventions improved our patient experiences, including problem-solving discussions at the bedside and the support of our executive team in implementing these changes. We continue using the inpatient prediction model to improve our patient experience measures. These continued discussions in visual management led to a patient experience board being implemented on each unit. This model has also improved our overall inpatient percentile rank in the NRC Health survey from the 49th percentile in 2019 to the 67th percentile in 2020.

Thinking Outside the Mask (N95 Alternative)

Ronald E. Sherman, BS, Financial Management & Executive MBA; Kirt Tassmer, BS, Industrial Engineering; Tony DePaola, CMRP; Rockman Ferrigno, MD; Margaret Rose, MD

Yale New Haven Health System

Background. In accordance with Centers for Disease Control and Prevention guidance, health care workers are required to have appropriate size, fit-tested respirators to ensure proper protection when providing selected, high-risk, aerosolized procedures. Our work to source high-quality N95 respirators revealed a supply chain disruption signifying that the supply of appropriately sized N95 respirators would not be enough to sustain our immediate health system needs during the peak of the early COVID-19 pandemic. Knowing a true crisis loomed for our frontline health care workers and their patients, our supply chain employees drew together as a team, bringing their diverse backgrounds and varied experiences to the table for brainstorming sessions. In collaboration with medical leadership from occupational health and infection prevention, the Yale New Haven Health System supply chain turned to non-health care industrial respirators to identify a novel option for respiratory protection from COVID-19 transmission and infection. Industrial settings use elastomeric respirators for protection against asbestos and other contaminants, which created alternative options for consideration. We evaluated multiple manufacturer samples of industrial elastomeric respirators for availability, fit, and function. Centers for Disease Control and Prevention approval of elastomeric respirators for aerosol respiratory protection from COVID-19 infection allowed us to formally incorporate elastomeric respirators into our health system respiratory protection plan.

Intervention Detail. In February 2020, elastomeric respirators became part of Yale New Haven Health System’s COVID-19 respiratory protection plan. Our Corporate Supply Chain Team identified an immediately available inventory of elastomeric respirators to source from multiple warehouses and locations throughout the world. Working with the elastomeric respirator manufacturers, infection prevention, central sterile supply (CSS), and occupational health, we established distribution plans and innovative, high-temperature, disinfection protocols. Once we had our approved respirators, we planned our initial trial for deployment to medical intensive care unit staff while also including other high-risk areas such as anesthesiology and emergency medicine. The respirators would be worn by staff during their scheduled shift, returned to CSS for disinfection, and then redistributed for reuse. We quickly scaled our distribution to reach 3500 frontline health care workers. In December 2020, we shifted course to distribute personal elastomeric respirators to our frontline health care workers. This required research and the creation of a new self-cleaning protocol and education strategy, and distribution of the protocol and strategy to our staff members. Our distribution plan included a carrying bag for the elastomeric respirator, replacement filters, and a personalized note thanking our health care heroes for their service. Staff morale was greatly boosted with their having immediate access to high-level respiratory protection.

Outcomes and Impact. Incorporating elastomeric respirators strengthened our respiratory protection plan and allowed our health care workers to be protected throughout the pandemic. By utilizing nondisposable respirators, we were able to reduce our N95 burn rate—conserving them for health care workers who could not be fitted into elastomeric respirators or other respiratory personal protective equipment. After testing 14 000 asymptomatic employees who did wear respiratory personal protective equipment, we had a positivity rate of 0.22%. Yale New Haven Health’s low rate of positive COVID-19 cases among employees is attributed to our ability to develop a respiratory plan that included a combination of N95 respirators, powered air-purifying respirators/controlled air-purifying respirators, and importantly, elastomeric respirators. As we improved our elastomeric processes and moved to personalized elastomeric respirators and self-cleaning protocols, we noticed an increase in employee satisfaction. Employees felt safe and confident that they had the necessary protection. Additionally, by reducing the labor burden on CSS of disinfecting elastomeric masks, we were able to sustain a high level of operating room productivity, enabling us to address the backlog of patients requiring surgery after months of restrictions on nonurgent surgical procedures. This experience created a direct line of communication between our clinical and supply chain teams during a crisis, resulting in a meaningful, cooperative dialog with our frontline health care workers.

Poster Presentations

An Initiative to Reduce High-Dose Opioid Prescriptions

Scott Weiner, MD, MPH; James Bryant, JD; Charles Morris, MD, MPH; Sunil Eappen, MD, MBA; Peter Mullins, MD; Christopher Gilligan, MD, MBA

Brigham and Women’s Hospital

Background. In 2016, the Centers for Disease Control and Prevention issued practice guidelines recommending avoidance of opioid prescriptions for ≥ 90 morphine milligram equivalents (MME) per day for chronic, noncancer pain without careful consideration and justification. While once thought to be the standard of care, multiple studies have demonstrated that doses at these high levels are associated with only minimal improvement in function and markedly increase the risk of adverse opioid-related events, including overdose and death. Health care systems struggle with what to do with patients who are on these high doses. Simply “cutting them off” or abruptly tapering opioid doses are not in the patient’s best interests. Likewise, counting on physicians to perform opioid tapers for their patients without a supportive system in place is also a recipe for failure. At our institution, between July 2019 and December 2020, there were 137 350 pill form, opioid analgesic prescriptions written, of which 15 864 (11.6%) were for > 90 MME/d. Of these, the most frequent prescribing clinics were the hospital’s pain clinic and an affiliated primary care practice with a large number of chronic pain patients. These 2 clinics wrote 5993, or 37.8%, of the high-dose prescriptions. A top-down decision was made by the hospital’s Board of Directors Compliance Committee to safely reduce the number of high-dose opioid prescriptions. As a result, we created this program to support our prescribers and patients in these clinics.

Intervention Detail. Beginning in July 2019, we instituted a multifaceted approach toward reducing prescriptions for high-dose opioids. Steps taken include: (1) prohibiting trainees/fellows from writing prescriptions for > 90 MME/d; (2) promulgating guidelines, including tapering instructions and best practices, such as a prescription drug monitoring program check prior to every opioid prescription, a pain treatment agreement for all patients on chronic opioids, a random toxicology screening test at least once a year, and a clinic visit at least once every 4 months; (3) random chart reviews to ensure compliance with guidelines; (4) sharing providers’ prescribing practices compared to their peers; (5) pharmacist assistance to help safely taper patients; and (6) psychosocial support for any patients having difficulty with tapering, including rapid access to a substance use disorder bridge clinic.

Outcomes and Impact. Between July 2019 and December 2020, prescriptions for high-dose opioids decreased from 452 per month to 270 per month, a decrease of 40.4% (–11.8/mo, linear fit P < 0.001). Prescriptions from the 2 most frequent prescribing clinics for doses ≤ 90 MME/d remained stable, with 1017 in July 2019 and 1015 in December 2020. The data were useful to monitor our progress. Most tapers should take place as a reduction of 10% per month, so we expected the observed gradual decrease over several months. Helpful features of this intervention included: (1) a mandate from the system, so that patients knew they were not being singled out but that the intervention applied to all high-dose opioid patients; (2) sharing prescribing information of individual providers compared to their peers; and (3) involving a multidisciplinary team to support both the provider and the patient, including rapid access addiction psychiatry consultation, access to a pain psychologist, and a pharmacist who is an expert in managing opioid prescribing and tapering. This intervention is ongoing and the work is now expanding beyond these 2 clinics.

Brace Yourself for the First Newborn Fall Risk Assessment Scale

Hannah Antista, BSN, RNC-NIC; Krystal Savage, BSN, RN; Kenneth Oja, PhD, RN; Taylor Diamond, BSN, RNC-NIC

Denver Health & Hospital Authority

Background. Inpatient newborn falls, or drops, can lead to severe physical injury or death. They can also cause emotional distress to the family, caregiver(s), and medical staff.¹ The terms “newborn fall” and “newborn drop” are used synonymously and are defined as an unplanned descent of the newborn to the ground with or without injury. A 2018 Joint Commission Quick Safety advisory regarding newborn fall prevention stressed the importance of standardizing assessments and procedures by creating a shared mental model to aid staff and improve outcomes.² Thus began the process of developing the DH-Newborn KAnt Fall Scale—a numerical scale that assesses both newborn and caregiver risk factors to determine if the newborn is of low, moderate, or high fall risk. The scale then provides intervention suggestions, including how many health care providers should assist in transfers. The Denver Health NICU’s Evidence-Based Practice Committee, which includes 4 neonatal intensive care unit (NICU) nurses, formed a task force for this project. This task force includes the NICU educator, a Denver Health research scientist, and other members of the interdisciplinary team. After receiving ethical approval from the institutional review board, our task force followed the Jones and Bartlett model for instrument development to establish the initial psychometric properties of the scale, including: (1) content validity (the relevance of each item); (2) interrater reliability (the reproducibility of results); and (3) construct validity (the instrument measures what it is intended to measure). Being part of the 2021 Vizient Connections Summit gives us the opportunity to increase awareness of this safety issue, while also providing an assessment and management tool to serve as the groundwork for future research and use. The hope is that other institutions will adopt this tool and conduct further testing to establish additional psychometric properties, with the goal being to provide safer care to the inpatient newborn population.

Intervention Detail. In order to identify a need for a newborn fall risk assessment tool and assess the staff’s baseline knowledge pertaining to newborn falls, our task force released an 8-question, pre-implementation survey to the Denver Health NICU staff. Based on the results, we determined that the unit needed a tool with the ability to assess fall risk factors and provide intervention recommendations. As we reviewed the literature, we identified one of the only studies to date that addresses inpatient newborn falls. Linda Helsey’s 2-year study in 2006 provided helpful insight as to why newborn falls occur.¹ With the help of the literature and survey results, we developed the first draft of the DH-Newborn KAnt Fall Scale. We tested the drafted instrument on more than 50 patients in a 1-month period to help us create an accurate numerical scale to accommodate a variety of newborns and caregivers. The final phase of developing an instrument is to establish its reliability and validity. We formed a group of 10 NICU nurses, known as content experts, who volunteered to assist with the data collection. Item-level content validity indices of 0.78 or higher and an average scale content validity index of 0.90 or higher were used to establish content validity. This process determined that 2 items were irrelevant, so they were removed from the scale. To assess interrater reliability, the content experts paired up and independently completed the newly revised tool on the same patient at the same time, remaining blind to each other’s assessment. The forms were stapled together and placed in a designated folder that was collected by the primary investigator. We followed the guideline of a 5:1 respondent-to-item ratio and collected a total of 120 pairs of completed assessments. Our primary investigator used SPSS Statistics software to calculate an intraclass correlation coefficient with a 95% confidence interval based on a 2-way random-effects model.

Outcomes and Impact. After implementing the fall risk assessment tool in Epic and providing staff education, our task force sent out the same 8-question survey to NICU staff. The results proved that the staff views the scale as being beneficial. For example, of the 23 nurses who completed the post-survey, 87.5% felt that the Denver Health NICU has a standard way of assessing newborn fall risk—a large improvement from the mere 18% in the pre-survey. The post-survey also showed that 100% of staff members agreed that they currently perform fall prevention interventions, whereas only 84% felt that way prior to implementing the DH-Newborn KAnt Fall Scale. The NICU nursing staff also expressed that the tool is easy to use. For example, all the risk factors are items that the infant’s nurse would know after an initial assessment. The infant risk factors are all objective items, reducing the risk of human error and variations in nurse interpretation or opinion. The caregiver risk factors contain both objective and subjective items, but these are things that can easily be assessed through a conversation with the caregiver. The nursing staff also agreed that completing the scale is not exceedingly time-consuming and can be done in a few minutes. Many of the interventions suggested by the DH-Newborn KAnt Fall Scale were being performed prior to its implementation, but were not being consistently documented or communicated. This tool helps create standardization in newborn fall prevention. A major limitation of the current study is that the tool was implemented prior to validity and reliability testing. Thus, the content experts may have been biased to the relevance of the items. As such, further testing of the DH-Newborn KAnt Fall Scale in other acute care hospitals is needed. Additional psychometric properties, including sensitivity, specificity, and predictive validity, should be established.

1. Helsley L, McDonald JV, Stewart VT. Addressing in-hospital “falls” of newborn infants. Jt Comm J Qual Patient Saf. 2010;36:327–333.

2. The Joint Commission. Preventing Newborn Falls and Drops. Quick Safety Issue 40. 2018. Accessed March 2018. https://www.jointcommission.org/-/media/tjc/newsletters/quick_safety_issue_40_2018_newborn_falls_dropspdf.pdf?db=web&hash =A91597BE199080F84BD4EA5261F3B48B.

Implementing a Maternal Rest Protocol to Prevent Newborn Falls

Kenisha J. Karlsson, BA in Kinesiology, BSN, MNN-C; David Mulkey, DNP, RN, CPHQ, CCRN, CHSE

Denver Health & Hospital Authority

Background. Adult inpatient fall prevention programs have been thoroughly studied. In contrast, there is little known about newborn falls. The rate of inpatient newborn falls has been reported as high as 4.6 per 10 000 live births, with an estimated 1600 falls annually in the United States.¹ Newborn falls often do not result in significant injuries, but major injuries such as intracranial hemorrhage can occur. According to the Association of Women’s Health, Obstetric and Neonatal Nurses, all newborns should be considered a high fall risk.² Just a couple of years prior to the Association of Women’s Health, Obstetric and Neonatal Nurses Practice Brief release, The Joint Commission issued an advisory that synonymized the terms newborn falls and newborn drops; therefore, a newborn fall and a newborn drop should have the same prevention process to reduce practice variation.³ A newborn fall is defined as a fall in which a baby being held or carried by a health care professional, parent, family member, or visitor falls or slips from that person’s hands, arms, or lap, regardless of the surface on which the child lands and regardless of whether the fall resulted in injury. Between January 2015 and September 2018, our mother-baby unit reported 9 newborn falls—1 of which resulted in a cranial fracture and brain hemorrhage. Historically, fall prevention at our institution focused primarily on adult risk factors, while those related to newborn falls had not been explored. A root cause analysis was completed and indicated that all 9 of the newborn falls at this institution occurred when a caregiver fell asleep while holding the newborn. Therefore, it was not known to what degree implementation of a maternal rest protocol would impact newborn falls. Including this information in the 2021 Vizient Connections Summit provides an opportunity for organizations to learn about an innovative strategy for improving newborn falls.

Intervention Detail. Interventions found in the literature included using a safe sleep pledge and a quiet time to promote maternal rest. Drawing on the successful experiences reported in the literature, a multidisciplinary team planned to pilot a maternal rest protocol using the model for improvement Plan-Do-Study-Act cycle proposed by the Institute for Healthcare Improvement. During the “plan” phase, the team created a process map to capture the full process and daily workflow. In addition, awareness of the current state of newborn falls was raised and its potential impact on patient care emphasized. The team sought advice from nursing personnel on the mother-baby unit through a survey regarding the most efficient way to implement quiet time hours. Survey response from the staff (n = 45) determined that quiet time should be implemented twice daily between the hours of 0200 and 0400 and 1400 and 1600. During this time, mothers roomed-in with their newborns. Only 1 support person was allowed in the room during quiet time hours. No staff—including medical residents, attending physicians, nurses, nursing assistants, or photography, birth certificate or lab personnel—could enter the room unless the patient requested, or there was an emergency. In the “do” phase, quiet time signs, in both English and Spanish, were placed around the unit, and all new mothers were educated about the quiet time initiative. A daily overhead announcement was provided at 1400. An 0200 overhead announcement was forgone to promote nighttime rest. All mothers received and signed a safe sleep pledge that established their commitment to the initiative. By signing the safe sleep pledge, caregivers agreed for each newborn to sleep alone, without blankets, pillows, or toys, on their back, in a safe crib, in a smoke-free environment.

Outcomes and Impact. During the “study” phase, the pre-intervention newborn fall rate was 3.18 per 10 000 live births. No newborn falls occurred during the 90-day pilot, and the team decided to “act” by adopting the maternal rest protocol as an official component of its mother-baby standard work. Post-intervention newborn fall rates per 10 000 live births have sustained at zero per 10 000 live births for 4 consecutive years and counting post-implementation. As of October 2021, there have been no newborn falls on the mother-baby unit. There was a 62.5% response rate on the staff feedback survey (n = 45). Overall, 95% of staff indicated that the maternal rest protocol should continue. The major barrier identified by staff was performing daily tasks, such as medication administration, laboratory blood draws, and vital signs. Caregivers reported that they were satisfied (95%) with the maternal rest protocol. Patients reported that quiet time hours allowed them to rest and promoted bonding with their newborn. Mothers who were breastfeeding reported that they were more motivated to continue breastfeeding after adequate rest during quiet time. The team continues to evaluate the effectiveness of the maternal rest protocol. Implementing a maternal rest protocol in the mother-baby unit at our institution eliminated newborn falls for 4 consecutive years to date. This data suggests that promoting maternal rest by using an official, unit-enforced quiet time and newborn safe sleep pledge was successful.

1. Helsley L, McDonald JV, Stewart VT. Addressing in-hospital “falls” of newborn infants. Jt Comm J Qual Patient Saf. 2010;36:327–333.

2. Association of Women’s Health, Obstetric and Neonatal Nurses. Prevention of newborn falls/drops in the hospital: AWHONN practice brief number 9. Nurs Womens Health. 2020;24:383–385.

3. The Joint Commission. Quick Safety Issue 40: Preventing Newborn Falls and Drops. 2018. Accessed January 2, 2020. https://www.jointcommission.org/resources/news-and-multimedia/newsletters/newsletters/quick-safety/quick-safety-40-preventing-newborn-falls-and-drops/.

ICU Delirium Care Redesign: A Multidisciplinary Team Approach to Harm Prevention

Adam M. Glenn, BSIE, LSSBB; Elizabeth Anderson, RN, BSEE

Duke University Health System

Background. There is variability in how delirium is identified, prevented, and treated within the intensive care units (ICUs) at Duke University Health System. Delirium is often associated with serious complications, including increased length of stay, readmission rates, and fall risk, as well as higher mortality rates. Delirium is common, especially in older patients, and it can affect 20% to 50% of critically ill patients.¹ In affected persons, delirium is underrecognized by as many as one-third of physicians and nurses. It is estimated that 30% to 40% of delirium is preventable.² The combination of delirium being common, underrecognized, and preventable makes it a prime opportunity for developing a coordinated, standardized approach for detection and intervention. Across Duke University Health System’s 7 adult ICUs, delirium screening rates vary from 20% to 90% of patients screened per shift, using the Confusion Assessment Method (CAM) ICU assessment. There is also inconsistent use and understanding about which patients meet the threshold to be considered unable to assess. Nursing staff members select between 15% and 48% of CAM ICU assessments as “unable to assess” across the units. This leads to patients being incorrectly or inconsistently identified as potentially delirious, in turn reducing the number of appropriate delirium interventions that patients receive.

Intervention Detail. The multidisciplinary Care Redesign Team focused on creating an electronic health record-driven, evidence-based approach to delirium detection and intervention across health system adult ICUs. The team began by conducting stakeholder interviews with all 7 adult ICUs, as well as distributing an online survey to determine additional opportunities. Stakeholder interviews included nurse managers, clinical nurse specialists, clinical leads, and frontline nursing. The interviews and online survey found there was a high level of variability and confusion around appropriate usage of the current delirium assessment tool, the CAM ICU assessment. Based on this frontline staff feedback, the team conducted a “go and see” approach to observe ICU workflow. The team came up with the solution to build each feature of the CAM ICU assessment into the electronic health record (Epic) through nursing flowsheet documentation—including a simple, detailed explanation of how to perform each step of the assessment and corresponding options for nursing staff to select for each step. Based on these selections, assessment results are automatically calculated and populated within the flowsheet. The team also performed an evidence review for valid reasons not to complete the assessment. These “unable to assess” reasons were built into the electronic health record as the first step of the assessment. If nursing determines a patient is unable to assess, they are required to select a valid reason for not assessing (Richmond Agitation-Sedation Scale –4, Richmond Agitation-Sedation Scale –5, language barrier, or advanced dementia). The team also reviewed the available nursing interventions listed within the “delirium actions taken” documentation and added an intervention for modifying the patient environment to support delirium mitigation. To support ongoing monitoring of ICU delirium assessments, a dashboard report for managers was created, as well as a nursing quality dashboard that can be used during each shift to monitor shift progress toward delirium goals.

Outcomes and Impact. Since implementing these interventions in fall 2020, delirium assessment rates for all ICU patients are greater than 90% per shift. This is a 26% increase in completion rate for CAM ICU delirium assessment compared to prior care redesign efforts. The project resulted in a 12.4% reduction in usage of the “unable to assess” response to the CAM ICU assessment, indicating that more patients are being appropriately screened. The impact of the screening enhancements and more accurately identifying potentially delirious patients has reduced average length of stay for delirious patients in all Duke University Health System ICUs by 20.7%, resulting in a $2.2 million estimated annual benefit from increased patient throughput.

1. Smith CD, Grami P. Feasibility and effectiveness of a delirium prevention bundle in critically ill patients. Am J Crit Care. 2016;26:19–27.

2. Inouye SK, Westendorp RG, Saczynski JS. Delirium in elderly people. Lancet. 2014;383:911–922.

Data-Driven Strategies for Clostridioides Difficile Reduction

Effie Rubia, MSN-PH, RN, CIC; Janice Elizabeth Sandiford, MSN, RN, NP-C, OCN; Mark A. Stern, MD; Adam Bressler, MD; Cornelius E. Brown, PharmD BCPS BCCCP; Adriana Karettis, Microbiologist; Egharevba Osarenomase, MD; Lesa Kelly, EVS

Emory Healthcare

Background. Clostridioides difficile is estimated to cause almost half a million illnesses in the United States each year and an estimated 29 300 deaths.^1,2 Individuals are 7 to 10 times more likely to get C. difficile while taking an antibiotic and during the month after.^2,3 Prolonged health care admissions also increase C. difficile risk. Early identification, effective cleaning practices, and prudent prescribing of antibiotics and proton pump inhibitors are foundational to C. difficile prevention. Our 3 hospitals are part of a large, regional health system. System hospitals are referred to as operating units, and for the purposes of this abstract, the term DeKalb Operating Unit (DOU) refers to our 3 hospitals. In 2019, the health system established a goal of achieving top national quartile performance for hospital-acquired infections by 2024. Incremental yearly goals were set for each hospital to achieve this. While being part of a large system provides a breadth of resources and knowledge, the DOU lacked the local coordination of efforts and participation of frontline staff essential for making meaningful improvement in our hospital-acquired C. difficile rates. The DOU office of quality had experienced successful, sustained improvements in other important outcomes by establishing powerful structures for improvement called tactical teams. This work was based on the Donabedian conceptual model that provides a framework for developing structures supporting processes that drive positive outcomes. Teams are facilitated in a standardized manner, focusing on a metric, goal, and plan. Ideally members are frontline staff who actually perform the work of the processes that are being improved. The teams serve as a coordination hub for improvement efforts, minimizing duplication and waste. Our performance at the beginning of this effort was worse than the national average. Quality leadership was engaged to support the development and success of a C. difficile tactical team.

Intervention Detail. The C. difficile tactical team included representatives from nursing, lab, pharmacy, environmental services, hospitalist medicine, and gastroenterology physicians. Additional individuals were invited to participate as needed or desired by the committee. DOU administrative team members served as ex officio members and were welcome at all meetings. Team members were oriented to the tactical team structure and their role as subject matter experts, including ensuring appropriate outcomes and processes measurements and engaging and educating staff and leadership. With initial support of the quality director, a frontline infection preventionist facilitated the team and served as an subject matter expert for best practices associated with C. difficile reduction. A gap analysis and case reviews revealed that we were not consistently identifying C. difficile that was present on admission. This not only limited our effectiveness at containing the spread of this infection but also resulted in overstating our true hospital-acquired infection rate. Additionally, our lab was following a different diagnostic algorithm than the rest of the health care system, resulting in a higher reported infection rate.

The following priority focus areas were identified:

To assure timely identification of present on admission C. difficile, the following initiatives were prioritized in a sequential manner:

• 1. Create a C. difficile algorithm, enabling nurses to order tests, collect stool samples, and implement transmission-based precautions.
• 2. Amend the patient admission to include symptoms of C. difficile. for early identification.
• 3. Engage patient care techs in monitoring for diarrhea.

Other improvements included:

• Developed standardized treatment protocols or stewardship of antibiotics and proton pump inhibitors.
• Improved cleaning of room surfaces using Environmental Protection Agency-approved, spore-killing disinfectant and an ultraviolet light device.
• Standardized lab protocols for C. difficile testing.

Outcomes and Impact. Despite FY2020 having been an unprecedented year due to the COVID-19 pandemic, the 3 facilities demonstrated a commitment to meet the goals and were able to dramatically decrease C. difficile rates by 40%, 70%, and 89%, respectively, with standardized infection ratios of 0.58, zero, and 0.39. This significantly exceeds the performance expectations set for the year. The tactical team now focuses on monitoring process metrics for sustainment and continued case review of any new occurrences. We believe that the tactical team structure enabled the organization to maximize our process improvement efforts. In complex and challenging times, maintaining focus on outcomes and the processes driving them is critical to quality improvement. Developing facilitation skills in frontline staff members can increase their confidence and effectiveness. The facilitation process breaks down hierarchical barriers and allows each member an equal voice and shared responsibility for success. As the team develops, each member recognizes that they have the power to improve their work and have an impact on important outcomes.

1. Centers for Disease Control and Prevention. What is C. diff? 2021. Accessed August 24, 2021. https://www.cdc.gov/cdiff/what-is.html.

2. Centers for Disease Control and Prevention. CLOSTRIDIOIDES DIFFICILE. Accessed August 24, 2021. https://www.cdc.gov/cdiff/pdf/Cdiff-Factsheet-P.pdf.

3. Centers for Disease Control and Prevention. Your Risk of C. diff. 2021. Accessed August 24, 2021. https://www.cdc.gov/cdiff/risk.html.

“A” Stands for Airway: Reducing Airway Emergencies Following Tracheostomy Placement

Kellianne Fleming, MSc, RRT-ACCS, RRT-NPS, Froedtert & Medical College of Wisconsin; Kathryn Lauer, MD, Froedtert & Medical College of Wisconsin; Thomas Carver, MD, Froedtert & Medical College of Wisconsin; Jennifer M. Vehring, RRT, Froedtert Hospital; Sarah Bazelak, BSRT, RRT, RRT-ACCS, AE-C, Froedtert Health; Sheila C. Blogg, MSN, BSN, BA, Medical College of Wisconsin; Staria Brickner, MSN, RN, CCRN, Froedtert Hospital; Inderjit Pooni, ACNS-BC, Froedtert Hospital; Timothy E. Klatt, MD, Froedtert & Medical College of Wisconsin

Froedtert & Medical College of Wisconsin

Background. New surgical airways are at risk for dislodgement and inadvertent decannulation, which can have life-threatening consequences. Through Vizient safety intelligence, our organization reported 3 serious adverse events in a span of 3 months as a result of tracheostomy dislodgement within 7 days of placement. The respiratory safety domain team created a multidisciplinary group to perform a root cause analysis that identified the following: (1) all of the events occurred with repositioning immediately prior to the onset of patient decompensation; (2) bedside clinicians recognized a change in patient clinical status; however, there was a delay in recognizing and confirming the tracheostomy was dislodged; (3) bedside clinicians needed more education regarding signs of tracheostomy dislodgement; (4) lack of an overall standardized response to these emergencies; (5) a gap in knowledge regarding the utility of end-tidal carbon dioxide capnography in patients with a tracheostomy; (6) the emergency response team needed to be alerted at the first signs and symptoms of respiratory distress related to a compromised airway; (7) we needed a method to identify patients with a new surgical airway in the electronic medical record and at the bedside; and (8) the emergency response team needed the addition of surgeons familiar with surgical airways.

Intervention Detail. Several processes were implemented following the root cause analysis. We assigned a third person to be at the bedside who is responsible for ensuring the airway remains secure during turning and repositioning following placement of a new tracheostomy. Second, we created a surgical airway emergency response team consisting of an anesthesia provider, intensive care unit pharmacist, a rapid response registered nurse (RN), a rapid response respiratory therapist, a transporter, a critical care medicine resident, an intensive care unit charge RN, and a trauma surgery attending. Third, a surgical airway emergency response algorithm (ventilated and nonventilated) was developed for the emergency management of patients with tracheostomies less than 7 days old. The algorithm is posted at the patient’s bedside and provides pertinent information regarding the date and type of tracheostomy. A “high-risk airway” banner was created for the electronic medical record that identifies patients with new surgical airways. Fourth, the respiratory care department pulls airway reports daily and reviews the location/patient room numbers of new surgical airways in twice-daily shift huddles. The respiratory therapists are responsible for huddling with bedside RNs at the start of each shift to review the algorithm in case of an emergency. Finally, the respiratory educator, in collaboration with the simulation center team, built education modules regarding tracheostomy emergency management, with 4 specific arms: respiratory therapist, acute care RN, critical care RN, and provider. Completion of online modules is required. Respiratory therapists are required to complete both an online module and in-person simulation training. This education has now become part of the RN Residency Program and annual competency training. Education touched over 1000 staff and providers.

Outcomes and Impact. To date, there have been no new sentinel events related to the loss of a new surgical airway. There have been 2 events related to defective tracheostomy tubes that were identified by the respiratory therapist (as third person) prior to turning patients, likely averting an serious adverse event. The algorithm has been activated once since roll-out and proved successful at guiding the bedside practitioners to secure and confirm the airway prior to the surgical airway emergency team’s arrival.

Sync’ing the STEMIs

Kristin Hayden, MSN, RN, CV-BC; Elizabeth Toledo, APN-C; Joseph Solda, RN, BSN, CFRN, NREMT-P CMTE

Hackensack Meridian Network

Background. According to the 2013 American College of Cardiology Foundation/American Heart Association clinical practice guidelines, acute ST elevation myocardial infarction (STEMI) patients should be transferred to a cardiac catheterization lab to undergo emergent coronary revascularization within 120 minutes.¹ Upon review of the 2019 National Cardiovascular Data Registry executive summary for Hackensack University Medical Center (HUMC), our median transfer STEMI time was above this benchmark at 137 minutes. We recognized that by improving our median STEMI transfer time, we would also see improvements in other outcomes, including increased emergency department (ED) throughput and decreased lengths of stay and overall mortality. First, to understand current processes, a failure mode and effects analysis was completed that encompassed the ED, transfer center, cardiac catheterization lab, code teams, anesthesiology, emergency medical services, and the capacity department. We learned of ineffective communication among our teams, with multiple phone calls being made by various team members throughout the process of obtaining medical information. This highlighted the need for a streamlined communication system capable of uniting multiple disciplines in a single location. The failure mode and effects analysis also identified a technological barrier within our electronic medical record that prevented the transfer team from preregistering patients. With granted security clearance, the transfer team would be able to safely register patients and bring them directly to the catheterization lab for the emergent procedure instead of the already-overpopulated ED. Lastly, we reviewed data to ensure that the on-call cardiac team would be available and capable of safely accepting the high-risk patients. A timeline was established with 3 phases to ensure a seamless roll-out of the new process.

Intervention Detail. The new process was proposed and later accepted after several brainstorming sessions with multidisciplinary leaders from identified departments. An essential piece to this new process was a commitment from the emergency medical services team to continue caring for the patient until the on-call STEMI team was truly prepared for a safe hand-off. Once this new process was established, we worked to obtain electronic medical record security clearance for the transfer center, allowing them to preregister and admit patients. This was the key piece in bypassing the ED and allowing patients to go directly to the catheterization lab. Next, all current on-call devices were upgraded to iPhones so that they could receive a secure text message with necessary patient information through an emergency one-touch notification system. We then created a standardized STEMI text message for team notification of a code STEMI, which included all necessary team members and potential consults. By streamlining this process, clinical team members could focus on their specialty, rather than making multiple phone calls. All staff contact information was updated in HUMC’s human resources system to ensure that all the proper phone numbers were in the emergency one-touch notification system and could escalate communication in the event that there was no response from the team member. Tabletop sessions using the emergency one-touch notification system were piloted with team leaders to ensure a smooth roll-out. Staff members were educated on how to respond to the new, secure platform and later, mock STEMIs were completed to ensure the bedside teams were competent. In February 2020, the pilot was initiated Monday through Friday, with on-hours from 7 am to 11 pm. Unfortunately, due to the COVID-19 pandemic, progression of the pilot was placed on hold; however, Plan-Do-Study-Act cycles continued. In June 2020, the pilot extended to 24/7.

Outcomes and Impact. Our goal was to reduce HUMC’s median transfer STEMI time by streamlining this process to meet the 120-minute standard. Five months after the new process roll-out, we reduced our median STEMI transfer time by more than 20% to 106 minutes. We will continue to monitor our progress, especially with the waxing and waning of COVID-19 patients. Simply based on numbers, we can see the impact this has had on patient care. By reducing our door-to-revascularization time, we are preserving valuable heart muscle for each patient. Ironically, these “saved” minutes transform into additional years lived with family and friends. As patients are diverted from the ED to the catheterization lab, this allows ED staff to focus on presently admitted patients who deserve the bedside team’s attention and care. Because of this pilot’s success, we have rolled the process out to ED STEMI patients, as well as inpatient STEMI patients. We are proud to share that this process will soon be rolled out to the stroke, pulmonary embolism, and cardiogenic shock teams, and later, added to our hypothermia protocol. Another qualitative result is the on-call STEMI team’s satisfaction and safety. These team members can now simply respond “yes” to a text message to acknowledge the notification and focus on arriving safely to the hospital to care for the patient. Upon arrival, they have a detailed patient report, to be followed up with a verbal hand-off. This ensures the team is prepared to provide patient-centered care. Knowing that they are supported while caring for a high-risk emergency patient is comforting and essential to our team members.

1. O’Gara PT, Kushner FG, Ascheim DD, et al. 2013 ACCF/AHA guideline for the management of ST-elevation myocardial infarction. J Am Coll Cardiol. 2013;61:e78–e140.

A Systemwide Initiative: Assessing Preparedness to Navigate Airway Emergencies

Pavithra Bora, MBA, CSSBB, Houston Methodist; Stuart Dobbs, MD, Houston Methodist; Debbi Garbade, MSN, RN, CPPS, CPHRM, CPHQ, CPSO, Houston Methodist; Randolph Steadman, MD, MS, Houston Methodist Hospital; Carla Braxton, MD, MBA, Houston Methodist West Hospital; Jason Knight, MD, MBA, Houston Methodist The Woodlands; Robert Phillips, MD, PhD, Houston Methodist

Houston Methodist

Background. Adequate preparation is necessary for successful navigation of airway emergencies. Implementing appropriate interventions quickly is critical to patient outcomes. Houston Methodist is an 8-hospital system with a flagship academic medical center, 6 regionally based hospitals, and 1 long-term, acute care hospital. We are committed to the principles of high reliability and quality patient care. We are preoccupied with failure and are constantly seeking ways to improve the quality of care and experience of care for our patients, staff, and providers. We experienced a serious safety event related to failed airway management. The root cause analysis and subsequent gap analysis identified several opportunities for process improvement, including communication across the care continuum, immediate access to necessary equipment, geographically distanced supplies, malfunction of equipment in the airway carts, and standardization across all of our hospitals. These opportunities led to a systemwide improvement initiative at Houston Methodist. The systemwide initiative was a multidisciplinary, collaborative effort led by the system chair of anesthesia, the chief quality officers (CQOs) from each of our hospitals, and representation from various committees. Our system patient safety department conducted a gap analysis that engaged multiple departments to understand and discuss variation in the process. This presentation is an important part of the 2021 Vizient Connections Summit because often in health care, we approach problems as if they were individual competency issues while underappreciating the system failures. Lack of standardization and/or availability of the necessary equipment when dealing with airway emergencies could lead to preventable harm. We worked to develop a complete process with standardized roles and equipment, and we appreciate the opportunity to share our learning and process improvement journey because it could benefit patients, staff, and providers elsewhere.

Intervention Detail. The system quality and patient safety department identified a need to improve outcomes related to emergency airway interventions. After consulting with system CQOs from each of our hospitals and various stakeholders, an improvement work group was formed that included patient safety, quality, anesthesia, certified registered nurse anesthetist, respiratory therapy, the emergency department (ED), the intensive care unit (ICU), supply chain, education, information technology, and infection prevention. In the initial meeting, the group reviewed system airway events and the airway categories for improvement obtained from our outcomes data for the past 2 years. Group members discussed training, competencies, readiness, communication, utilization, standardization, and location of emergency airway equipment. Checking the functionality of the equipment and situational awareness were areas determined to need improvement. Closed-loop communication about the presence of difficult airways, necessary equipment and supplies, and standardization were also targeted for improvement. Multiple meetings were set to evaluate options and move forward with process improvements. All CQOs and their quality departments were tasked with determining what their facility kept in the difficult airway carts. Each hospital completed a process assessment questionnaire and also provided the list of contents they kept in the bags and carts. All of this information was compiled for the work group review and feedback. The process assessment questionnaire is a gap analysis designed to help users understand their hospital’s process and needs. The questionnaire included clarifying questions such as: do they have mobile intubation kits or/and difficult airway carts? How many do they have versus how many do they need? Where are they located, stored, and maintained? And what is the process of bringing the kits when a code blue is called? The equipment lists for the airway kits, difficult airway carts, and process assessment reports were presented for work group review. Significant variability between our hospitals was identified in the emergency airway management kits, including variable equipment options, the ways in which we identify a patient with a difficult airway, and ongoing communication about the presence of a difficult airway during hand-offs between care teams.

Outcomes and Impact. The multidisciplinary work group formed in May 2020 to standardize airway management systemwide and recommend guidance on the minimum equipment list for emergency response bags, difficult airway carts, and airway alert signage communication. Patients with identified and/or anticipated difficult airways will have an “Airway Alert” sign posted at the head of the bed and the sign will travel with the patient. These signs are especially important in non-ICU areas, as staff members are not as accustomed to specialized airway needs. The alert displays clinical indications to be checked off on the signage. Some clinical indications include tracheal stenosis, new tracheostomy, laryngectomy, anatomic variant, surgical or facial deformity, and/or obstructive sleep apnea—as well as the date of the last successful intubation. Recommendations were presented at System Mortality Committee, System Respiratory Council, System Chief Quality Officer and Chief Nursing Officer Committee, System Quality and Patient Safety Steering Committee, System Education Committee, System ICU Council, and System ED Council meetings. Upon feedback and approval from these committees we passed along the guidance to the System Education Committee to operationalize. The systemwide collaboration has had a positive impact, reinforcing the culture and our commitment to the principles of a high-reliability organization. Over time, we envision that the consistency in our approach and standardization of equipment in all our hospitals will help our rapid response teams, intensivists, and ED physicians as they respond to individual patient needs in the face of increasing stress within our system brought on by the COVID-19 pandemic.

A Tale of Two Metrics: “Observed” and “Expected”—Improving Vizient-Reported Mortality

Sathya Vijayakumar, MS, MBA; Kearstin Jorgenson, MSM, CPC, COC; Guido Bergomi, MHA

Intermountain Healthcare

Background. Intermountain Healthcare, an integrated, 24-hospital health care system covering Utah, Nevada, and Idaho, began participating in Vizient in 2018. Despite efforts to improve the mortality index for many years, Intermountain was at an average 38th percentile for mortality in the 2019 Vizient Quality and Accountability (Q&A) Study. Leveraging data from the Vizient Clinical Data Base to define targeted strategies moved us to the 50th percentile for mortality in the 2021 period zero Vizient Q&A calculator (the results of 2020 Q&A measurement year). This huge improvement in just a year during a global pandemic is the result of strategically addressing very specific opportunities in the Observed and Expected metrics. Our integrated Office of Patient Experience worked across quality and physician advisory services functions to develop a novel improvement algorithm that identifies areas of highest opportunity and defines effective actions for improvement in appropriate areas. The 3 main components of this algorithm are discovery, analysis, and action. In the “discovery” phase, we review mortality data from the Vizient Clinical Data Base and identify opportunities as being in Observed or Expected mortality based on benchmarking. The crux of our novel algorithm is in the “analysis” phase, where we evaluate a set of defined criteria for improvement in Observed versus Expected mortality. Once we evaluate all the defined steps, we work with our service lines to identify actions from a standard template and then support our operating structures in implementing those items. This approach of using a standard algorithm with predefined, clear steps for intervention has helped us drill into specific service lines at any of our 24 hospitals (many of which use a different Vizient risk model group). This has helped us scale our improvement tactics and provide the best patient care without increasing implementation costs.

Intervention Detail. The algorithm for Observed looks at 4 main ideas: (1) are we an outlier for early deaths?; (2) are we an outlier for deaths on transfer?; (3) are we an outlier for late deaths (> 12 d)?; and (4) if none of these items are true, then we begin looking at clinical care opportunity and diagnosis-related group-level data for actionable items. If 1 or more of the questions are a yes, we look at the following: (1) appropriate patient placement and hospice recommendations; (2) transfer optimization; and (3) evaluate for hospice. The algorithm for the Expected follows a standard path of starting with the spread of the relative expected mortality followed by diagnosis-related group-level expected mortality comparison to cohort and the respective top performers in that cohort. We then look at the most common risk variables and comorbidities that represent the highest opportunity. Our analytics team has created a “missingness” dashboard that we then refer to in order to see where we are missing documentation, look at diagnoses per account if needed, and then prioritize sending queries on those conditions. As a process measure, we look at postdischarge query rate, physician query response rate, and the increase in number of times the missing condition was captured. We are also partnering with our clinical documentation improvement vendor to develop appropriate natural-language understanding-based provider nudges for improved documentation. The ”action” part of our algorithm consists of having a standard template of data sharing with leadership that highlights opportunities in a similar manner with appropriate recommendations for improvement. This prevents different leaders across the system (different facility/service line) from interpreting their opportunity and action items differently. It has also helped us track process and outcome measures more consistently.

Outcomes and Impact. Upon applying our novel algorithm to systemwide Vizient mortality improvement initiatives, as a system, Intermountain Healthcare improved from the 38th percentile in mortality in 2019 to the 50th percentile in the 2021 period zero Vizient Q&A ranking. Specifically, we would like to point out in the service lines of neurology (18th to 29th), neurosurgery (11th to 28th), cardiology (36th to 40th), and medicine general (29th to 43rd), overall mortality percentile improvement from 2019 period 4 Vizient Q&A calculator to 2021 period zero Vizient Q&A calculator, respectively. We can also drill down to the improvement in Observed versus Expected for these respective service lines during these Vizient Q&A measurement periods. We define delta Observed and delta Expected here to be the difference between Intermountain Healthcare as a system and the Vizient benchmark. A reduction in delta Observed implies Intermountain Healthcare has improved its Observed mortality. An increase in delta Expected implies Intermountain Healthcare has improved its Expected mortality. Changes in delta Observed for service lines of neurology (3.77 to 0.55), neurosurgery (9.22 to 3.2), cardiology (1.75 to 1.4), and medicine general (0.76 to 0.72) all show an improvement in observed mortality. Change in delta Expected for service lines of neurology (–0.06 to –1.64), neurosurgery (0.97 to –0.94), cardiology (–0.81 to –0.34), and medicine general (–0.79 to 0.58) show that there is still opportunity for improvement in Expected mortality in the neurology and neurosurgery service lines. As seen from our results, this novel algorithm has proven to be a very useful tool for Intermountain Healthcare to scale the Vizient mortality improvement work to many hospitals and service lines across different Vizient model groups while keeping our metrics reliable and constant and our costs down.

Getting Better at Getting Better: Intermountain Healthcare Drives Rapid Improvement

Stacy Reed, RRT, MHA, Intermountain Healthcare; Kearstin Jorgenson, MSM, CPC, COC, Intermountain Healthcare; Sathya Vijayakumar, MS, MBA, Intermountain Healthcare; Mark Ott, MD, Intermountain Medical Center

Intermountain Healthcare

Background. Intermountain Healthcare is a Utah-headquartered, nonprofit health system of 24 hospitals and more than 180 clinics serving Utah, Nevada, and Idaho. Intermountain Medical Center (IMED) is the system’s flagship hospital, with 510 licensed beds. It is a level 1 trauma center, a Comprehensive Stroke Center, a transplant center, and home to the Intermountain Neurosciences Institute. IMED serves as the system’s quaternary care facility, offering a full array of specialty and subspecialty services. In its first year participating with Vizient, IMED was a 3-star hospital and was 34th overall in the Vizient Quality and Accountability (Q&A) ranking. While achieving top marks in efficiency and effectiveness, opportunity for improvement existed in all other categories. In order to better serve its patients and community, hospital leaders challenged all caregivers to “become the standard” for the system and the industry. This elevated vision was shared with all caregivers in a series of forums that underscored the fact that IMED’s work is about “helping people live the healthiest lives possible.” Leaders also established Vizient as a key performance indicator to help monitor progress in getting better at getting better. Caregivers and leaders across Intermountain came together over the last 2 years and used the Intermountain operating model (IOM) to guide improvements that ultimately moved IMED into the top 10 Vizient Q&A ranking. The 3 key components of the IOM include strategy deployment, visual management, and daily improvement.

Intervention Detail. (1) Strategy deployment: A vertical and lateral strategy deployment was conducted between department leaders and panels of subject matter specialists. This new process connected frontline caregivers with all tiers of leadership on key performance indicators and improvement projects identified with Vizient data. (2) Visual management: Tracking methods such as huddle boards monitored measures including nursing bundles and key drivers, leading to improvement in patient centeredness, infection control, and clinical best practices. (3) Daily improvement: Daily tiered huddles created escalation pathways for immediate resolution of concerns. They also provided a forum for reporting lessons learned within hours of concluding a root cause analysis. Kata is a rapid cycle improvement strategy discussed in huddles and applied to time-sensitive measures, such as angiography and neurology device times, as well as sepsis bundle compliance. Concurrent reviews are another daily improvement strategy used to improve query response rate, patient safety indicator reporting, and appropriate removal of patients from the readmission population. Regularly scheduled step back reviews give higher tiered leaders an opportunity to step out into departments for review of process and outcome measures tracked on huddle boards. This helped maintain clarity, alignment, accountability, and the closing of loops on all goals and initiatives. Operations reviews are also cadenced and give specialty teams, such as the hospice work group, infection control, and Intermountain Physician Advisory Services, an opportunity to report progress on value improvement projects.

Outcomes and Impact. Applying the IOM principles to the Vizient work led to increased collaboration and daily rapid cycle improvements at the unit level. Departments and service lines were able to track their progress through concurrent reviews and daily huddles to escalate barriers and improvement gaps to appropriate tiers of leadership for support. Thanks to caregivers’ dedication, hard work, and adherence to the IOM, IMED ED experienced the following successes from the 2018 to the 2020 Vizient Q&A Study results:

• Improved overall Vizient rank from 34 to 9.
• Improved Vizient mortality rank from 67 to 45.
• Improved Vizient safety rank from 60 to 7.
• Improved patient centeredness from 70 to 64.
• Saved $3.5 million in provider education costs.
• Increased care provider query response rate from 81.6% to 91%.
• Increased query rate by clinical documentation improvement team from 10.5% to 17.2%.
• Improved expected mortality from 2.75% to 6.09%.
• Improved early death rate from a high of 1.39 to below the cohort, with a low of 0.71.
• Removed 28 patient safety indicators.
• Removed 6 patients from the readmission population.
• Removed 29 patients previously admitted to inpatient for comfort measures to a more appropriate status.

Unique Approach Helping Those Most in Need During the Pandemic

Kelly Zabriskie, MLS, BS, CIC, FAPIC, Jefferson Health; Efrat Kean, MD, Thomas Jefferson University Hospital; Kristen Vogl, MBA, Thomas Jefferson University Hospital

Jefferson Health

Background. Long-term care facilities (LTCFs) represent a major driver of COVID-19 morbidity and mortality. State and local governments have invested significant resources to control outbreaks within these facilities. The Pennsylvania Department of Health issues regular updates to its COVID-19 regulations for LTCF, including instructions regarding surveillance testing, cohorting positive or exposed individuals, personal protective equipment (PPE) use, and quarantine periods for staff, but the ability of some facilities to comply with regulations is limited by access to resources, knowledge, and time. While the state conducts infection control surveys, they are insufficient to ensure compliance with all guidelines—particularly during an outbreak, when staffing is limited and changing protocols create chaos. As part of the Regional Response Health Collaboration Program (RRHCP), our team at Jefferson Health was tasked with supporting all the LTCF assigned to us in our region. While skilled nursing facilities are required to designate staff as a point person for infection control, state law does not require trained infection preventionists, and other facilities—including personal care homes and assisted living facilities—do not have to meet this metric at all. As a result, these facilities have difficulty accessing necessary resources to comply with state COVID-19 guidance. The RRHCP program allows hospitals to provide facility consultation, including clinical, operational, and administrative support. This model has been highly effective at preventing and containing COVID-19 outbreaks, has allowed the hospital to form strong relationships with LTCF in this region, and has provided resources and guidance to facilities that house our vulnerable community members. This model for public-private partnerships has strong potential for future applications, both in LTCF infection control and quality—a longstanding concern before the COVID-19 pandemic—and in allowing these facilities to access resources and expertise that are readily available in hospital settings but less accessible in the community.

Intervention Detail. Our team at Jefferson Health approached the issue of infection control in LTCF using 3 major targets: prevention, preparedness, and response. We conducted initial site visits at all assigned facilities to assess readiness to prevent or respond to an outbreak. Our team then provided recommendations to facilities on appropriate infection control protocols, PPE, and other key strategies to prevent and contain outbreaks. This included evaluating the staff’s knowledge of infection control protocols in general, as well as COVID-19 regulations. We also offered advice for creating a COVID-19 plan tailored to each facility. Our team also provided PPE to facilities that were having difficulty procuring their own and arranged for fit-testing of N95 masks to ensure adequate preparedness. Our rapid response teams, consisting of a physician, a nurse with long-term care experience, and an infection preventionist, were on call 24/7 to respond to emerging outbreaks and provide additional advice and supplies to achieve rapid containment. Finally, crisis staffing was provided to stabilize facilities short on staffing to ensure quality resident care and the ability for facilities to maintain infection prevention and containment. With an established presence in the community, we understand specific challenges faced by the LTCF in our region. This enables more aligned relationships with facility administrators than would be possible with a state health department program. In addition, these relationships have allowed us to achieve a high degree of success in outbreak prevention and containment.

Outcomes and Impact. To date, the Jefferson Health RRHCP team has conducted 720 assessments and consultations, distributed 4 156 000 PPE items, conducted 66 rapid responses, performed 309 testing missions and 67 staffing missions; and provided N95 mask fit testing to over 6500 facility employees. In addition, we have maintained an on-call schedule in which a physician is always available to facilities remotely to provide guidance or answer questions. While it is difficult to quantify the effect we have had overall, within each facility we have seen our involvement and the interventions of our rapid response teams achieve rapid reductions in case numbers, as exemplified in a large personal care home that experienced an outbreak in September 2020. Our involvement providing resources, staffing, education, and a continued presence enabled the outbreak to quickly resolve over the next 4 weeks. Overall, the response from facility administrators has also been overwhelmingly positive. One facility, representing a community of group homes that services individuals with developmental and intellectual disabilities, sent a letter in February 2021 to the Pennsylvania Department of Health expressing that, “the Jefferson team has been an invaluable resource. Knowing they are available to partner with us should we be presented with a pandemic-related challenge we are not equipped to handle on our own is not a luxury but rather an absolute necessity.”

SAFE @ HOME O2: Ambulatory Management of Supplementary Oxygen for COVID-19 Pneumonia

Josh Banerjee, MD, MPH, MS; Soodtida Tangpraphaphorn, MPH; Catherine Canamar, PhD; Anabel Lemus, BSN, RN, CCDS, CPHQ, CEN; Chase Coffey, MD, MS; Laura Sarff, DNP, RN, MBA, CPHQ, NEA-BC; Brad Spellberg, MD

LAC+USC Medical Center

Background. Faced with the predicted surge in COVID-19 patients, our team at LAC + USC designed expected practices to supply COVID-19 pneumonia patients with supplemental oxygen upon being discharged to home or quarantine. Patients on 3 liters of oxygen or less and stable without other indication for inpatient care were discharged from either emergency or inpatient encounters with home oxygen equipment, education resources, and nursing phone follow-up within 12 to 18 hours of discharge. Nurses provided continued phone follow-up as indicated, always with physician backup. By giving supplemental oxygen outside of inpatient admission, our team hoped to reduce morbidity and mortality due to COVID-19 pneumonia while also freeing up inpatient capacity to patients in more urgent need of placement. The goal of this project was to reduce COVID-19 mortality and prevent readmission to our hospital within 30 days of discharge. To measure our outcomes, we examined patient lengths of stay, all-cause mortality, and hospital readmission rates. Our team also collected constructive feedback from patients about which visual elements helped them learn how to operate the home oxygen equipment most effectively. Based on their feedback, we were able to create an instructional video to teach new SAFE @ HOME patients how to use their equipment.

Intervention Detail. The SAFE @ HOME O2 Expected Practice included several principal care expectations. Patients with COVID-19 pneumonia who were clinically stable and requiring < 3 L per minute of oxygen to achieve = 92% oxygen saturation should be discharged and managed in an ambulatory setting with strict return precautions, in the absence of any other indication for acute care. Our facility stocked portable oxygen and pulse oximeters from the durable medical equipment (DME) vendor in the emergency department (ED). This ensured safe ambulatory management, allowing patients to be sent home directly from the ED after hours without waiting until the DME vendor’s normal business hours the next day. Patients discharged on oxygen must receive a nurse phone call, with physician support if needed, within the first 12 to 18 hours postdischarge. The clinical follow-up program consisted of a nurse manager who assigned cases and performed quality assurance, 3 nurses who shared a schedule to perform follow-up calls, a physician on call for support, and the DME vendor support team that was accessible 24/7 for DME troubleshooting. Calls were performed 7 days a week and continued daily until patients demonstrated that they understood how to use oxygen equipment or presented indications for return to acute care. Patient education reinforced during these calls was consistent with that provided at discharge. Patients were given self-management instructions, including a weaning protocol. Respiratory therapists from the vendor also performed periodic patient checks to help reinforce instructions. Collectively, the team determined appropriate timing of oxygen discontinuation. In addition to verbal education, patients received a printed educational handout and an educational video regarding safety parameters and home oxygen self-management. Supplemental education materials were available in English and Spanish, but both patients and staff had access to on-call translation services for other language.

Outcomes and Impact. From March 20, 2020, to August 19, 2020, 621 COVID-19 pneumonia patients were discharged from the ED (24.0%) and inpatient (76.0%) encounters with home oxygen. Inpatients discharged with home oxygen had a mean length of stay of 5.9 days 95% confidence interval (5.3–6.5), compared to a baseline length of stay of 7.6 days for all COVID-19 patients 95% confidence interval (7.0–8.2), effectively vacating beds 1 day sooner than expected. All-cause mortality for SAFE @ HOME patients was 1.3%. No deaths occurred in the ambulatory setting. The 30-day return to hospital admission rate was 8.5%. There was no significant difference in hospital mortality for patients readmitted after discharge on home oxygen compared to patients admitted without preceding home oxygen (15.1% versus 14.1%; P = 0.6; a = 0.05). COVID-19 patients discharged on home oxygen had mortality rates lower than those published in large surveys of outpatients with COVID-19, and the 30-day readmission rate was below local and nationally reported rates for Medicaid patients. Ambulatory management of COVID-19 patients with home oxygen has an acceptable safety profile and should be considered broadly as a strategy for optimizing outcomes, by ensuring the right care in the right place at the right time and preserving access to acute care during the pandemic. LAC + USC is continuing to follow the expected practices to discharge patients with home oxygen throughout the duration of the COVID-19 pandemic and intends to sustain the practice beyond the pandemic.

Reduction in Unnecessary Blood Transfusions in Hospitalized Patients

Amy Slenker, MD, CPHQ; Warren D. Behr, MBA; Joshua Rosentel, RN, BSN, CPHQ; Matthew McCambridge, MD

Lehigh Valley Health Network

Background. Red blood cell (RBC) transfusion is the most common procedure performed in United States hospitals¹ and is 1 of the top 5 most overused procedures, according to The Joint Commission.² Approximately 10 million units of RBCs are transfused in the United States yearly³ with an annual estimated cost of $5 billion to $12 billion.⁴ Transfusion guidelines emphasize that RBC transfusion should be given for symptoms of anemia and should not be based on hemoglobin concentration alone.⁵ Many clinical trials have shown that restrictive RBC transfusion strategies with a hemoglobin threshold of 7 grams per deciliter (g/dL) to 8 g/dL are noninferior to liberal transfusion strategies of 9 g/dL to 10 g/dL for different patient populations.^6–12 Despite data that RBC transfusions are associated with worse outcomes,^13–15 there remains considerable individual variation in clinical practice, and reductions in inappropriate blood transfusion remain an opportunity for many hospitals in the United States.¹⁶ The introduction of a patient blood management program and different computer-based strategies to promote appropriate blood utilization have been successful in reducing unnecessary blood transfusions.^17–18 Encouraging data from the National Blood Collection and Utilization Survey has documented a decline in RBC transfusions of 13.9% between 2013 and 2015 and 6.1% between 2015 and 2017.³ At Lehigh Valley Health Network, we implemented a successful patient blood management program with a multifaceted strategy to reduce unnecessary RBC transfusions in hospitalized patients with anemia.

Intervention Detail. The Lehigh Valley Health Network patient blood management program began with the development and approval of RBC transfusion guidelines for hospitalized patients with asymptomatic anemia in April 2019. This guideline development was subsequently followed by education to key groups that utilize RBC transfusion in June 2019, including attendings and residents from cardiovascular surgery, internal medicine, family medicine, general surgery, emergency medicine, hematology and oncology, and obstetrics and gynecology. Electronic medical record transfusion orders were modified to promote appropriate blood utilization in July 2019 and now include: (1) an embedded link to the transfusion guidelines; (2) embedded clinical decision support in the order that states, “in hemodynamically stable, nonbleeding patients, a hemoglobin threshold of 7 g/dL (or 8 g/dL with cardiovascular disease) decreases transfusion requirements and reduces adverse outcomes. Single-unit transfusions are usually preferable”; (3) updated clinical indications for RBC transfusion; (4) a default of 1 unit for RBC transfusion; and (5) a best practice advisory to alert providers when ordering blood products if the patient’s hemoglobin is greater than 7 g/dL. Finally, a blood utilization analytics dashboard was created for general use in July 2019, and feedback reports were created and sent on a monthly basis to the key groups noted starting in November 2019.

Outcomes and Impact. The impact of this program has been tremendous. There has been resounding provider support for this program with notable results. After the implementation of our multifaceted intervention, we experienced an 18% decrease in RBC transfusions at our 2 main campuses in FY 2020 from the baseline in FY 2019. RBC transfusions remained stable through the midpoint of FY 2021, with a 16% decrease compared to FY 2019. There was a 5% decrease in RBC transfusions for patients with a hemoglobin of > 8 g/dL from FY 2019 to FY 2020, which was sustained through FY 2021. Additionally, from FY 2019 to FY 2020, there was a 12% increase in the percentage of single-unit transfusions. From FY 2020 to FY 2021, there was an additional 3% increase in single-unit transfusions. Estimated savings from this program include a decreased acquisition cost of $638 280 (approximately $270 per unit) and a total estimated cost savings of approximately $1.8 million in the following fiscal year (mean of $761 per unit, including direct and indirect costs).⁴ Because of its early success, the blood management program has expanded to operative blood conservation strategies that include preoperative anemia management, patient education, and expanded feedback reports to surgeons and anesthesia. Additionally, we have started a program for our local patient population who decline blood products. The program continues to look for better ways to improve the care of our patient population.

1. Agency for Healthcare Research and Quality. Healthcare Cost Utilization Project, Statistical Brief #149. Most Frequent Procedures Performed in Hospitals, 2010. Accessed February 5, 2021. http://www.hcup-us.ahrq.gov/reports/statbriefs/sb149.pdf.

2. The Joint Commission. Proceedings From the National Summit on Overuse. 2012. Accessed February 10, 2021. https://www.parsemus.org/wp-content/uploads/2015/06/AMA-National_Summit_Overuse.pdf.

3. Jones JM, Sapiano MRP, Savinkina AA, et al. Slowing in decline in blood collection and transfusion in the United States - 2017. Transfusion. 2020;60(suppl 2):S1–S9.

4. Shander A, Hofmann A, Ozawa S, et al. Activity-based costs of blood transfusions in surgical patients at four hospitals. Transfusion. 2010;50:753–765.

5. Carson JL, Grossman BJ, Kleinman S, et al. Red blood cell transfusion: a clinical practice guideline from the AABB*. Ann Intern Med. 2012;157:49–58.

6. Hebert PC, Wells G, Blajchman MA, et al. A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care. Transfusion requirements in critical care investigators, Canadian Critical Care Trials Group. N Engl J Med. 1999;340:409–417.

7. Hajjar LA, Vincent JL, Galas FR, et al. Transfusion requirements after cardiac surgery: the TRACS randomized controlled trial. JAMA. 2010;304:1559–1567.

8. Carson JL, Terrin ML, Noveck H, et al. Liberal or restrictive transfusion in high-risk patients after hip surgery. N Engl J Med. 2011;365:2453–2462.

9. Villanueva C, Colomo A, Bosch A, et al. Transfusion strategies for acute upper gastrointestinal bleeding. N Engl J Med. 2013;368:11–21.

10. Holst LB, Haase N, Wetterslev J, et al. Lower versus higher hemoglobin threshold for transfusion in septic shock. N Engl J Med. 2014;371:1381–1391.

11. Murphy GJ, Pike K, Rogers CA, et al. Liberal or restrictive transfusion after cardiac surgery. N Engl J Med. 2015;372:997–1008.

12. Carson JL, Sieber F, Cook DR, et al. Liberal versus restrictive blood transfusion strategy: 3-year survival and cause of death results from the FOCUS randomised controlled trial. Lancet. 2015;385:1183–1189.

13. Isbister JP, Shander A, Spahn DR, et al. Adverse blood transfusion outcomes: establishing causation. Transfus Med Rev. 2011;25:89–101.

14. Musallam KM, Tamim HM, Richards T, et al. Preoperative anaemia and postoperative outcomes in non-cardiac surgery: a retrospective cohort study. Lancet. 2011;378:1396–1407.

15. Bernard AC, Davenport DL, Chang PK, et al. Intraoperative transfusion of 1 U to 2 U packed red blood cells is associated with increased 30-day mortality, surgical-site infection, pneumonia, and sepsis in general surgery patients. J Am Coll Surg. 2009;208:931–937.

16. Frank SM, Savage WJ, Rothschild JA, et al. Variability in blood and blood component utilization as assessed by an anesthesia information management system. Anesthesiology. 2012;117:99–106.

17. Sadana D, Pratzer A, Scher LJ, et al. Promoting high-value practice by reducing unnecessary transfusions with a patient blood management program. JAMA Intern Med. 2018;178:116–122.

18. Thakkar RN, Lee KH, Ness PM, et al. Relative impact of a patient blood management program on utilization of all three major blood components. Transfusion. 2016;56:2212–2220.

Uniting Physical Therapy and Emergency Medicine to Reduce Admissions

Gillian Beauchamp, MD; Richard MacKenzie, MD, MBOE, FACEP; Joshua Rosentel, BSN, RN, CPHQ; Daniel Sawyer, PT, DPT; Ryan Vetter, MS, OTR/L

Lehigh Valley Health Network

Background. The length of time between a medical back pain patient seeking treatment and an evaluation by a physical therapist has shown to have a significant impact on patient outcomes.¹ Swift physical therapy (PT) initiation leads to lower surgery rates, cost, and opioid use. In addition, because inpatient PT evaluations normally require a minimum 1-day stay, evaluation in the emergency department (ED) could potentially reduce admissions, resource utilization, length of stay, and cost.² As of Q3 FY 2021, Lehigh Valley Health Network (LVHN) has been a consistently high outlier for back pain admissions on the Centers for Medicare & Medicaid Services Program for Evaluating Payment Patterns Electronic Report for 8 of the last 12 quarters. The medical back pain metric on the Program for Evaluating Payment Patterns Electronic Report report is a count of all discharges for medical diagnosis-related groups in major diagnostic categories (MDCs) 08 as the denominator and those discharges in Medicare severity diagnosis-related groups (MS-DRGs) 551 and 552 as a numerator. LVHN also has a high percentage of outpatients receiving MRI diagnostic imaging for back pain prior to a PT evaluation, placing us above both state and national averages. An analysis was also done using the Vizient Clinical Data Base Report Builder. The Centers for Medicare & Medicaid Services medical back pain metric was replicated using a patient outcomes report and a member cohort for comparison was created with hospital profiler that included the Vizient Quality and Accountability Comprehensive Academic Medical Centers (AMCs) for the appropriate year. In FY 2019, out of 92 comprehensive AMCs, LVHN had the third-highest rate of admissions in MS-DRGs 551 and 552 out of the total MDC 08 population. The utilization of PT evaluations in the ED was also studied among the same comparison cohort. Use of this ancillary service was low across the entire cohort.

Intervention Detail. LVHN began the project in March 2019 by pulling data regarding arrivals to the ED for back pain and the most common hours of the day when these arrivals occurred. This informed the most opportune times that PT could be staffed for ED evaluations. Additionally, this provided an expected volume of patients for arrival. Meetings between ED and PT personnel began in June 2019 and protocol development started for PT evaluations in the ED. Over the next several months providers revised the protocol and a final algorithm was completed in early 2020. Education for ED providers began in February 2020. However, we soon shifted focus to manage the COVID-19 pandemic, and most quality improvement work was halted. Like many other organizations, the pandemic forced us to halt all elective procedures and visits. Summer 2020 began a gradual return to a normal state. A dashboard was built to track data and analytics regarding back pain patients arriving to the ED, their disposition status, diagnostic testing, and PT evaluations, as well as how often each provider was utilizing PT in the ED for appropriate patients. The data showed that while patients who received PT care in the ED had positive results, the algorithm was underutilized. Provider outreach and reeducation was necessary but was again interrupted by a second peak of COVID-19 admissions. As COVID-19 cases began to reduce in late January 2021 through early February 2021, ED and PT personnel collaborated to produce reeducation information that was distributed via email to ED providers by the ED vice chair of quality.

Outcomes and Impact. LVHN’s data and analytics dashboard has provided unique insight into the benefits of PT evaluations in the ED. While usage of the algorithm for back pain patients has not yet occurred at the desired rate largely due to the pandemic, several tracked metrics have shown clear benefits for patients. On average, patients evaluated by PT in the ED using the algorithm wait 1 hour for the order to be placed and an additional 1.9 hours for the evaluation to occur, compared to 14.5 hours and 25.5 hours, respectively, when evaluated on an inpatient or observation unit. We have also received positive satisfaction surveys for patients evaluated by PT in the ED. Patients have specifically cited satisfaction in the rapid PT response time, use of PT over opioid analgesics, fewer diagnostic imaging tests, and overall improvement in their symptoms. Patients evaluated by PT in the ED also exhibited a lower 7-day return rate of 5.88% compared to 10.63% for those evaluated by PT after admission. Admission rates remained relatively stable between FY 2020 and FY 2022 to date at 34.5%. While several metrics have shown improvement, LVHN continues to experience a high admission rate for back pain patients. FY 2021 Q1 and Q2 reports run in the Vizient Clinical Data Base utilizing the Quality and Accountability Comprehensive AMCs cohort continue to show LVHN as having some of the highest admission rates for MS-DRGs 551 and 552 compared to MDC 08 discharges. We recognize the PT in the ED algorithm as another tool to reduce back pain admissions and will continue with data and analytics tracking and provider education, outreach, and feedback to ensure expanded use of the algorithm, which will be gradually implemented in all health system EDs.

1. Fritz JM, Childs JD, Wainner RS, et al. Primary care referral of patients with low back pain to physical therapy: impact on future health care utilization and costs. Spine (Phila Pa 1976). 2012;37:2114–2121.

2. Magel J, Kim J, Fritz JM, et al. Time between an emergency department visit and initiation of physical therapist intervention: health care utilization and costs. Phys Ther. 2020;100:1782–1792.

Interdisciplinary VTE Task Force: Cleaning Up the Clotter

Sameh Samy, MBBCh, MSA, CPHQ; Zachary Lockerman, MD, MBA, FACG; Angela Sung, PharmD, BCPS

Maimonides Medical Center

Background. Pulmonary embolism and deep venous thromboembolism, collectively referred to as venous thromboembolism (VTE), are potentially severe complications that can occur following major surgical procedures, potentially leading to patient death. Over time, health care organizations have recognized the importance of preventive measures to minimize postoperative VTE risk and are implementing these measures throughout their facilities. One of the ways Maimonides Medical Center (MMC) observes our performance is through the Centers for Medicare & Medicaid Services (CMS) Five-Star Quality Rating System. The CMS Star Rating from July 2019, which reflects data from 3rd quarter 2016 to 2nd quarter 2018, indicated our VTE rate to be 4.73 per 1000 discharges, which is worse than the national rate of 3.85 per 1000 discharges. In addition, our VTE rates over the last 3 years increased each year, while the national rates decreased. These findings led to MMC’s project of creating an interdisciplinary task force and action plan to decrease postoperative VTE events and improve patient outcomes. The project is important as it highlights institutionwide efforts to identify causes and solve problems in a collaborative manner. It is also relevant to today’s COVID-19 issues and post-infection complications. As institutions may have more patients at risk of developing VTE due to COVID-19, it is crucial for the VTE prophylaxis process to be refined to protect patients.

Intervention Detail. MMC’s VTE Task Force was organized in September 2019 and comprised of members from multiple departments, including quality management, surgery, orthopedics, gynecology, perioperative services, anesthesia, pharmacy, nursing, and information technology. The task force identified areas for improvement and the action plan utilized Plan-Do-Check-Act methodology. Improvement strategies included: teambuilding, process mapping, staff involvement, cause and effect analysis for sources of variations, goal-setting, assigning responsibilities, creating an implementation timeline, and establishing measures of effectiveness. The main interventions included:

• Implementing a mandatory process for VTE risk assessment upon patient admission.
• Updating the institution’s VTE prophylaxis protocol to include the most current guidelines.
• Implementing preoperative VTE prophylaxis for indicated major procedures.
• Conducting root cause analysis audits on all postoperative VTE cases to identify improvement opportunities.
• Creating an educational module on VTE prophylaxis protocol for relevant departments.
• Standardizing utilization and interpretation of venous duplex studies.

Each intervention’s progress is followed and discussed at the monthly VTE Task Force meeting. Project progress is measured by a monthly, internal, no-harm dashboard that includes all CMS patient safety indicators. The dashboard shows the postoperative VTE cases by month and variance from previous years and allows team members to discuss results and next steps.

Outcomes and Impact. The project goal was to decrease the VTE rate by 20% in 1 year. The current data, monitored on a monthly basis through our internal, no-harm dashboard, showed a 45% reduction of VTE events in 2020 as compared to 2019. In addition, data on risk-adjusted postoperative VTE from the Vizient Clinical Data Base demonstrated a significant reduction of 54% in VTE cases from 2019 to 2020. The observed/expected ratio also decreased from 1.68 to 0.92 in the same time period, and the downward trend continues in 2021. The significant reduction in the VTE rate not only reflects positive patient safety outcomes, but also the significance of working in a multidisciplinary group comprised of different professions and disciplines.

Total Hip and Total Knee Replacement Complications and How the ED Can Help!

Marie Reisdorfer, MS, Mayo Clinic; Betzalel Reich, MD, Mayo Clinic Health System; Jeannie Tenold, MS, Mayo Clinic

Mayo Clinic

Background. In 2018, Mayo Clinic Health System (MCHS) Austin experienced an increased rate (3%) of elective total knee and total hip complications. The complications varied from a pulmonary embolism and pneumonia to sepsis and periprosthetic joint infection/wound infection. In 2019, MCHS Austin started an interdisciplinary work group focused on improving hospital quality, with orthopedics complications being 1 focus area. The group completed a chart review for all 2018 and early 2019 cases to identify gaps in care and brainstormed ideas on how orthopedics could partner closely with the emergency department (ED) on post-surgery care.

Intervention Detail. The chart review revealed that all patients were seen via the ED. Joint and wound infections were the greatest complaints that resulted in the orthopedic complication. A gap was identified where ED and orthopedic providers should be comanaging patients prior to antibiotics being started and subsequent to readmission. ED and orthopedic leaders developed a suspected infection treatment algorithm that was deployed throughout the region. A real-time, electronic health record alert was developed to show activity of orthopedic patients 90 days post-surgery. The quality department managed those alerts and shared when a patient entered the ED or was admitted. That patient’s chart was reviewed to ensure that the treatment algorithm was followed. If not, ED leadership would coach ED colleagues on the importance of following the best practice treatment algorithm.

Outcomes and Impact. Using the Vizient Clinical Data Base, data was analyzed and tabulated to show the statistically significant reduction of elective orthopedic complications for MCHS Austin. Following implementation of the suspected infection treatment algorithm in all EDs throughout the region, elective total knee and total hip complication decreased by 92% (from January 1, 2020 to September 1, 2021). Patients who previously experienced periprosthetic joint or wound infection decreased by 82% during the same time period. Decreases were also noted for all other complications. With the use of a standard treatment algorithm, we avoided unnecessary use of antibiotics and an unneeded inpatient stay. Readmission data was also pulled via the Vizient Clinical Data Base to analyze the impact on 30-, 14- and 7-day readmissions. From January 2018 to January 2020, MCHS Austin improved the 30-day readmission rate from 3.96 to 3.36, the 14-day readmission rate from 2.57 to 2.06, and the 7-day readmission rate from 1.41 to 1.36. The dual impact is positive both for the patient and the Austin facility. Patients with complications are being cared for at the most appropriate and cost-effective level of care while avoiding unnecessary readmissions, and inpatient bed management is maximized to ensure we have space for other patients.

Hypoglycemia Command Center and Insulin Management Learning Response Team

Bela Patel, MD, FCCP, CMQ, FCCM, Memorial Hermann-Texas Medical Center, UT Health McGovern Medical School; Jeffrey Chen, MD, Memorial Hermann-Texas Medical Center, UT Health McGovern Medical School; Michelle Narat, MS, Memorial Hermann-Texas Medical Center, UT Health McGovern Medical School

Memorial Hermann-Texas Medical Center

UT Health McGovern Medical School

Background. Hypoglycemia occurs in 7% to 38% of patients admitted to general inpatient wards, and up to 40% of patients who have 1 hypoglycemic event are at risk of having another event.^1,2 Inpatients with hypoglycemia have more adverse outcomes, higher mortality, and longer admissions.^3,4 At Memorial Hermann-Texas Medical Center in Houston, Texas, we estimate that within a 12-hour period, we have 2 to 5 hypoglycemic inpatient events—regardless of whether patients are on an insulin regimen. The rate of hypoglycemic events following insulin administration is one of the lab metrics within the Vizient safety domain. From April 2018 to March 2019, the Vizient 2019 Quality and Accountability Study calculator showed that 5.4% of Memorial Hermann-Texas Medical Center’s patients on insulin had a hypoglycemic event (glucose ≤ 50) on the day of or day after insulin administration. Many factors may contribute to inpatient hypoglycemia occurrence, making effective and proper glucose management and prevention difficult and complex. A key factor in facilitating a multidisciplinary approach to improve this lab metric was the need to orient various groups such as bedside nursing, rapid response, endocrinology, primary team physicians, and hospital leadership to a single, centralized monitoring tool that would heighten awareness of hypoglycemic events in near real time. This command center-style tool, in the form of a Tableau dashboard, also aligned the definition of hypoglycemia to a glucose level of less than 60 for all disciplines involved.

Intervention Detail. This project strives to improve hypoglycemia in the insulin population through a systems-based approach to respond to active hypoglycemic events in the hospital. First, a situational awareness Tableau dashboard was developed to monitor hypoglycemic events in real time. The dashboard displays patients over the age of 18 with a glucose level < 60, with emergency room, intensive care unit, and operating room patients excluded. The patient’s unit location and time since hypoglycemia are clearly displayed; symbols denote which patients are still hypoglycemic (and need immediate action), which have received insulin in the past 26 hours, and a trend of all glucose values since each patient’s admission. Within 15 minutes of a glucose < 60 being collected, the patient will populate onto the dashboard, and the information is made available to a larger multidisciplinary response team. In the initial 2019 Plan-Do-Study-Act improvement cycle, rapid response communicated with the bedside nurse to ensure appropriate short-term interventions were conducted. In an insulin patient, the primary team physician is immediately notified to evaluate the root cause of hypoglycemia, and the endocrine service resident is available for further assistance. During this first improvement cycle, we recognized the need to strengthen bedside nursing response. To reduce the time to nursing intervention, nurses check premeal, point-of-care glucose instead of patient care assistants. Additionally, repeat glucose checks occur less frequently in initial hypoglycemic events when the glucose level is between 50 and 60, versus initial events with glucose levels less than 50. As a result, in Q4 2020, nursing implemented an aggressive “q15min glucose < 60” campaign to reinforce our hypoglycemia protocol. This includes diabetic educators that provide face-to-face training for each bedside nurse, as well as nursing managers using the hypoglycemia dashboard to track the bedside process.

Outcomes and Impact. Our use of a real-time, situational dashboard, coupled with the involvement of bedside nursing, rapid response, physicians, endocrinology, and diabetic educators, had a direct impact on the Vizient hypoglycemia lab metric. After the first Plan-Do-Study-Act cycle, the metric value decreased from a baseline of 5.4% (April 2018 to March 2019) to 4.58% (April 2019 to March 2020). However, beginning in April 2020, COVID-19 completely paused our multidisciplinary efforts to monitor and improve hypoglycemia. As a result, the hypoglycemia metric value climbed back up from 4.58% to 5.7%—likely attributed to competing priorities and large numbers of traveling staff. Revisiting dashboard-driven improvement efforts in Q3 2020, more focus was placed on streamlining the bedside nursing response so that rapid response could serve as a safety net instead of first-line response. Having nurses checking premeal sugars instead of patient care assistants has reduced the time to intervention and physician notification. The “q15min glucose < 60” campaign has put a spotlight on the hospital’s hypoglycemia protocol and simplified the way nurses approach hypoglycemia. The diabetic educators have reeducated all nurses on 2 of our hospital units with the highest hypoglycemia rates. Using the dashboard, nursing unit managers can view and track events over time to identify patterns to address. Although we are still seeing high numbers of COVID-19 patients and new staff, the Vizient hypoglycemia lab metric has dropped from 5.7% to 5% since resuming improvement efforts, and we look forward to further improvement as we see the effects of our additional pilot and expanded nursing efforts throughout our hospital.

1. Turchin A, Matheny ME, Shubina M, et al. Hypoglycemia and clinical outcomes in patients with diabetes hospitalized in the general ward. Diabetes Care. 2009;32:1153–1157.

2. Umpierrez GE, Isaacs SD, Bazargan N, et al. Hyperglycemia: an independent marker of in-hospital mortality in patients with undiagnosed diabetes. J Clin Endocrinol Metab. 2002;87:978–982.

3. Khazai NB, Hamdy O. Inpatient diabetes management in the twenty-first century. Endocrinol Metab Clin North Am. 2016;45:875–894.

4. Hulkower RD, Pollack RM, Zonszein J. Understanding hypoglycemia in hospitalized patients. Diabetes Manag (Lond). 2014;4:165–176.

Clearing the CMI Clouds: Standardized Documentation Tools Improve Severity of Illness Capture

April Junker, RN, Nebraska Medicine; Justin Birge, MD, MS, Nebraska Medicine; Micah Beachy, DO, FACP, SFHM, Nebraska Medicine—University of Nebraska Medical Center; Tammy Winterboer, PharmD, BCPS, CPHQ, Nebraska Medicine

Nebraska Medicine

Background. Translating severity of illness (SOI) from clinical to coding terminology can sometimes be a challenge for clinicians. SOI has far-reaching implications on quality and reimbursement. Nebraska Medicine sought to improve SOI capture through standardized electronic health record tools, with the goal of creating efficient and well-adopted workflows. One year after implementation, the cardiothoracic surgery service line demonstrated > 90% workflow adoption, increased mean case mix index (CMI), decreased observed-to-expected mortality and length of stay (LOS) indexes, and reduced median note-editing time. Both quality and reimbursement rely heavily on SOI documentation at the time of admission. Because SOI definitions are dynamic, organizations and clinicians face challenges while striving to maintain performance excellence. To successfully achieve and maintain excellence, health systems may rely on recurrent education or communication on the importance of documentation rather than allowing performance to erode over time. This is especially true at academic medical centers where house officers—who will have expected annual turnover—may be the primary documenting clinicians. Furthermore, SOI calculators and/or coding rules often depend on specific definitions or terminology that may be unfamiliar or irrelevant to documenting clinicians. When key terms are not used, SOI, anticipated resource need, and the expected portion of observed-to-expected indexes may be impacted. Translating familiar clinical concepts to appropriate use of these key terms is an area of opportunity for health systems. Pairing a standardized, team-based, documentation workflow with a dynamic documentation support tool is one potential solution. This workflow and system tool also lends itself to periodic updates as risk calculators and coding rules change, allowing the clinician to focus on maintaining usage of the tool without needing to be reeducated on the new coding specific language. Implementation of such a workflow and documentation tool would positively impact both quality metrics and CMI, while improving documentation efficiency in a sustainable way.

Intervention Detail. In January 2020, Nebraska Medicine implemented a standardized, team-based approach to documentation in the pre-identified clinical specialty area of cardiothoracic surgery. This inpatient-focused documentation effort included physician, fellow, resident, and advanced practice provider (nurse practitioner and physician assistant) authors. The new documentation approach included a problem-based model and efficiency tools for documenting clinician-identified common problems. Standard note templates for the specialty area were created for initial, subsequent, and discharge documentation. The initial note template included the integrated, dynamic documentation support tool to assist in capturing specific, appropriate variables identified by the health system team as opportunities to more accurately reflect SOI. This tool uses focused rules to prompt documenting clinicians to confirm or clarify the presence of these variables (based on available electronic health record data) and document an appropriate, efficient plan, when necessary. The impact of this intervention was monitored over 1 year, including CMI, observed-to-expected mortality and LOS indexes, and note-editing times.

Outcomes and Impact. Overall, mean CMI for the cardiothoracic surgery team increased from 8.4 to 10.15 (21.4%) during the 1-year observation period. Although the total case volume was smaller in 2020 (694 versus 816), the mean CMI was higher than each corresponding month in 2019. Additionally, the observed-to-expected mortality (0.91 to 0.78) and observed-to-expected LOS (0.96 to 0.93) ratios improved. Because advance practice providers were the primary authors for cardiothoracic surgery, advanced practice provider champions were engaged to ensure appropriate workflow and content development, and to serve as peer support during the implementation period. Workflow adoption and note template utilization for the cardiothoracic surgery team was greater than 80% during the first week of implementation and sustained at greater than 90% during the 1-year observation period (5139 total notes). Because of the intent to create documentation efficiencies, note-editing time was also measured for all relevant visit types, including 1 year prior to implementation. A 9-minute and 54-second reduction in median note-editing time was observed across note types when comparing 2019 to 2020. At 1-year post-intervention, the service line had reduced total note-editing time by approximately 62 hours per month. Due to this success, critical care surgery, emergency general surgery, and trauma surgery service lines began a similar intervention January 2021.

Multivisit Patients: Cross-Continuum Care to Reduce Hospitalizations

Kate Thomas, MSN, RN

Northwestern Medicine Lake Forest Hospital

Background. In spring 2018, Northwestern Medicine Lake Forest Hospital started a journey to achieve top decile performance for 30-day all-cause unplanned readmissions. At the time we were rarely achieving top quartile and overall, were sitting under the median. We started with the standard approach of looking at the Centers for Medicare & Medicaid Services condition- and procedure-specific rates. In general, we were performing quite high. We also looked at hospitalwide service line performance. Lastly, we looked at readmission rates per patient. This multipronged approach gave us 3 main buckets of opportunity to reduce 30-day all cause unplanned readmissions—sepsis, chronic obstructive pulmonary disease, and multivisit patients (MVP). There is no single golden key to tackling readmissions—this portfolio of work is what led us to our success. We used the define, measure, analyze, improve, and control methodology to dive into the multivisit patient population. Our team was not only multidisciplinary but also cross-continuum. We knew these patients required better coordination and we needed to start looking at the problem from their perspective—not as inpatient versus outpatient episodes of care. We defined “multivisit” as 3 or more readmissions in 12 months. We found that 3% to 15% of our unique patients accounted for up to 28% of total readmissions. To better understand the specific reasons these patients were utilizing acute care services we conducted multidisciplinary case reviews for each patient. Not surprisingly, we found these patients had multiple comorbidities, although interestingly, they had good access to specialists. In addition, these patients typically readmitted much later in the 30-day window than the rest of our patient population (27% at week 4, compared to 20%, respectively). The main contributing factors identified were social determinants of health and lack of care coordination. Once we moved into the “improve” phase, we maintained our original team and created a biweekly working meeting to review patients and ongoing care plans.

Intervention Detail. To begin our formal “improve” phase, we ensured that we had a multidisciplinary team that spanned the care continuum. Next, we outlined explicit roles and responsibilities. The major players were our ambulatory care coordination (ACC) team, inpatient social workers (SWs), primary care operations, and our hospitalist champion. Next, we outlined the ideal process for coordinating care of these patients. The major line of communication was between ACC and SW. Our first guiding principle is that for each MVP admission to the hospital, the SW and ACC have a warm hand-off with the patient. In addition, we found that only about half of the MVPs had a Northwestern Medicine primary care physician (PCP), which was likely leading to fragmentation in care. For those without a PCP, our primary care operations created an avenue to establish care with a Northwestern Medicine PCP. For those with a non-Northwestern Medicine PCP, our hospitalist called the patient’s PCP to discuss their case and potential transfer of care to a Northwestern Medicine PCP. Our hospitalist is also accountable for creating patient care plans, which are in the patient’s medical record for review by any clinician. This added visibility for our emergency department (ED) clinicians as well. Any MVP who arrives to our ED is seen by an inpatient social worker. This, combined with the care plan, provides the ED with better outpatient discharge plans and follow-up pathways. To keep the momentum going, we created an accountability structure in which the team met every other week. We always started with data and how we were performing. We would then launch into case reviews to resolve identified barriers in real time. A large amount of our work comes out of the ASPIRE work done by Amy Boutwell, MD. Many of the guiding principles in the ASPIRE tool kit were adopted into our own operations.

Outcomes and Impact. MVPs who go through our program have a 56% lower visit/patient ratio than those who do not go through the program. In order for MVPs to go through our program, they must have a Northwestern Medicine PCP. For those who do not, we attempt to establish care with a Northwestern Medicine PCP. To date, we have a 40% Northwestern Medicine PCP conversion rate (17 total patients). Our primary outcome metric is visits/patient. We have both an intervention and a nonintervention group. Intervention group members are referred to the ACC population. The ACC population had an average visit rate of 0.33 visits per patient (ranging from 0.14 to 0.65) during calendar year 2020. The non-ACC population has an average visit rate of 0.75 visits per patient (ranging from 0.40 to 1.22) during the same time period. Our ACC population stays in the cohort until 6 months after they graduate the program, allowing us to hold ourselves accountable for successfully integrating patients back into the community. Patients graduate once they complete all goals that are developed through a shared decision-making process. Our operational process metrics for MVPs are Northwestern Medicine PCP conversation rate, MVPs without a PCP, and MVPs with a non-Northwestern Medicine PCP. The latter 2 are actionable lists within our electronic report. We then have 4 primary process metrics for our ACC population: (1) percentage engaged (monthly outreach and goals set); (2) warm hand-off prior to discharge; (3) 14-day appointment scheduled prior to discharge; and (4) 14-day appointment attended. All metrics are in an electronic dashboard that is reviewed monthly and accessible to all team members. Data transparency has been a key contributor to the program’s long-term success.

Quality Improvement Methods Accelerate and Optimize COVID-19 Regulatory Reporting

Teresa Pollack, MS CPHQ; Anne M. Bobb, MBA, RPh, CPHQ; Rachael Beres, BSN, CMSN; Sally Buttry, QMS; Sheila Neiner, BSN, RN, CPHQ; Christina Silkaitis, MT(ASCP), MBA, CIC, FAPIC; Cindy Barnard, PhD, MBA, MSJS

Northwestern Medicine

Background. Northwestern Medicine (NM), a growing, integrated, academic health system in Illinois, includes 10 hospitals and over 200 ambulatory sites. Public health reporting is a vital core of pandemic control and public safety. In Illinois and nationally, reporting has been inconsistent and incomplete, contributing to delayed interventions and profound racial and socioeconomic inequities in mortality and morbidity.^1–4 NM infection preventionists report communicable diseases to the Illinois Department of Public Health through the online Illinois National Electronic Disease Surveillance System (I-NEDSS). This regulatory reporting provides clinical data to local and state health departments to identify and track infectious diseases and improve, prevent, and control outbreaks. Early in the COVID-19 pandemic, I-NEDSS launched a comprehensive COVID-19 data collection tool with almost 10 times the data burden (110 fields) of other diseases. NM staffing demands in spring 2020 were significant, requiring difficult decisions to allocate quality and infection prevention staff to the most critical needs as we adapted clinical, electronic medical record, and administrative systems to the massive demands of COVID-19. To meet public health needs and optimize use of staff in this national emergency, NM quality division organized a 10-person urgent intervention team to manage the reporting burden and ensure timely, complete reporting for critical public health response. The insights and strategies developed in this emergency are directly relevant to other urgent improvement cycles, emergencies, and disaster response and have lessons to teach for any quality improvement process. As a learning health system, NM is committed to developing insights from successful work and disseminating them internally and externally.

Intervention Detail. Launch: By March 2020, COVID-19 cases increased rapidly and in-house testing was implemented and ramped up. Standardized reporting was implemented to trigger I-NEDSS abstraction. Performance management was updated every 1 to 2 days with new expectations for daily case abstraction goals. Growth: Using rapid improvement cycles, the team implemented feedback and revision to embed best practices and continuously optimize. By May, the team completed 67 cases/d, but did not have the capacity to meet case volumes. Hitting the productivity wall: Rapid improvement cycles and qualitative feedback from the pilot team prepared our leaders to scale training and team capabilities to onboard 50 additional staff members. A tiered regional leadership team was recruited. Video and mixed media training equipped the expanded team effectively. Survey results found that 72% of the new leaders felt very/extremely confident that they received the right resources and training. However, time commitment and effort remained a concern. In June, 58% of participants reported that workload could not be comfortably maintained for longer than 2 months, and 12% reported that their workload was already unsustainable. Technology and process breakthrough: Project leaders designed an intervention to automate the abstraction process. A robotic process automation (RPA) was built by the information systems team to electronically emulate the actions of human data entry into the I-NEDSS portal. This technology is novel for regulatory reporting purposes. The quality team used the rapid improvement cycle to address internal validation, operations, and testing. The RPA process launched in August and saved an estimated 12 minutes per case. From September to mid-October, the positive case volume averaged 114 cases per day, for an average time savings of 22.8 work hours daily—and some days exceeding 24 hours saved. Stabilization: Ultimately, the team was able to manage the workload with approximately 3 full-time equivalents, down from a peak of more than 18 full-time equivalents across more than 80 trained staff members.

Outcomes and Impact. Unfortunately, while the NM team improved productivity, data demands grew more substantial. The I-NEDSS data fields expanded as emerging evidence grew for COVID-19 reporting. By October 2020, the reportable fields grew 62% (to 179 fields) and the caseload grew exponentially. The expanded quality and infection prevention team reported 14 150 positive COVID-19 cases from May to mid-October. The leadership team stayed in close communication with Illinois Department of Public Health epidemiologists and ranked data fields from low to high priority. In mid-October, during a Midwest surge, NM saw 470 cases/d with a positivity rate of 19.4%. The I-NEDSS automated system positioned NM well, as the RPA process now fulfilled the data requirements for outpatient cases. At this point, the RPA process saved an average of 276 human staff hours/day. The quality team continued to abstract the more complex reporting needs for emergency and inpatients (25% of positive cases). Beyond the vital public health and regulatory compliance benefit, this substantial quality improvement initiative for COVID-19 reporting positively impacted staff engagement during a time of remote work and isolation for many team members. Survey results revealed that 62% are likely to connect to a colleague outside of their local team, and 38% are likely to design a future project modeled on the I-NEDSS work. Quality improvement methods were critical to COVID-19 regulatory response. These rapid improvements kept the quality team agile and flexible as emerging evidence, public guidance, and regulatory reporting changed every 7 to 14 days. Innovative organizational leadership designed internal and external systems, engaged experts, and implemented a novel technology to partner with public health agencies.

1. Mahr J. One Case? or 38? Pandemic Exposes Inconsistent, Incomplete Data on COVID-19 Cases in Illinois. Chicago Tribune. 2020. Accessed April 28, 2020. https://www.chicagotribune.com/coronavirus/ct-coronavirus-missing-inaccurate-data-20200425-7d7a3mzf4vgundc5jgzeaznfay-story.html.

2. WBEZ Chicago. Illinois Has holes in its COVID-19 Data. Will That Hinder Planning for Future Outbreaks? 2020. Accessed May 29, 2020. https://www.wbez.org/stories/illinois-has-holes-in-its-covid-19-data-will-that-hinder-planning-for-future-outbreaks/20bfea8f-c140-404d-bf2e-4c8795e2ce8c.

3. American Hospital Association. HHS Issues Updated COVID019 Data Reporting Guidance. 2020. Accessed December 18, 2020. https://www.aha.org/news/headline/2020-12-10-hhs-issues-updated-covid-19-data-reporting-guidance.

4. Meadows AJ, Oppenheim B, Guerrero J, et al. Estimating infectious disease underreporting at the country level: a model and application to the COVID-19 pandemic. The Lancet, preprint October 8, 2020. Accessed December 1, 2020. Available at SSRN: https://ssrn.com/abstract=3706059.

Antibiotics: How Immediate Care Centers Reduced Prescribing by 38% (Relative)

Dharmesh Patel, MBA, CNMT, R.T.(N)(ARRT); Jeffrey Linder, MD, MPH, FACP; Lubna Madani, MD; Stephen Persell, MD, MPH, FACP; Phillip Roemer, MD, FACP; Sonali Oberoi, MHA, OTR; Mark Greg, PharmD, RPh; Teresa Ng, BS

Northwestern Medicine

Background. Inappropriate use of antibiotics creates bacteria that is resistant to antibiotics—increasing community health risks, exposing patients to unnecessary adverse effects, and raising the health care system’s preventable costs. According to the Centers for Disease Control and Prevention, outpatient settings are accountable for an estimated 80% to 90% of human antibiotics use. Furthermore, at least 30% of the antibiotics used in outpatient settings are unnecessary and account for more than 60% of United States antibiotics expenditure.¹ The inappropriate use of antibiotics is particularly magnified in urgent care settings. According to a recent study, at least 46% of antibiotics prescribed in urgent care were inappropriate for respiratory diagnoses.² Immediate/urgent care centers at Northwestern Medicine prescribed antibiotics for upper respiratory infection, pharyngitis with negative strep, acute bronchitis, and influenza at a rate of 34% in FY 2019.

Intervention Detail. Interventions included:

• Data sharing: Regional-level, comparative, antibiotics prescription data was shared with clinicians and leadership through multiple regional and quality committee meetings.
• Individual clinician prescription rate and peer comparison: A quality measure summary report was developed using the enterprise data warehouse, which provided individual clinician antibiotic prescription rates and compared them with the overall practice and regional rate. This report included 4 antibiotic measures, the denominator for the last 12 months, and individual and regional-level antibiotic prescription rates. Regional leadership reviewed the report with individual physicians and advance practice professionals annually to discuss potential barriers and opportunities for improvement.
• Commitment poster: The Centers for Disease Control and Prevention commitment poster was enlarged to 18 × 24, signed by clinicians working in the clinics, and posted in each exam room facing the exam table.
• Patient education pamphlets: The trifold patient education pamphlet was made available to practices, explaining the potential harm of using antibiotics for viral illnesses.
• Clinician education: All immediate/urgent care clinicians were assigned to watch an internally developed, 3-minute video showcasing a physician/patient interaction. The video showed a physician explaining a viral illness diagnosis, why an antibiotic prescription was not necessary, the expected duration of illness, and symptom management options.
• Survey: A survey was administered to practicing physicians and advance practice professionals (n = 38) to better understand why they prescribed antibiotics for viral illnesses. The survey listed 4 possible reasons for antibiotic prescriptions: fear of bad patient satisfaction score, patient perception of the value of clinic visit, treat superimposed secondary bacterial infections, and wait and fill if the symptoms worsen. According to survey results, the main reason (31%) why clinicians prescribed antibiotics for viral illnesses was fear of bad patient satisfaction. Following this finding, data analysis of clinicians’ antibiotic prescription rate and patient satisfaction rate was performed and shared, which uncovered no correlation between antibiotic prescription rate and patient satisfaction.

Outcomes and Impact. The cumulative baseline data for upper respiratory infection, pharyngitis with negative strep, acute bronchitis, and influenza indicated that immediate/urgent care centers at Northwestern Medicine were prescribing antibiotics at the rate of 34% in FY 2019. Following implementation of the interventions, the overall rate of antibiotic prescriptions was reduced to 21% at the end of FY 2020 and 12% at the end of FY 2021—a total relative reduction of 65%. The patient satisfaction rate of immediate/urgent care clinicians increased from 83% in 2019 to 89% in 2020 and stayed the same in 2021. During the early phase of the COVID-19 pandemic (April 2020 through June 2020), we observed an uptick in antibiotic prescribing. Nevertheless, antibiotic prescriptions returned to pre-COVID-19 levels after June and continued to decline. To sustain improvements and further reduce inappropriate use of antibiotics in immediate/urgent care centers, interventions surrounding clinician engagement, data sharing, and patient education will continue.

1. Centers for Disease Control and Prevention. Measuring Outpatient Antibiotic Prescribing. 2020. Accessed March 15, 2021. https://www.cdc.gov/antibiotic-use/community/programs-measurement/measuring-antibiotic-prescribing.html.

2. Palms DL, Hicks LA, Bartoces M, et al. Comparison of antibiotic prescribing in retail clinics, urgent care centers, emergency departments, and traditional ambulatory care settings in the United States. JAMA Intern Med. 2018;178:1267–1269.

Power of Prediction: Implementing a Predictive Model Workflow Drives Sepsis Care

Sheila Neiner, BSN, RN, CPHQ, Northwestern Medicine; Jennifer Altounian, BSN, RN-BC, Northwestern Lake Forest

Northwestern Medicine

Background. Early sepsis identification had decreased since going live with the Epic electronic medical record in March 2018 across Northwestern Medicine (NM) hospitals. Best practice advisory (BPA) alerting for potential sepsis used antiquated systemic inflammatory response syndrome (vital signs and white blood cells) criteria, leading to a 70% increase in alerts from the prior electronic medical record, subsequently decreasing registered nurse (RN) response when receiving an alert. In addition to the large increase in BPAs, accuracy of predicting sepsis dropped to only 20%. The BPA merely prompted the RN to complete a sepsis assessment without provider notification. No provider BPA was established and the process relied on RN notification to the provider. Given the low utilization of the BPA due to alert fatigue, timely assessment was lacking, resulting in delayed sepsis identification and treatment. There was a need to reduce alert fatigue and gain confidence in the sepsis BPA to improve early, goal-directed therapy for sepsis. It was identified that the systemic inflammatory response syndrome (SIRS)-based BPA was a burden for nursing, caused disruptions in care, and did not accurately identify septic patients. In July 2019, we set out to determine if Epic’s predictive model scoring is a valid tool to guide early intervention. A multidisciplinary team, including physicians, nurses, information services, and quality leaders, was convened to validate the predictive model scoring and redesign the workflow.

Intervention Detail. The multidisciplinary team first looked at all criteria that would be taken into account for the predictive score. The criteria used by the Epic model includes trends based on patient data to predict the likelihood of developing sepsis. Specifically, this data includes demographics, medications, labs, flowsheet documentation, lines, drains, airways, and comorbidities. In comparison, the SIRS criteria BPA only used vital signs and white blood cell values. Epic’s sepsis prediction model is an automated algorithm that runs in the background every 15 minutes. A score is computed that represents how likely the patient is to be septic or develop sepsis. Scores are used to drive decision support tools. With mapping confirmed, we then validated the build by logging prediction scores behind the scenes, collecting 6 months of data. Data analysis of the model’s performance showed that the false positive rate was reduced by 33%. With confidence in the model, we proceeded to design a BPA and workflow that would guide the clinicians. Score implementation needed to be done alongside a revamp of the sepsis workflow utilizing the Epic tools. Two pilot sites (Lake Forest and Kishwaukee Hospital emergency departments and intensive care/medical/maternal units) would design a BPA and workflow for both RNs and providers using the enhanced criteria prediction score, which would then trigger a BPA that guides clinicians in early sepsis treatment interventions. In the new workflow, the provider confirms the BPA, a time zero displays, and a storyboard is linked to a guide/task list to assist in sepsis interventions—also replacing redundant screening documentation that was required in the previous workflow. Providers and RNs received training on the new sepsis workflow and the importance of early intervention. The new BPA and workflow were then activated in real time, replacing the existing SIRS-based BPA.

Outcomes and Impact. The pilot sites went live in August 2020. Early analysis showed positive prediction for sepsis of 30% using threshold scores (= 5 for medical/surgical units, = 7 for intensive care units, and = 3 for maternal units). In comparison to the SIRS criteria, the predictive model was 10% more specific. With the redesigned workflow, the number of documentation clicks required by RNs was reduced by 90%. The tools displayed vital signs and lab values for sepsis risk automatically without needing to redocument. The BPA also displays important organ dysfunction criteria. The new tools are actionable and do not require leaving current documentation activity to take action on the BPA. Once sepsis has been confirmed within the BPA, information is placed on a storyboard to guide clinicians in sepsis treatment. A sepsis time zero tracker is also displayed to assist in 3- and 6-hour interventions for sepsis. Post-implementation data comparison of 1 month before go-live and 1 month after go-live showed that the BPA triggered to the RN 31% less. Looking at interventions, we saw an improvement in obtaining initial lactate of 29%, blood culture collection improved by 28%, and antibiotic administration improved by 19%. Initial 3-hour sepsis treatment has improved, which resulted in an overall decrease in the Vizient mortality index for septicemia for both pilot hospitals (NM Kishwaukee from 0.78 mortality index to 0.60 mortality index and NM Lake Forest from 0.62 mortality index to 0.46 mortality index). In a post-survey to nursing, over 80% felt the new workflow is easy to use to determine sepsis treatment, while 78% felt that the task list is useful in the management and treatment of sepsis.

Reducing Telemetry Overuse in a Nonintensive Care Unit

Jacquelyn Reisenhofer, RN, BSN, PCCN; Valerie Kaura, MSN; Cheary Shelim, RN, DNP

Redlands Community Hospital

Background. Current policy does not reflect the most recent American Hospital Association guidelines for telemetry, and checking the patient for necessary continued telemetry use is not routinely performed. Patients who are admitted with telemetry orders usually stay on telemetry until discharge, resulting in patient flow issues and increased costs. This was a nursing-driven project in which nursing identifies patients for whom it is appropriate to ask for a telemetry downgrade from their physician. Leadership facilitated a partnership between nursing, case managers, and physicians to develop criteria, which were then rolled out to telemetry staff. The customer’s voice is captured through increased independence (eg, ability to shower without an order, removal of bulky/heavy equipment that restricts patient experience and safety). Unnecessary testing is also prevented that could lead to overtreatment of the patient for benign rhythms, which can be harmful. Downgrading patients who no longer need telemetry monitoring is part of the 2021 Vizient Connections Summit because it decreases costs for the organization, improves patient experience, decreases errors in treatment, and ensures patients are at the right level of care—which improves patient flow.

Intervention Detail. We utilized the Plan-Do-Study-Act method to create a multidisciplinary team, including bedside nursing, nursing leadership, physicians, and case management, to develop criteria for patients who would qualify to be downgraded to medical/surgical. The pilot started in November 2017 on a 19-bed telemetry unit, then completed on 40-bed telemetry unit. Inclusion criteria included adult patients in 2100 (a 40-bed telemetry unit) at the time of the study with telemetry orders > 48 hours. Exclusion criteria: Adult patients in 2100 at the time of the study who: (1) do not have telemetry orders; (2) have telemetry orders < 48 hours; (3) have multiple transfers between levels of care; or (4) did not have a discharge date at the end of the study and used the last date of the study as their cutoff period. A 48-hour check table was created to help nursing and monitor techs identify patients who qualify for telemetry discontinuation. The check table included diagnoses, date of admission, current rhythm, whether the patient has had a rhythm change in the last 12 hours, and if so, what their change in rhythm was. Staff was educated that all adult patients on 2100 on telemetry monitoring for more than 48 hours with no recent (last 12 h) rhythm changes would be selected for the registered nurse (RN) to discuss with the physician the need for continued monitoring. Stroke was identified by cardiology as a diagnosis that would not discontinue monitoring due to post-stroke risk of arrhythmia. Monitor techs identified the patients on monitors > 48 hours, charge nurses reviewed this information with RNs, and RNs would have daily discussions with physicians on the need for continued monitoring. A literature review was conducted using multiple sources.^1–5

Outcomes and Impact. Data analysis included pre- and post-intervention of total telemetry patients during the study, the number of downgrades by day, and financial implications. Pre-intervention total patients during study: total telemetry patients 853, patients > 48 hours 395, discharges 89. Post-intervention total patients: total telemetry patients 996, patients > 48 hours 428. Pre-intervention number of downgrades by day: n = 118, 353 patient telemetry days total, 14 patients downgraded, 52 days saved by downgrade. Post-intervention: n = 428 patients, 111 patients downgraded. The normal histograms of the data show that the 2 distributions are positively skewed. They approximate a Poisson distribution, with high frequency at the beginning, dropping off to much less. There is a high, positive, and significant correlation between pre- and post-data, as statistics do not always capture the whole picture. Our current data analysis is completed and results demonstrate the following:

• Financial implications:

• –  Pre-data monthly cases 14, monthly patient day reduction 52, hour per patient day reduction 2.42, average hourly rate $35.92, monthly cost reduction $4520, annual cost reduction $54 242.
• –  Post-data: monthly cases 111, monthly patient day reduction 94, hour per patient day reduction 2.42, average hourly rate $43.38, monthly cost reduction $9868 annual cost reduction $118 417.

• Increase from 3% to 10% in projected annual cost savings.
• Projected annual reduction of 1154 patient days.
• Projected annual reduction of 2793 nursing hours per patient day.
• Unit processes now include a daily check for telemetry use in patients who have telemetry orders for > 48 hours.

1. Choosing Wisely. Society of Hospital Medicine – Adult Hospital Medicine: Five Things Physicians and Patients Should Question. 2013. Accessed November 14, 2017. http://www.choosingwisely.org/societies/society-of-hospital-medicine-adult/.

2. Sandau KE, Funk, M, Auerbach, A, et al. Update to practice standards for electrocardiographic monitoring in hospital settings: a scientific statement from the American Heart Association. Circulation. 2017;136:e273–e344.

3. American Association of Critical-Care Nurses. AACN practice alert: accurate dysrhythmia monitoring in adults. Crit Care Nurse. 2016;36:e26–e34.

4. Perrin K, Ernst N, Nelson T, et al. Effect of a nurse-managed telemetry discontinuation protocol on monitoring duration, alarm frequency, and adverse patient events. J Nurs Care Qual. 2017;32:126–133.

5. Bubb C. A timely practice: a nurse-driven telemetry discontinuation protocol. Pa Nurse. 2011;66:6–10.

To the Heart of Pain: A Multimodal Regimen in Cardiac Surgery

Britany Eichenauer, MSN

Saint Luke’s Hospital of Kansas City

Background. In 2017 the United States Department of Health & Human Services declared the opioid epidemic a public health emergency. This declaration, along with growing concern from public and regulatory bodies, prompted health care organizations to analyze their opioid prescribing practices and potential alternatives. In 2001, the first enhanced recovery after surgery (ERAS) protocol included an alternative approach; specifically, a multimodal pain management plan for colorectal surgery patients. The first cardiac surgery ERAS guidelines were published in 2019 with indications for a multimodal pain regimen (strong benefit with moderate quality evidence). The guidelines were intentionally vague based on the lack of studies on the multimodal regimen benefit.^1,2 Our project was designed to determine current utilization of opioid resources within the cardiac surgery domain by days within stay. The data showed there was an opportunity to improve our opioid management practices and implement a multimodal pain management strategy to reduce our opioid usage. To make the needed changes, a large multidisciplinary team was formed to review available literature, guide decision-making, and implement proposed changes.

Intervention Detail. Based on the new cardiac surgery ERAS guidelines, a baseline analysis of opioid resources was conducted. Utilizing the Vizient Clinical Data Base, the population was isolated by Medicare severity diagnosis-related group. The identified comparative group was the U.S. News & World Report’s “U.S. Best Heart Hospitals 2018.” Data were analyzed for the calendar year 2019. To establish a true comparison, case mix index and length of stay were examined. Data were further analyzed by reviewing the use of high-cost medications and the mean resource units per case of opioid agonists. Out of the 20 centers within the comparison, our organization was the 7th-highest opioid consumer. In further investigation, the data were disaggregated using the “days within stay” grouper. Data were separated by Medicare severity diagnosis-related group into 3 categories: coronary artery bypass graft (CABG) only, valve only, and all DRGs (216, 217, 218, 219, 220, 221, 231, 232, 233, 234, 235, 236) to determine if there was a subpopulation with higher opioid usage. This report showed that, by comparison, our organization performed better on postoperative day (POD) zero and in line with comparison hospitals for days 1 and 2. The true opportunity was the use of opioid medications during days 3 through 13, with a gradual increase of resources day over day. Through data review, it was determined that our organization had an opportunity to reduce opioid consumption in the postsurgical population phase. A multidisciplinary team was established to begin reviewing current available literature, our internal process, and identification of potential opportunities. Team members included pharmacy, leadership, anesthesia/critical care, electronic medical record analysts, nursing, physicians, surgeons, and advanced practice providers. Due to a lack of standardized dosing or medication regimens, a subgroup reviewed multiple literature sources to ensure our changes were safe and evidence-based. A pharmacy-led team specifically guided dosing of gabapentinoids. Based on the mixed published results of gabapentinoids being used for postoperative pain, the team decided to guide dosing based on age and renal function. The surgical team, including cardiothoracic surgeons, advanced practice providers, and nursing, focused on a review of the current order set. Multiple opportunities were identified, including scheduling Tylenol, removing intravenous opioids after POD 3, offering nonpharmacological intervention options, and strategically guiding age-based dosing.

Outcomes and Impact. Due to the size of the multidisciplinary team, the extensive literature review, and the COVID-19 pandemic, the order set change went live in August 2020. Through the review, high-cost medications, such as IV Tylenol, were determined not to have a substantial return on investment. The sole focus was on transitioning our medication orders to reflect a multimodal strategy. After go-live, project coordinators reviewed any needed revisions or questionable adverse effects related to the regimen (eg, prolonged ventilation times, confusion, somnolence). Based on bedside clinician feedback and potential correlated complications, no edits were required and no complications were matched to the new regimen. The data was first reviewed in January 2021 in 3 groups: CABG only, valve only, and combined CABG/valve DRGs. For all DRGs, our organization outperformed the comparative group on POD zero and equal to POD 1 and 2 in the initial report, but we still saw a reduction of 17% on POD zero, 34.5% on POD 1, and 12.5% on POD 2. Our target of PODs 3 through 10 saw a 33% reduction in opioid usage. Our overall post-implementation opioid usage by case showed a 32% reduction in opioid resources for the hospital stay. Case mix index appeared similar between comparative group and internally year over year, but length of stay was down by 1 day. Through the “days within stay” grouper, the CABG-only population saw a reduction in opioid resources of 30% on POD zero, 43% on POD 1, and 22% on POD 2. The valve subpopulation also saw improvement with a reduction on POD zero of 6%, 28% on POD 1, and 3% on POD 2. Based on our experience, other specialties within our organization are looking to incorporate our work into their opioid reduction strategies. They are also using the Vizient Clinical Data Base opioid resource utilization to analyze their baseline and post-implementation data. It is our hope that by sharing our experience as part of the 2021 Vizient Connections Summit, we can encourage others to examine their processes to reduce the opioid burden within cardiac surgery.

1. Brown J, Singh K, Dumitru R, et al. The benefits of enhanced recovery after surgery programs and their application in cardiothoracic surgery. Methodist Debakey Cardiovasc J. 2018;14:77–88.

2. Engleman DT, Ali WB, Williams JB, et al. Guidelines for perioperative care in cardiac surgery: enhanced recovery after surgery society recommendations. JAMA Surg. 2019;154:755–756.

Protecting High-Risk Patients From Unsafe Chemicals: IVF Bag Conversion

Colleen Groll, MSN, LEED AP BD&C, Seattle Children’s, and Mellissa Nguyen, Vizient

Seattle Children’s

Background. According to the Environmental Protection Agency, contamination from unsafe chemicals is pervasive in society. At particularly high risk to the negative impacts of unsafe chemical exposure are pregnant women, infants, and young children.¹ Prenatal and early life exposure to phthalates, for example, has been linked to asthma, allergies, and cognitive and neurodevelopmental problems such as hyperactivity, anxiety, depression, and aggression. Phthalates have also shown to disrupt reproductive development in boys. In addition to concerns of unsafe chemical exposure in high-risk patient populations, health care staff are becoming increasingly concerned about their risk of exposure to unsafe chemicals during their daily work. A national survey of nurses by the Environmental Working Group suggests there are links between chemical exposure at work and serious health problems such as cancer, asthma, miscarriages, and birth defects.² While the number of legislative policies being discussed at state and national levels is growing, many health care organizations are proactively creating their own policies to protect patients and staff from unsafe chemical exposure based on the “do no harm” bioethical principle of medicine. Health care organizations have found success through adoption of policies to screen unsafe chemicals in the supply chain as both patient and staff safety initiatives, as well as moral and ethical responsibilities. Once educated properly, both patients and staff see the effort to mitigate unsafe chemical exposure as a necessary and protective measure. Despite these efforts, however, multiple studies and research show that legislation and policies alone are having a minimal effect on preventing unsafe chemicals from polluting the bloodstreams of pregnant women, unborn fetuses, infants, young children, and adults. A recent study revealed that hazardous chemicals can be found in at least 114 different products being used in the average hospital pediatric care room.³

Intervention Detail. For many health care organizations, the challenge of screening and removing unsafe chemicals from the health care supply chain is multifold, including: (1) a lack of industry-wide standardization; (2) today’s methodology is highly manual, unsustainable, and not data-driven; (3) safer chemical information is rarely considered at the product and equipment levels; (4) chemical safety initiatives are often not framed in terms of negative health impact; and (5) return on investment is not clear. At Seattle Children’s, we: (1) acknowledged and amplified the concern of unsafe chemicals in the IV fluid (IVF) bag category; (2) increased transparency into the chemical composition of products being used in current IVF bags by requesting attribute information at the product level; and (3) established standardized, environmentally preferred attribute information that was matched to real-time hospital purchasing data to evaluate safer chemical usage in the supply chain at the organization, facility, department, product category, and individual product levels for the IVF bag conversion. For the conversion, several suppliers submitted information regarding the requested environmentally preferred attributes, pricing, and product supply reliability. Several Vizient suppliers were evaluated based on their submitted information. Value analysis included stakeholders in supply chain, strategic sourcing, pharmacy, anesthesia, and nursing. We determined that converting to B Braun, a Vizient-awarded supplier that produces polyvinyl chloride (PVC)- and phthalate-free IVF bags in continental United States manufacturing facilities, would save more than $200 000 annually.

Outcomes and Impact. An IVF bag conversion to B Braun is a great example in which the 3 P’s of sustainability has been achieved: people, planet, and prosperity. This prevents high-risk pediatric patients from being exposed to IVF bags that contain PVC and phthalates, thereby reducing potential long-term health effects from exposure to these chemicals. Product manufacturing within the continental United States reduces the carbon footprint for product shipping and reduced usage of these chemicals of concern allows for more environmentally sound product disposal. The conversion to B Braun saved more than $200 000 annually. At the time of abstract submission, the first phase of conversion was at the beginning stages. The expected outcome is completed conversion and staff training on handling of the new IVF bags that do not contain PVC and phthalates.

1. U.S. Environmental Protection Agency. Information About Chemical Hazards in Child Care Settings for Parents. 2021. Accessed March 11, 2021. https://www.epa.gov/childcare/information-about-chemical-hazards-child-care-settings-parents.

2. Environmental Working Group. Chemical Exposures on the Job May Be Linked to Diseases in Nurses [press release]. 2007. Accessed March 11, 2021. https://www.ewg.org/news/news-releases/2007/12/11/chemical-exposures-job-may-be-linked-diseases-nurses.

3. Rossi MS, Ratliff W. Chemical Footprint of Products Commonly Used in Pediatrics Departments. Clean Production Action. 2017. Accessed March 11, 2021. https://www.cleanproduction.org/resources/entry/report-chemical-footprint-products-in-pediatrics.

The Journey to Eliminating Hospital-Acquired Pneumonia

Tonya Meyer, RRT, CPHQ, CLSSGB; Regina Moore, RRT, CLSSGB

SoutheastHEALTH

Background. In 2018, Southeast Hospital experienced an $819 000 hospital-acquired conditions pay-for-performance penalty for FY 2019. This penalty was also impacted by a high pneumonia readmission rate of 18.9% (compared to the national average of 16.9%) and a high pneumonia mortality rate of 18% (compared to the national average of 15.9%).¹ Through strategic planning, the executive leadership team requested quality improvement efforts to focus on patient outcomes and safety—which in turn would positively impact hospital-acquired condition reduction efforts. The Sepsis Performance Improvement Team was assembled as part of the strategic plan to achieve the goals set by leadership. As the sepsis team reviewed patient cases and established goals, several sepsis cases were recognized as originating from infections that had been acquired during the hospitalization. It was also identified that Southeast Hospital experienced its worst performance for hospital-acquired pneumonias (HAP) in 3Q 2017, with 45 cases ending calendar year 2017 and 210 total cases in nonventilated patients—resulting in a potential cost to the organization of $6.3 million (cost estimate of $30 000 per case). Considering oral bacteria and microaspiration as contributors to pneumonia and their impact on patient safety, the team decided to focus on ensuring appropriate oral care was performed as a regular part of the patient’s activities of daily living. This project was supported by an education opportunity through the Missouri Hospital Association for Lean Six Sigma certification. Although oral hygiene may seem rudimentary, this project proves that small initiatives can have extremely positive impacts for patient safety and an organization as a whole.

Intervention Detail. The “define” phase of the project began in May 2018 and followed Lean Six Sigma principles. According to evidence-based literature and in relation to the replication of bacteria in the oral cavity, research indicated that providing oral care 4 times daily is most beneficial in pneumonia prevention.² A clearly defined process, output, defect definition and opportunity were defined and aligned closely with research resources. The intensive care unit and the cardiothoracic unit (CTU) were chosen for the pilot. During the “measure” phase of the project, a data collection plan was written, with opportunities defined for both ventilated and nonventilated patients. To ensure the validity of data collection an attribute measurement systems analysis was completed for repeatability and accuracy, with both abstractors achieving 100%. Daily audits were completed for 2 weeks during the “analyze” phase, revealing 34% compliance in oral care opportunities with a sigma score of 1.0 for baseline process capability. According to statistical significance using chi-square, with a P value of 0.0, the greatest opportunity was identified for both nonventilated and bilevel positive airway pressure populations in CTU. At this point in the project, it was discovered that there really was not a standard process for oral care within the organization. Therefore, at the start of the “improvement” phase, a supplier, inputs, process, outputs, and customer was completed with multidisciplinary team participation to determine the standard operating procedure. Following the supplier, inputs, process, outputs, and customer, clinical staff members were educated on the new process, which included the use of oral care kits instead of a toothbrush and toothpaste. In February 2019, when 110% improvement in oral care was achieved (73% compliance), the project moved into the “control” phase. Audits and reporting were assumed by clinical staff leaders from both the intensive care unit and CTU.

Outcomes and Impact. The official Lean Six Sigma project was completed in February 2019, but positive change continued through all nursing units, with onboarding of the last units in August 2019. Similar barriers were encountered on all units, some of which included staff buy-in as to the “why,” staff responsibility/accountability for patient populations, and supply needs and challenges. Although the project began with an inpatient focus, a much larger community need was discovered through encountered barriers with patient cooperation (attributed to social determinants and underfunded populations). The organization was supported through a grant opportunity to provide oral care kits to all patients discharged from the medical and surgical floors, preoperative, same-day surgery, and gastrointestinal lab. During the project period (May 2018 through August 2019), the following improvements were achieved: (1) a 54.83% improvement in HAPs in the first project’s first 6 months, with an estimated cost avoidance of $677 246 ($30 000 per case); (2) a 15.76% improvement for 30-day pneumonia readmissions in the 65 and over Medicare population; (3) a 34.16% improvement in pneumonia mortality; and (4) a 39.39% improvement in sepsis mortality. This project continues to positively impact patient safety by decreasing HAPs and pneumonia readmissions. In turn, the organization also experienced a decrease in pay-for-performance penalties, including: (1) a 69.23% improvement in HAPs from Q4 2019 to Q4 2020, with an estimated cost avoidance of $3.93 million (updated cost estimate of $39 838 per case) and (2) a 9.21% improvement in 30-day pneumonia readmissions (65 and over Medicare only) from Q4 2019 to Q4 2020. Providing improved oral care will be an ongoing effort, sustained through continued staff and patient education, weekly rounding for sepsis, and oral hygiene education, as well as by providing resources to assist patients with successful transition after discharge.

1. Centers for Medicare & Medicaid Services. Hospital Compare Report Run Date: October 3, 2017. https://www.medicare.gov/care-compare/.

2. Giuliano KK, Penoyer D, Middleton A, et al. Original research: oral care as prevention for nonventilator hospital-acquired pneumonia: a four-unit cluster randomized study. Am J Nurs. 2021;121:24–33.

Using QR Code Technology to Enhance the Patient Education Experience

Maribeth Cambridge, MS, RN, CCRN-K; Brian Lee, MSN, RN, CMSRN; Sharon Hampton, PhD

Stanford Health Care

Background. COVID-19 has limited in-person interactions between patients, families, and health care providers, highlighting the need to examine medical education materials to ensure they are easily accessible and appeal to multiple learning styles. Providing patient education while in the hospital can be challenging because patients are stressed, anxious, and in the process of recovering from surgery or illness. When education materials are consistent and available in multiple modalities in the inpatient and outpatient settings, patients reported that it made understanding complex information easier. QR code technology allowing videos to be embedded into standardized discharge instructions for the ear, nose, and throat patient population optimized patients’ comprehension and met the needs of both auditory and visual learners.

Intervention Detail. Nurses on an ear, nose, and throat unit at a large academic hospital enlisted the help of a patient advisory panel to understand how to meet learning needs. The feedback received from the panel validated the need to standardize education materials, appeal to multiple learning styles, and be easily accessible. We identified that existing educational videos were not being utilized by patients or registered nurses, as they were not easily accessible in the inpatient setting. Written handouts that provide step-by-step instructions of tracheostomy care, tube feeding, and medication side effects were updated with QR codes that link the patient to a video on the respective subject. registered nurses were educated on using hospital-issued smartphones and tablets to utilize QR codes and incorporate videos into their discharge teaching. Patients can use their personal mobile device to reference the videos in the hospital and at home.

Outcomes and Impact. At the time of this submission, 37 patients had been surveyed during their postoperative visit on the effectiveness of the standardized education materials that included QR codes. Patients provided comments that the videos were helpful and easy to access with the QR code. Additionally, there were no readmissions related to patient knowledge deficit on tracheostomy, tube feeding, or medication management. Nursing reported that the QR codes made it easier to incorporate videos into their discharge teaching, which led to an increase in patient engagement and willingness to learn how to manage complex needs at home. Other hospital departments have now expressed interest in using QR codes to improve their patient education experience.

In and Out: That’s What PCI’s All About

Aaron Sy, MS, Stanford Health Care; Purnima Krishna, MSN, MBA, RN, Stanford Health Care; Dominique Watt, MSN, RN, PCCN-K, CNL, Stanford Health Care; Yohan Lee, MBOE, Stanford Health Care; Amy C. Lu, MD, MPH, Stanford School of Medicine; Jennifer Tremmel, MD, MS, Stanford School of Medicine

Stanford Health Care

Background. In 2019, our institution enrolled in the Centers for Medicare & Medicaid Services Merit-Based Incentive Payment System. Merit-Based Incentive Payment System is a mandated quality payment model in which reimbursements are adjusted based on performance measured in 4 domains—quality, improvement activities, promoting interoperability, and cost. Elective percutaneous coronary intervention (PCI) was 1 of 8 cost measures used by Centers for Medicare & Medicaid Services to evaluate performance. According to Medicare’s risk-adjusted cost measure score, Medicare paid an average of 6% more per episode to our institution compared with the national average. Therefore, we wanted to focus on decreasing the cost of our elective, outpatient PCI cases. After meticulously analyzing the financial and clinical data, we found that most of our total direct cost was due to increasing recovery room costs and additional costs accrued during the overnight stay. Interventional cardiology and the operations team decided to concentrate their efforts on developing an evidence-based, consensus-driven, clinical pathway. While the interventional cardiology service had started same-day discharge (SDD) for PCI in the early 2010s, utilization among physicians was variable and system-based barriers existed—particularly with procedural scheduling and discharge planning. The high-impact opportunities were optimizing preoperative scheduling and the efficient use of recovery rooms. This abstract will allow others to learn how to implement similar principles and workflow for various cohorts to optimize use of hospital resources.

Intervention Detail. The clinical team first pinpointed the major logistical barriers of SDD for PCI (SDPCI). These included: (1) patients not having an adult caregiver at the time of discharge; (2) patients residing several hours away from the hospital; (3) patients’ desire to stay overnight; (4) procedures performed too late in the day to allow for SDD; and 5) patients’ pharmacies not having or insurances not covering the antiplatelet. Task forces were identified from each area—interventional cardiology, surgery schedulers, catheterization angiography pre- and post-intervention, and perianesthesia—to perform a root-cause analysis and help with designing the intervention. Literature reviews revealed that some organizations had successfully increased the rate of SDPCI by creating guidelines to identify SDPCI-eligible patients. We decided to expand on those criteria to develop a scheduling algorithm to schedule SDPCI-eligible patients earlier in the day, allowing recovery beds to be vacant for overnight patients later in the day. We then modified the nurses’ workflow to address the antiplatelet issues. Historically, the clinical team would call the patient’s pharmacy for availability of the antiplatelet near the time of discharge, but discharge would be delayed if the medication was unavailable. With the new workflow, nurses contacted the pharmacy at an earlier point; this verification process expedited discharge. Next, we created expectations for SDD throughout the patients’ care, and patients were encouraged to prepare living arrangements accordingly. After discharge, nurses would call SDPCI patients the next day before noon to follow up. The project’s next step is to develop a data dashboard containing key performance indicators such as SDPCI rate, readmission rate, and cost savings.

Outcomes and Impact. These interventions have been successful in significantly increasing our rate of SDPCI. Prior to the initiative, FY2018 and FY2019 had SDPCI rates of 32% and 39%, respectively. After implementation, the SDPCI rate increased to 66% in FY2020 and is currently at 80% for FY2021—a significant improvement compared to pre-implementation (P < 0.05). The annual volume of elective outpatient PCI cases has remained similar from FY2018 to FY2021, at about 180 elective outpatient PCI procedures each year, even during COVID-19. Although SDPCI rates have increased, there has been no significant change in post-procedure complications. The 30-day readmission rate and 30-day emergency department visit rate from FY2018 to FY2021 have remained the same. Concurrently, the average cost per case during FY2020 was 31% less for SDPCI than for overnight PCI, which translates to a substantial cost difference. In addition, Stanford Health Care experienced a decrease of 1.5% in the average total direct cost of care per elective outpatient PCI from FY2019 to FY2020. While this may seem like a small amount, it was accomplished during adjustments to our cost model and the opening of our new state-of-the-art hospital in November 2019. Moreover, by increasing the rate of SDPCI, more beds were available for other patients, which optimized our resources for other needs. This has been particularly invaluable during the COVID-19 pandemic. Not only were beds made available, but we had an infrastructure to support early discharge for patients so they could minimize their time in the hospital and feel safer recovering at home.

Optimizing Patient Access to Cancer Genetic Counseling Through Digitization

Courtney Rowe-Teeter, MS, CGC; Gayani Kadurugamuwa, MPH; Vikashini Savadamuthu, MPA; Kerry Kingham, MS, CGC; Sarah K. Garrigues, BA

Stanford Health Care

Background. Stanford Health Care aspires to provide expert care to all communities and includes patient access in its operational goals. The Stanford genetics team recognized that the lengthy 2.5-month wait for a new patient clinic appointment did not align with this mission. The involuntary shift to 100% video visits during the COVID-19 pandemic served as an opportunity to assess the effect of digital health on patient access, evaluate its long-term staying power, and pioneer telehealth at the organization. The interdisciplinary team’s project was selected for Stanford Health Care’s 5-month internal process improvement program for clinicians, nonclinicians, and other staff. The team met weekly and applied methodologies taught through the course. Under the supervision of a process improvement coach, the group completed assignments and tested the impact of potential interventions against its goal of reducing the average number of days from referral to completed new patient visit by 25%. Project progress and findings were presented to sponsors, including clinical and administrative cancer care leaders. Using Lean methodology, the team identified key drivers that were necessary to attain success: (1) provider availability across sites; (2) streamlined patient scheduling and family health questionnaire collection; and (3) ensuring clinic patients were seen in a timely fashion and that any upfront, necessary paperwork and requirements were clear, straightforward, and did not prove burdensome. These key drivers were at the forefront when designing interventions. We continue to reference these principles when iterating processes to further improve patient access. For example, the group plans to survey patients for their understanding and clarity of pre-visit expectations. This presentation is an important part of the 2021 Vizient Connections Summit because the COVID-19 pandemic has catapulted the industry toward digital health adoption. Given this recent transition, many may not fully understand what reinforcing processes are required to ensure successful virtual health uptake. This project can serve as a guidepost for other organizations to understand what it takes to successfully hardwire digital health and use its advantages to enhance patient access.

Intervention Detail. While we knew there was significant delay from date of referral to when the patient was seen for a genetics visit, we had not previously quantified it. The specific report needed was not available in Epic. Therefore, we created an Epic report, pulling all patients seen for genetics visits and comparing it to a separate, inbound referral report to determine days from referral to date patient seen. Our baseline average from December 2, 2019 to March 16, 2020 was 51 days—a nearly 3-month average wait. We performed a root cause analysis to understand the reasons for the delay by interviewing our new patient coordinators (NPCs). The following reasons contributed to delays: (1) patients prefer to be seen closer to home, but there may be a longer wait at their local clinic site; (2) it was challenging for NPCs to ensure they were offering the true “next available” slot, as it was time-consuming to toggle between 4 different clinic-scheduling templates; and (3) patient delays in returning the required family health questionnaire. We focused on key drivers 1 and 2 for this project. Cancer genetics visits are ideal for video, given that a physical exam is not required. Yet, according to the 2020 National Society of Genetics Counselors’ Professional Status Survey, only 27% of genetic counselors providing in-person visits were also providing video visits.¹ Our team was already conducting some visits via video and was aiming to increase this percentage when the COVID-19 pandemic involuntarily required us to move to 100% video visits on March 13, 2020. This switch allowed patients to be seen by the next available provider(s), regardless of patient and/or provider location. We also implemented a systems-level intervention in Epic, allowing NPCs to see all providers on 1 scheduling template, ensuring they offered the next available appointment.

Outcomes and Impact. By transitioning to 100% video visits and due to Epic template optimization, we were able to offer the first available video visit to patients regardless of patient and/or provider location. Overall, we reduced the average time from referral to new patient visit from 51 days in March 2020 to 30 days in November 2020—a 41% reduction. We concurrently tracked weekly inbound referrals, which remained steady, thus confirming the increased accessibility was not due to decreased inbound referrals. Cancer genetics risk assessment has become an integral part of many oncology patients’ care, with genetic testing results often influencing surgical and/or chemotherapeutic management. By improving accessibility to our services, patients were able to receive timely, high-quality genetic risk assessment and genetic test results. While many in-person oncology visits still occur, patients appreciated the genetics visit being virtual, as it eliminated travel and allowed the comfort of taking the visit from home. Our operational change also improved clinician satisfaction by ensuring equity in patient volumes across all providers. We also saw a drop in average no shows per week from 4 to 2, leading to a gain of productive clinician time. We continue to track data in our run chart tool to monitor and quantify changes in wait time for new patient appointments. This information will help inform staffing needs and determine the appropriate percentage of visits to remain via video once it is safe to return to an in-person model. We are also excited to launch a virtual family health information collection platform, which sends automatic patient reminders to complete forms and provides a streamlined process to return the forms directly to clinicians, lessening the burden on the NPCs. We anticipate this platform will further contribute to decreased days from referral to visit and further strengthen the efficacy of video visits.

1. National Society of Genetic Counselors. Professional Status Survey 2020. Accessed August 2020. https://www.nsgc.org/Policy-Research-and-Publications/Professional-Status-Survey.

Rapid Implementation: Same-Day Discharge Pathway for Orthopedic Joint Patients During the COVID-19 Pandemic

Amy C. Lu, MD, MPH, Stanford School of Medicine; Alicia Wilson, BSN, Stanford Health Care; James Huddleston, MD, Stanford Health Care; Yohan Lee, MBOE, Stanford Health Care; Saud Khan, MBBS, MHA, Stanford Health Care; Dominique Watt, MSN, RN, PCCN-K, CNL, Stanford Health Care; Bridget Bergstrom, RN, Stanford Health Care; Joe Schomburg, MBA, MSN, RN, Stanford Health Care; Eric Bernier, RN, MSN, Stanford Health Care; Britomar Gomez, BSN, RN, Stanford Health Care; Stella Ng, BS, Stanford Health Care

Stanford School of Medicine

Stanford Health Care

Background. In recent years, evidence has suggested that patients undergoing a total hip arthroplasty (THA) or total knee arthroplasty (TKA) can safely be discharged the same day of procedure. Typically, these initial studies were done in selected patient populations with dedicated resources and time for optimal pathway implementation. We designed a pathway for rapid implementation, scaling, and iteration for same-day discharge of patients undergoing THA or TKA within a several-day time frame. The time constraints were due to a surge in COVID-19 cases in mid-December 2020, with a need to assist with safe patient throughput while continuing to provide high-quality care and ensure patient satisfaction. This was accomplished by setting preoperative expectations, gaining stakeholder and care team support around a novel discharge process, and daily iterations to troubleshoot workflow challenges. Additionally, the pathway was successful because we used recently developed, evidence-based THA and TKA short-stay pathways as a template for our same-day discharge pathway. We were able to utilize partnerships with multidisciplinary teams, including the orthopedic surgery service line, perianesthesia, and the pathways teams, with rehabilitation and care coordination available to troubleshoot any existing challenges in real time. Clinical and quality process, outcome, and balancing metrics were measured to monitor for any adverse events. Given the unprecedented and unpredictable nature of the COVID-19 pandemic, it is imperative that health systems address innovative solutions to rapidly scale and implement care redesign efforts. Our proposal describes a novel and feasible approach to these challenges and is something the 2021 Vizient Connections Summit audience will find valuable as an example of a continued care redesign innovation.

Intervention Detail. The interprofessional stakeholder team used a multipronged approach to assess suitable patients for the pathway across the continuum of care. A critical component to the pathway’s success relied on appropriate preoperative patient selection, based on criteria set forth by the orthopedic service line and the hospital pathways team. The inclusion and exclusion criteria addressed both the clinical and psychosocial support requirements of patients. The next phase included incorporating the existing preoperative and intraoperative components of the existing total joints pathway into the new same-day discharge workflow. The multidisciplinary work group selected a target measure of increasing the percentage of same day discharges (SDDs), in addition to quality and clinical metrics like length of stay, readmissions, and postoperative complications. Patient satisfaction was measured qualitatively through postoperative phone calls on postdischarge days 1 and 2. Key interventions implemented during this time included: (1) standardizing workflows to reduce variation in care and provide best practices; (2) providing prescriptions to patients prior to surgery; and (3) conducting daily huddles between multidisciplinary team members to review patient cases and plan accordingly, which allowed for conversations between nursing, residents, advanced practice providers, anesthesia, and surgeons. This facilitated identification and tracking of barriers and issues such as appropriate staff resourcing and preparations for various patient language needs. Additionally, a post-anesthesia care unit (PACU) discharge checklist specific to same-day discharge pathways was created. The checklist facilitated coordination with case management, rehabilitation services, and PACU nursing, improving communications between the outpatient and inpatient settings. Subsequent workflow changes included ensuring durable medical equipment was stocked in the PACU or delivered to the patient’s home prior to discharge. Patient-facing education materials were developed to promote safety and set patient expectations from the preoperative to postoperative setting. These included details on pain management, medications, and mobility.

Outcomes and Impact. In less than 2 weeks and with a multidisciplinary team effort, we were able to shift over 60% of the patient volume to SDDs. Our outcomes were as follows: (1) SDDs during pre-implementation (November 22, 2020 through December 13, 2020), SDDs were 1.9% of the patients undergoing THA or TKA (SDD N = 1, Total N = 54) and (2) SDDs during post-implementation (December 14, 2020 through January 21, 2021) were 64.2% (SDD N = 45, Total N= 70). During this time, 1 readmission was noted from this cohort, with the patient being discharged after a 1-day hospital stay. No other severe adverse events were noted. Qualitative patient experience reports through postoperative phone calls revealed extremely high patient and caregiver satisfaction scores. Pathway challenges unique to the COVID-19 pandemic included the inability for the patient’s family or caregiver to be present during postoperative recovery. Given the decrease in elective cases during the surge, our sample size was smaller than our typical surgical volume. The rapid scaling and implementation translated to manual data capture of clinical and quality metrics, including adverse events. Future improvements include capturing patient satisfaction data, digitizing patient-facing education, dashboard development to measure ongoing progress, and an electronic flag for automatic patient cohort identification on the perioperative status board. The unprecedented events of the COVID-19 pandemic presented a golden opportunity to rapidly develop a safe and effective SDD pathway for our highest-volume surgical patient population. This was beneficial for patient outcomes, patient and clinical team satisfaction, and hospital throughput and efficiency. Future efforts include continuation and expansion of these SDD pathways to additional surgical procedures, such as gynecology, bariatric surgeries, and interventional cardiology procedures.

A Team-Based Approach to Improving Oncology Mortality

Kimberly O’Neill, RN, BSN, CCDS, Stony Brook Medicine; Kelly James-Walsh, BS, Stony Brook Medicine; Mark J. Sands, MD, MBA, FACR, Stony Brook Medicine; Edward Sun, MD, MBA, FASGE, Stony Brook University Hospital

Stony Brook Medicine

Background. Stony Brook University Hospital (SBUH) is a 624-bed tertiary care hospital serving Long Island, New York. SBUH began a clinical documentation integrity (CDI) program in 2012 with 4 CDI specialists. By 2018, the program expanded to 16 specialists who performed real-time review of inpatient discharges with diagnosis-related group (DRG) risk model analysis to ensure appropriate and accurate risk adjustment. Additionally, all mortalities were reviewed to identify opportunities for documentation improvement. Despite these efforts, CDI specialists could not review all inpatient discharges or mortalities given the high patient volume. In January 2020, our chief medical officer convened a multidisciplinary mortality work group that included the director of clinical documentation integrity, liaisons from our hospital quality measurement and analytics division, and physician champions representing each of the 13 service lines comprising the Vizient Quality and Accountability Study mortality domain. This work group met every 6 to 8 weeks moving forward to review updated data and discuss new initiatives and progress with ongoing projects. It was noted that the Vizient-defined service lines are based on DRG and that the degree to which the Vizient-defined service lines correspond with the treating services at SBUH is variable. The data analysis performed by this group and the educational efforts that resulted from this work directly contributed to an improvement in our mortality index for the oncology service line from 2020 Q3 to 2020 Q4. These improvements continued through 2021 Q1 and 2021 Q2. Our team-based approach to improving mortality outcomes could be of great value to other 2021 Vizient Connections Summit attendees looking to build a successful mortality work group.

Intervention Detail. Third-quarter 2020 mortality data from the Vizient Clinical Data Base was analyzed by our mortality work group. The mortality observed-to-expected (O:E) for the oncology service line in 2020 Q3 was 1.5, with 19 deaths and 367 total cases. Detailed examination of the mortalities revealed that 17 were primarily cared for by internal medicine physicians, while 2 of the mortalities were attributed to surgical services, based on the Medicare severity-DRG. The expected risk of mortality for the medicine patients was 13.3, while that of the surgery patients was 1.9. As such, an opportunity was identified to provide focused documentation education to surgical services, caring for patients with oncologic conditions. Increased efforts were made with the CDI specialist assigned to the oncology service to educate and query all oncology providers, including surgical clinicians. CDI specialists assigned to surgical services were also trained to focus on surgical patients with cancer diagnoses. These CDI specialists rounded with providers, attended virtual oncology meetings, and communicated with providers via email and phone calls to discuss patient scenarios and obtain clarification of documentation. CDI tip cards were created based on Vizient risk model groups, with a list of the most common diagnoses (such as hypoalbuminemia, cachexia, and neoplastic pain) relevant to patients with cancer. Targeted documentation tips were emailed to both individual physicians and services. Documentation tip cards were laminated and posted in clinician common areas. The mortality work group physician champions, including a medical oncology physician and a surgical oncology physician, served as education leaders and resources for their colleagues. Particular attention was paid to educating clinicians regarding documentation of these diagnoses when present on admission, given the impact of present on admission on risk adjustment, coding of complications, and reimbursement.

Outcomes and Impact. As a result of the analysis performed by the mortality work group and the subsequent efforts of our CDI specialists and oncology physician champions, the oncology mortality index improved from 1.24 (2020 Q3) to 1.07 (2020 Q4). Even more telling, this mortality O:E improved despite the absolute number of deaths increasing from 19 to 23. The overall case mix index for the oncology service line increased from 2.23 (2020 Q3) to 2.37 (2020 Q4). Stratified by medicine and surgery, 16 oncology mortalities belonged to medical services and 7 oncology deaths belonged to surgery in 2020 Q4. The expected risk of mortality for the medicine mortalities increased from 13.3 in 2020 Q3 to 16.26 in 2020 Q4. The expected risk of mortality for the surgery mortalities increased from 1.9 in 2020 Q3 to 5.19 in 2020 Q4. The oncology mortality O:E continued to improve from 1.07 in 2020 Q4 to 0.97 in 2021 Q1, and even further to 0.68 in 2021 Q2, reflecting a persistent positive effect from our interventions in 2020 Q3. Using the Vizient Quality and Accountability Study to identify areas of opportunity and then culling data from the Vizient Clinical Data Base allowed our team to focus our efforts and devote resources to improving the oncology mortality index through targeted education and focused CDI interventions. Clinicians have approached our CDI specialists asking for more tip cards. As clinicians become more receptive to queries, the number of queries is anticipated to decrease in number, and CDI workflows promise to become more efficient and streamlined, leading to more appropriate hospital reimbursement. Future potential efforts include initiatives that integrate natural language processing technology to perform automatic chart reviews and real-time risk-adjustment on discharges.

Preventing Clostridioides Difficile Infection: Implementation of Two-Step Testing

Paul Murphy, BS, CSSBB; Sadia Abbasi, MD

Stony Brook Medicine

Background. Stony Brook University Hospital (SBUH) is a 624-bed academic medical center providing tertiary care for the Long Island, NY region. Hospital leadership identified an opportunity to develop and implement prevention strategies to reduce hospital-onset Clostridioides difficile infections (CDIs). Hospital-onset CDIs (HO-CDIs) are included in National Healthcare Safety Network (NHSN) reporting and are tracked by the New York State Department of Health as part of our mandatory reporting. Ultimately, this metric is included with other hospital-acquired infections as publicly reported data and various national hospital-rating scorecards. CDIs contribute to prolonged hospital stays, inappropriate antimicrobial use, increased readmissions, and high costs for any institution. A quality improvement team focused on decreasing HO-CDIs in early 2016 and included a multidisciplinary team with representatives from executive leadership, health care epidemiology, nursing, quality improvement, hospitalists, infectious disease, and laboratory staff, among others. The CDI prevention team functions as part of the overall hospital quality strategy, aligned with and having reporting responsibilities to a Patient Safety Steering Committee. Hospital-onset CDI rates improved through 2019 with implementation of information technology alerts and laboratory testing criteria, standardized annual education materials, antibiotic stewardship efforts, and improved environmental cleaning. However, implementation of a 2-step laboratory test in January 2020 improved our HO-CDI rates to historic lows. This testing algorithm utilizes the traditional polymerase chain reaction (PCR) test paired with an enzyme immunoassay (EIA) test to further clarify all inpatient and emergency department PCR+ results. The resulting EIA test serves to indicate an active infection or a potential colonization, helping to clarify clinical treatment plans and patient care surrounding C. difficile.

Intervention Detail. Following initial implementations and standardization of adherence to Centers for Disease Control and Prevention testing guidelines regarding laxative avoidance and symptom definitions, antibiotic stewardship, and environmental services utilization of the Bioquell machine for room cleaning, further work on testing modalities was necessary. In January 2020, SBUH began utilizing a 2-step testing algorithm, including a PCR and EIA for all inpatient and emergency department locations. The PCR test is a highly sensitive and specific test for toxigenic genes. It cannot distinguish between active infection and asymptomatic carriage. The EIA test has a low sensitivity, but high specificity for toxin production. It helps to determine an active infection or asymptomatic carriage. Utilizing both PCR and EIA testing helps clinicians determine an appropriate treatment plan based on active infections or asymptomatic carriage. The SBUH laboratory prepared the set-up with testing algorithms and computerized physician order entry functionality with the assistance of the information technology department. The group prepared evidence to support the safe and cost-efficient implementation for approval from the medical board in late 2019. As built, the alert ensures a single C. difficile test is ordered with appropriate clinical suspicion and within Centers for Disease Control and Prevention testing guidelines. Should that test result be PCR+, the same sample is automatically reflexed to an EIA test shortly after the initial PCR result. The primary PCR+ result will be present in the chart with language to identify the forthcoming EIA result as a distinct, second, pending result. Each laboratory result contains language as to the specificity and sensitivity of the test. As the EIA results after the PCR, this becomes the final result for reporting purposes. All patients with PCR+ results are monitored and evaluated for opportunities in antibiotic stewardship and clinical care with the understanding that treatment is a clinical decision guided by the laboratory results.

Outcomes and Impact. SBUH implemented the 2-step testing in January 2020 with the surge of a national pandemic shortly thereafter. National agencies suspended reporting of many quality metrics, including the NHSN. However, SBUH maintained tracking of C. difficile results and outcomes. Through December 2020, the hospital HO-CDI rate decreased by 77.6% from 2019 rates. Overall HO-CDI volumes decreased from 125 in 2019 to only 28 cases in 2020. NHSN prepared a standardized infection ratio (SIR) to evaluate infections with a risk calculation for expectation. Despite being waived in Q1 2020 and Q2 2020, SBUH reported a dramatic SIR improvement, with Q3 2020 and Q4 2020 near 0.35, improved from an annual SIR of 0.914 in 2019.

The Skinny on Reducing Bariatric Surgery LOS

Grace Parrish, BSN, RN, Tampa General Hospital; Ashley Mooney, MD, University of South Florida/Tampa General Hospital; Christopher DuCoin, MD, University of South Florida

Tampa General Hospital

Background. The Bariatric Program at Tampa General Hospital continuously strives to deliver exceptional care to the bariatric population. In the pursuit of excellence, our leadership, performance improvement specialists, and surgeons collaborated to identify opportunities for improvement utilizing the Vizient Clinical Data Base. The initial focus was to benchmark quality outcomes against other Comprehensive Academic Medical Centers using procedures for obesity Medicare severity diagnosis-related groups. Recognizing that bariatric patients are a high-risk population, we found it beneficial to focus on the Vizient Length of Stay (LOS) Risk Model. As the comparison data was analyzed, the opportunity to reduce LOS was identified. Until this point, our process was to perform upper gastrointestinal series (UGI) post-procedure to identify potential staple line leaks. This was identified as a large contributor to this population’s existing LOS. In researching evolving best practices, we identified that routine UGI series are no longer performed due to the limited sensitivity in distinguishing leaks. Instead, they are performed on patients exhibiting signs of potential leaks. Once the outcome opportunities and best practice recommendations were identified, our bariatric surgeons eliminated routine UGI testing. In addition, they began utilizing a bougie intraoperatively to confirm staple line integrity and ensure a negative leak test. Instituting this process change yielded a reduction in LOS. With the support of leadership, the surgical team, and performance improvement specialists, our facility established the new method and continued to track data to ensure that the changes would effectively and safely decrease LOS without increasing readmissions, emergency room visits, or observation (excess days). This project is important to the 2021 Vizient Connections Summit because analyzing and comparing data systematically is the key to identifying areas of improvement and implementing changes that generate sustainable results.

Intervention Detail. In June 2019, Tampa General Hospital collaborated with the University of South Florida in an initiative to reduce LOS in the bariatric population. The surgical team, performance improvement specialists, and leadership created a partnership to identify opportunities to reduce LOS using the Vizient Clinical Data Base. To break down barriers, hospital leadership provided support and guidance throughout the project. The key was to systematically analyze our data and benchmark our outcomes against other comprehensive academic medical centers. Once the opportunity was recognized, research in best practice guidelines was initiated, allowing us to identify that routine, postoperative UGIs were unwarranted. The change in standard was implemented and we also instituted use of a bougie intraoperatively to confirm staple line integrity and ensure a negative leak test. After analyzing our data, we began the test of change. By September 2019, we instituted this process change with all elective gastric bypass and gastric sleeve procedures. We collected data monthly and compared the outcomes to prior months and other academic medical centers. Progress was tracked by the performance improvement team and shared with the surgical team and leadership. In addition, to ensure that the institution of the new practice did not affect other measures, we defined readmissions and excess days as our balancing measures. We then tracked and compared those measures against prior months and other comprehensive academic medical centers.

Outcomes and Impact. At Tampa General Hospital, the bariatric population witnessed a significant reduction in observed LOS and LOS index. By December 2020, the average observed LOS dropped by 43% (2.40 to 1.37) and the LOS index decreased by 40% (1.36 to 0.81). The improvement brought our institution down to the 20th percentile among the comprehensive academic medical center cohort. In addition, our readmissions and excess days were not impacted by the change in practice. In summary, our patients were able to return to their families and their environment swiftly and safely, increasing patient and provider satisfaction. While internal data can suggest success or deficiencies locally, competitive benchmarking is an essential tool in empowering organizations to improve, increase patient safety and satisfaction, and set industry standards.

Solving Emergency Department Left Without Being Seen

Rawle A. Seupaul, MD; Carly Eastin, MD; Travis Eastin, MD; Randy Maddox, MD; Crystal Sparks, MSAM

UAMS

Background. Emergency department (ED) left without being seen (LWBS) is an important measure of quality, safety, and efficiency. Prior investigations have noted that up to 11% of these patients require hospitalization within 7 days, including emergency surgery, underscoring the significant risk of morbidity and mortality associated with LWBS.¹ Furthermore, the inability to assess and treat this population represents a significant financial liability compounded by diminished confidence among the community of patients served. The Centers for Medicare & Medicaid Services’ LWBS statistics are now available on its hospital compare website in publicly accessible domains, allowing consumers to shop before they buy services at a particular health care facility. Our patients’ complaints about their ED experience were overwhelmingly weighted toward prolonged wait times to see a physician and the decision to leave before a physician evaluation occurred. With the support of hospital and service line leadership, a pilot of placing a physician in ED triage was implemented. Based on the pilot’s success, the health system invested in physician, nursing, and other ancillary staff resources to expand the model to 12 hours a day, Monday through Friday. Our volume of patient complaints relevant to LWBS noticeably decreased, while the quality and quantity of ED patient assessments improved. Addressing LWBS in emergency medicine will be an important aspect of quality and safety improvement in the ensuing years. Models such as ours will be increasingly attractive because they do not require exorbitant construction costs for ED expansion. Sharing our successes and failures with this model could help Vizient members develop strategies that will work in their health care environment to efficiently reduce the harms associated with emergency department LWBS.

Intervention Detail. Our LWBS rate (6.5%) was among the worst in the country (> 75^th percentile), based on the 2019 Academic Association of Chairs in Emergency Medicine’s 2019 administrative database, where the median was 2.6%.² Because of this unacceptably high rate, we reviewed the medical literature to determine if there were existing models that could work in our setting. Based on that review and personal communications with numerous academic emergency medicine leaders across the country, we developed a physician-in-triage (PIT) model. Our PIT model was implemented Monday through Friday for 12 hours per day starting in June 2019. To properly staff this model, additional physician, nurse, and ancillary staff resource positions were required. Attending physicians evaluated each patient as they presented to triage, ordered initial tests, and wrote brief notes. Patient demographics, LWBS rates, elopement rates, left-without-treatment rates (defined as LWBS + elopement), and door-to-physician times were recorded on all ED patients as part of routine measures of PIT model performance for two 9-month periods: before PIT and after 12-hour PIT implementation. The chart of every eloped patient was reviewed within 24 hours to address any clinically significant concerns. To eliminate potential effects of reduced patient volumes during the COVID-19 pandemic, we analyzed data through February 2020. Fisher exact test was used for categorical outcomes and Student t test for continuous variables. Patient satisfaction data were reported as the National Research Corporation net promoter score for “would recommend facility.” Average collections for ED patients either admitted to the hospital or discharged home were $12 250 and $310.32, respectively.

Outcomes and Impact. Overall, 81 853 patient visits were analyzed: 40 520 patients before PIT and 41 333 patients in the 12-hour PIT group. Breakdown by sex and emergency severity index were similar in each group. The LWBS rate decreased from 6.5% before PIT to 2.6% after 12-hour PIT (P < 0.005). The elopement rate increased from 0.55% before PIT to 3.31% after implementation of 12-hour PIT (P < 0.005). Despite the elopement increase, the proportion of patients left-without-treatment rates decreased from 7.1% to 5.9% (mean difference, –1.1%; 95% confidence interval, –1.5 to –0.7), indicating that more patients finished their ED encounter. Of the 233 eloped patients during PIT hours, 20 (8.5%) required hospital admission within 72 hours, including 1 patient requiring urgent surgery for an ectopic pregnancy. Mean door-to-physician time decreased from 55 to 42 minutes. Our National Research Corporation net promotor score increased from 26.5 to 47. ED admissions and patients seen and discharged increased by an average of 59.5/mo and 185.6/mo, respectively, resulting in a net increase in collections of $9 437 632.

1. Baker DW, Stevens CD, Brook RH. Patients who leave a public hospital emergency department without being seen by a physician. causes and consequences. JAMA. 1991;266:1085–1090.

2. Scheulin J, et al. 2019 Academic Association of Chairs in Emergency Medicine Benchmark Survey. Accessed November 2021. https://www.saem.org/about-saem/academies-interest-group-affiliate/aaaem/benchmark-survey2/benchmark-survey/ed-benchmarking-survey-summary-results.

Three People Walk Into a Clinic: The Patient, the Provider, and the Scheduler

Michelle Havinga, MLS(ASCP)CM, LBBP; Alicia Messer; Ashley Balsanek

University of Iowa Health Care

Background. Statement of problem: Low adherence of annual primary care provider (PCP) visits has the potential to lead to a decrease in opportunities for preventive care gap closure. In value-based contracts it is essential to build and maintain the patient-PCP relationship that connects patient to care. Unique challenges encountered: (1) attribution methodology allows criteria to draw from urgent care/quick cares, which can designate patients to our facility without any previous visits; (2) potential access challenges can drive patients to utilize an urgent care/quick care instead of aligning patients to a PCP to start to build the relationship; and (3) this project was performed during the COVID-19 pandemic. Project objectives include: (1) schedule patients for their annual visits to strengthen the patient-PCP relationship; and (2) utilize validated data to track and trend quantitative progress toward goal metrics, such as increase patient adherence to annual physician visits. Key messages: (1) maintaining consistent communication with patients through various platforms is essential to the success of this project. We needed to meet patients where they were in terms of communication, which included telephone calls and MyChart messages and scheduling functionality; and (2) increasing patient adherence to annual visits to help grow the patient-PCP relationship and ensure review of preventive care gaps by our teams, which can lead to better patient health outcomes.

Intervention Detail. The objective of our work group was to improve patient adherence rates of annual PCP visits. The team started by analyzing data to identify quantitative measures, reviewing the potential opportunity for patients needing visits, deciding the call list criteria, and developing scripting for approval by clinic medical directors. Challenges and solutions included:

• 1.  Deciding how to best communicate with our patient population:

• Leveraged tools such as MyChart via Epic communication and direct outbound phone calls.
• Patients can schedule annual exams directly in MyChart at their convenience.

• 2.  Language barriers—when making direct phone calls, patients could forego an annual visit or treatment because of language barriers:

• Interpreters can partner with our accountable care organization support services staff to schedule patients with providers who speak the same language and/or ensure that interpreters are available at the time of their appointment.

• 3.  Patients from rural locations—rural area patients commute long distances to see their PCP, which could require a day off work or special transportation needs (drivers).

• Video visits have helped provide a solution to get rural patients seen—meet the patient where they are.
• Arrange annual visit and preventive care appointments the same day to decrease the amount of travel a patient needs to do for care (eg, scheduling annual visit and mammogram on the same day).

For the following quantitative measures, we utilized a population health dashboard developed by the population health team. This dashboard lives within the Epic platform and allows for real-time review of both annual visit rates and quality measure performance.

Quantitative measures monitored for success include:

• PCP visits
• Well-child visits (infants)
• Well-child visits (3–6 y old)
• Success rate of outbound calls

Outcomes and Impact. This project was deemed a success and we are evaluating other workflows that can be developed to leverage the same methodology to other populations needing regular visits, such as chronic care patients. Qualitatively, patients have expressed their gratitude for us caring about their health and for contacting them, especially during the COVID-19 pandemic. This has been an opportunity to create greater relationships between the patient, provider, and scheduler. Quantitatively, we have seen our chosen visit measures for success improve by margins of 14% to 22%. During a year of COVID-19, we felt like these results were quite remarkable due to the fact that some patients wanted to avoid public spaces.

• 1.  Visit measures:

• PCP visit: 17%
• Well-child visit (infants): 14%
• Well-child visit (3–6 y old): 22%

• 2.  Core measures:

• Breast cancer screening: –2%
• Colorectal cancer screening: 6%
• Pneumococcal vaccine: 12%
• Future fall risk screening: 1%
• Controlling high blood pressure: –1%
• Hemoglobin A1C under control: 1%
• Diabetes eye exam: –5%
• Tobacco use: screening and cessation: 1%
• Influenza immunization: 14%
• Statin therapy for cardiovascular disease: 12%

• 3.  The success rate of outbound calls is approximately 35.3%, which is equivalent to 2351 patients scheduled as a result of this project in calendar year 2020, which was during the COVID-19 pandemic.

Rapid Development of Support for Physicians and Providers As Response to COVID-19 Pandemic

Becky Lowry, MD, The University of Kansas Health System; Terry Tsue, MD, FACS, The University of Kansas Health System; Talal W. Khan, MD MBA, University of Kansas Medical Center

The University of Kansas Health System

Background. International dialogue on physician well-being was concerning prior to the onset of a global pandemic. Alarming rates of physician distress and suicide over the last decade led to exploration of system and individual drivers and analysis of associated emotional and economic costs for individuals, patients, and health systems. Health care professionals are struggling with the paradox of providing care for increasingly complex patients, while navigating burdensome electronic medical record systems and clinical practice environments increasingly focused on productivity and cost reduction. Academic medical centers face unique challenges around understanding and addressing physician well-being and support. Literature has clearly identified that physician well-being is a shared responsibility of both individuals and system leaders. The SARS-CoV-2 (COVID-19) pandemic introduced new and unprecedented strain on health care workers. Early pandemic international literature confirmed significant psychosocial burden on medical professionals. Awareness of the profound mental health impact of the pandemic has led to calls for investigative research and sharing of experiences by those who have lived through the myriad challenges presented by this unique public health menace. We describe here our process for a rapid needs assessment and creation of a pandemic resiliency and well-being support infrastructure for physicians and health care staff at a single academic medical center.

Intervention Detail. In March 2020, executive leadership from our health system and physician group partnered to create the Resiliency and Support Steering Committee (RSSC) to lead the rapid development of a pandemic needs response. RSSC efforts were inclusive across disciplines but our submission describes initiatives targeting physicians and advanced practice providers (APPs). Informed by international literature, national colleagues, and institutional needs awareness from pre-pandemic well-being efforts, the RSSC identified the following key priority areas: (1) psychological care; (2) medical care; (3) basic care; and (4) communication. Subcommittees were also formed in these priority areas. Direct outreach to both leaders and frontline workers within disciplines allowed efficient assembly of subcommittee work groups with multidisciplinary leaders across our identified focus areas. An institutional review board approved a brief, pandemic-focused needs survey that was designed and distributed. Respondents were asked about needs and prioritization of support preferences in the following areas: (1) proactive stress and psychological care support; (2) reactive stress and psychological care support; (3) medical care support; (4) basic life care support; (5) basic professional care support; (6) child care needs; (7) financial impact support; (8) other (an open-ended option to describe other needs); and (9) communication preferences. Subcommittee work groups met virtually at least weekly to review existing resources and evolve support in response to real-time experiential feedback. A shared-drive database for support development tracking was created so that initiatives and outcomes could be communicated in real time. All RSSC and subcommittee team leads across the 4 priority areas were granted access. This platform facilitated timely communication and collaboration across subcommittees, helped avoid parallel efforts, and optimized synergism.

Outcomes and Impact. Actualizing new initiatives in an academic medical center with over 1500 physicians and APPs can be tedious and always necessitates stakeholder identification and involvement, as well as dedicated coordination. By incorporating real-time feedback of frontline physicians and providers into response development guided by their clinical leaders, RSSC efforts could focus on support initiatives prioritized as most meaningful by recipients. Physicians and providers in our health system received similar needs surveys in April 2020 and August 2020. Response rates for these groups in April 2020 were: 26% (271/1048) faculty physicians, 36% (202/567) resident/fellow physicians, and 44% (213/488) APPs. August 2020 response rates were: 30% (321/1054) faculty physicians, 1.4% (8/578) resident/fellow physicians, and 16% (82/498) APPs. Of note, dissemination of the August 2020 survey to resident/fellow physicians was unfortunately interrupted, resulting in negligible response. Survey revisions in August combined reactive and proactive psychological care support into a single question stem; inquired about consideration of job change; and requested input on needs around diversity, equity, and inclusion in terms of self-care and advocacy or ally development. Physicians and APPs in many areas had similar needs prioritization. However, there was a statistically significant variance seen using chi-square testing in prioritization of reactive psychology support preferences, medical care support preferences, and communication platform preferences. Several key efforts were developed out of the RSSC’s work. The first was a targeted and intentional communication strategy, including a new, centralized, well-being support webpage. The second strategy was the aggregation of reactive psychological support in a variety of formats, from group to peer to 1:1 support. Third, a rapid access scheduling line to connect with primary medical care was developed for health care providers. Finally, a repository of basic care support resources, from grab-and-go food to laundry care and grocery delivery, was developed for facilitation of home and family support.

A Multidisciplinary Service Line Approach to Optimizing Neurology Documentation and Coding Practices

Christine Boerman, RN, SCRN, CNRN; Debra E. Roberts, MD, PhD; Dimitrios Manou, MD

University of Rochester Medical Center

Background. Case mix index (CMI) is defined as the average relative weight of diagnosis-related groups (DRGs) assigned to inpatient discharges. This metric, designed to reflect the resource intensity associated with treating inpatient cases, drives inpatient reimbursement models. Using the Vizient Clinical Data Base, we compared our stroke patient population’s CMI to the top 20 U.S. News & World Report neuroscience programs—noting that our program’s CMI was 17% behind the comparison group. Given the direct link between CMI and DRG assignment, we analyzed variables that impact DRG assignment to understand the underlying drivers contributing to overall CMI. An example was a discrepancy in the risk adjustment within base DRG 22 (acute ischemic stroke with thrombolytic), a triplet DRG having 3 levels of risk adjustment to reflect increasing resource intensity. Our peer institutions captured 35% of discharges under the highest risk-adjusted Medicare severity (MS)-DRG, compared to 16% captured within our institution. Instead, our institution was discharging up to 79% of stroke cases under the second-highest risk-adjusted MS-DRG. Adjusting our risk-adjusted mix to match that of our peer institutions would result in up to $150 000 in incremental revenue opportunity. Alongside the DRG data analysis, we sought to gain insight into current clinician documentation and coding processes. We met with our institution’s inpatient coding team to understand their approach and policies. We also performed a series of chart reviews aimed at summarizing clinician documentation practices. Ultimately, we found a number of variables that were negatively impacting DRG assignment, including (1) a lack of coding policy standardization; (2) low clinician query rates; (3) common usage of the “copy forward” function in documentation; and (4) vague notes describing concomitant comorbidities and conditions. Beyond these complexities, we also noted a cultural chasm between clinicians and coders contributing to a lack of communication that likely exacerbated these issues.

Intervention Detail. As part of an institutional initiative aimed at capturing potential inpatient margin improvement opportunities, we established a multidisciplinary team consisting of a physician, a nurse, an analyst, and coding professionals. Initial data analysis showed a disproportionate assignment of MS-DRG 062 (ischemic stroke, precerebral occlusion or transient ischemia with thrombolytic agent) to inpatient discharges within base DRG 22 compared to similar peer institutions. We partnered with the inpatient stroke coding team to perform chart reviews aimed at finding documentation deficiencies for select inpatient cases discharged to MS-DRG 062. Based on feedback from the coding team and internal review, changes were implemented to clinician note templates to encourage specificity and consistency within documentation. The note templates were redesigned to organize notes into sections grouped by body system to encourage complete and consistent documentation; each section was also updated to include a prompt for documenting the assessment and treatment plan to help coders quickly identify and code conditions that were addressed throughout the hospitalization. Alongside changes to clinician note templates, we provided a series of educational sessions for faculty, advanced practice providers, and residents to communicate updates to documentation policies, as well as promote buy-in toward new best practices. To help simplify complete and consistent documentation, pocket cards with frequently missed diagnoses were created and distributed to clinicians for quick reference. In addition to departmental changes, we were able to leverage benefit from institutional improvement initiatives. Our institution adopted new software, ChartWise, to standardize clinician querying, optimizing the feedback loop between clinicians and the clinical documentation improvement team. Our institution was also invested in increasing the capture of Elixhauser diagnoses; working with this improvement team, we made further changes to clinician note templates to promote awareness of Elixhauser diagnoses to encourage documentation targeted at capturing these diagnoses.

Outcomes and Impact. Ongoing efforts to improve clinician documentation specificity and completeness have returned promising results for our institution. Across our stroke patient population, the average number of diagnoses per case increased from 14.7/case in 2018 to 20.7/case in 2020. Across our institution, the average number of Elixhauser diagnoses per case increased from 2.8/case in 2018 to 4/case in 2020. Since U.S. News & World Report scores are calculated based on historical rolling averages, we anticipate these improvements will translate to better rankings in future years. Our improvement efforts also translated to increases in CMI and reimbursement. Overall, our stroke patient population CMI increased by 14% from 2018 to 2020, which translated to $2.3 million in incremental annual revenue. We achieved these improvements by navigating the cultural chasm between clinicians and coders to create a multidisciplinary team targeted at enhancing documentation and coding practices. Through continuous process improvement and leveraging feedback from the inpatient coding team, we implemented changes to clinician note templates that encourage specific, consistent, complete documentation for ideal hand-offs to the coding team. We also instituted a series of education sessions designed to inform on best practices for clinician documentation. Overall, we observed notable improvements to CMI and reimbursement within our stroke patient population by way of a multifaceted approach that can be readily applied to other departments and institutions interested in observing similar improvements.

Blast to the Past to Change Your Future (Rankings)

Adam McDougal, MSIM, University of Utah Medical Group; Terrell Rohm, MBA, University of Utah Health; Michael Lowe, PhD, University of Utah Health

University of Utah Medical Group

Background. The goal of hospitals is to provide excellent care with positive outcomes for patients. One way to evaluate patient care is through national rankings, scorecards, and registries for benchmarking quality metrics against similar hospitals. These rankings are used by patients when deciding where to go for health care, as well as by health care staff when choosing where to work. The main benchmark tool for the University of Utah Health is the Vizient Quality and Accountability dashboard. The system quality group has tried to maintain a balance in supporting the hospital’s overall goal to perform well in Vizient benchmarking while also helping some service lines perform well in the different rankings or registries they are historically more concerned with. When U.S. News & World Report released its 2019 and 2020 neuroscience rankings, the University of Utah fell from “high-performing” to not being ranked at all. This was frustrating for the neurosciences department because the department performs well in other national rankings and awards, and the University of Utah overall is consistently in the annual top 10 for the Vizient Quality and Accountability ranking for Comprehensive Academic Medical Centers. It was also frustrating for system quality because the “U.S. News” ranking methodology historically has been difficult to gain visibility into and difficult to reproduce. The quality improvement group decided to use the Vizient Clinical Data Base to better understand the U.S. News & World Report rankings, and to use this issue as an opportunity to engage more with the neurosciences department.

Intervention Detail. System quality developed an analytical plan, which heavily relied on the Vizient Clinical Data Base, to better understand why the neurosciences department fell in the U.S. News & World Report hospital rankings. In 2019, system quality created a proxy data set in the Vizient Clinical Data Base to resemble the Medicare claims data that is used by U.S. News & World Report for outcomes data. We restricted the payor source to Medicare and used the diagnosis-related groups outlined in the “U.S. News” methodology summary. The Vizient Clinical Data Base contained 22 of the 25 (88%) ranked hospitals for neurosciences in U.S. News & World Report for 2019, which meant our proxy data set in the Vizient Clinical Data Base was similar in composition to the “U.S. News” data set. In 2020, instead of creating a proxy data set, we were able to use the U.S. News & World Report specialties template. For the comparison group, we used the “U.S. News Top 20 Specialty HCOs: Neuro/Neurosurgery.” The primary measures we examined were mortality (which included the mortality index and an approximation of the 30-day mortality rate), the complication or comorbidity/major complication or comorbidity capture rate, and discharge status. Three of these 4 measures were pulled from the Vizient Clinical Data Base and the fourth (30-d mortality rate) was pulled from our electronic medical record. Using the proxy data set in 2019 and the “U.S. News” template in 2020 to evaluate mortality index, the University of Utah neurosciences department was in the bottom 50% with the comparison hospitals. The estimate of the 30-day mortality showed spikes in Q2 2017 and Q3 2017 and another large spike in Q3 2018. However, the complication or comorbidity/major complication or comorbidity capture rate for this time period was high when compared with similar hospitals. The 2 mortality measures were consistent with what would be expected after a drop in “U.S. News” rankings.

Outcomes and Impact. Using the Vizient Clinical Data Base and our electronic medical record, our system quality department was better able to understand why the neurosciences department dropped in the U.S. News & World Report rankings and present specific data points for improvement to the neurosciences department. Because U.S. News & World Report relies on a 2-year look-back period, it is possible to estimate future rankings—assuming the methodology remains consistent. This information was shared with neurosciences, which led to greater engagement by neurosciences senior leaders in quality improvement discussions. An action plan is being developed by system quality and the neurosciences department to improve key quality measures. The system quality group found the Vizient Clinical Data Base, as opposed to other national data sources, to be useful in assessing changes in quality improvement measures because the data is more current. There were some limitations in using Vizient to approximate U.S. News & World Report rankings. The majority, but not all, of the “U.S. News”-ranked hospitals are in the Vizient Clinical Data Base so the comparison group is an approximation. Another difference is that the Vizient Clinical Data Base does not track 30-day post-hospital discharge mortality, the mortality measure used in “U.S. News” rankings. The time period used for the 2019 “U.S. News” rankings outcome measures was not fully covered by Vizient, but in 2020 this was no longer an issue. In spite of the limitations, we found that the Vizient Clinical Data Base, together with internal data sources, was an effective tool to better understand the outcomes component behind the “U.S. News” rankings—and we anticipate that the Vizient Clinical Data Base could be used to approximate other data sources used in national rankings. This method of focusing in on key measures is especially relevant in this current period of restricted financial and human resources due to the demands of COVID-19.

High-Value Hemodialysis Hand-Off

Margaret Diane Giles, MSN, RN, CNL; Kristen Barnes, MSN, RN, CNL; Tammy Law, MSN, RN, CNL

Wellstar Douglas Hospital

Background. A subcommittee of the hospital length of stay (LOS) group comprised of certified clinical nurse leaders (CNLs) conducted discharge barrier tracking audits that identified common barriers to timely discharge and treatment of hemodialysis (HD) patients. Barriers to timely initiation of HD treatments included: (1) the HD nurses were unaware of pertinent information required to efficiently prioritize the HD patient schedule list for the day because of inadequate communication. The HD nurse needs relevant information, including expected day of discharge, conflicting procedures, isolation status, critical labs, and HD access issues. Late initiation of HD often resulted in an additional overnight stay; (2) a second communication error finding revealed that bedside nurses were unaware of what time the patient was expected to be dialyzed, which left patients unprepared (eg, uninformed, unfed, requiring hygiene, lack of transport help)—resulting in late meals and further delays in getting them ready for their HD treatments; (3) Fresenius Kidney Care reported over 95 hours of delays in initiating HD treatments in 3Q 2019; contributors to these delays included transportation challenges, conflicting procedures and meal delays during the 6-month period prior to study initiation; and (4) Gemba analysis revealed that patients were not able to order hot breakfasts before 8 am due to insufficient kitchen staff; patients often refused HD until they had their hot breakfast. This study highlights the value of the CNL role in implementing outcomes-based practice and quality improvement strategies. The need for more clinical expertise and leadership at the point of care is abundant. An important component of the CNL role, as a systems analyst and risk anticipator, is to improve the quality of patient care delivery with the aim of preventing medical errors and harm.

Intervention Detail. Gemba analysis results, process mapping tools, and discharge barrier tracking audits were shared with the hospital shared governance committee for review and input. An affirmative vote was received prior to initiating the proposed process changes and enhanced education for HD nurses, bedside nursing staff, and unit secretaries. The new bidirectional communication tool was introduced to inform the HD staff of pertinent information necessary to prioritize the daily HD patient list. Night shift charge nurses from all acute care units, the intensive care unit, and the emergency room were educated on the tool and instructed to fax completed forms to the HD department by 0600 each morning. Pertinent information required by HD staff included expected discharge day, scheduled surgeries, procedures, patients with fluid overload or missed HD, critical labs, hemodynamic instability, ventilator dependency, and isolation. HD staff was taught to use the tool, prioritize the patient list, and re-fax the list to the units by 0730 with expected chair times listed for each HD patient. Unit secretaries receive the fax and inform the bedside nursing staff of patient chair times so that breakfast, medications, and patient preparations are timely. To address the meal-related delays in starting HD procedures, collaboration with food and nutrition management resulted in adjusted staff hours to provide early morning coverage for hot breakfast preparation for first chair HD patients. The CNLs conducted daily audits to track compliance using the new HD Patient List tool. Once staff education was completed, manager notification was initiated if the new process was not followed. This provided immediate feedback for reviewing and correcting the reasons for problems as needed with reeducation or support. LOS data was tracked for the HD population by hospital LOS quality members and using the Fresenius Kidney Care reports delay data.

Outcomes and Impact. Post-intervention, Q1 2020 results showed a decrease of 53.9% in hours of delayed HD initiation due to conflicting procedures and meal delays. This is especially significant considering that the volume of HD procedures increased by 17.5% from Q4 2019 to Q1 2020. In addition, first quarter, post-study process changes resulted in a cost savings of $5084.13. This amount seems insignificant because it represents only the cost savings from the charges made from the time of delayed starts. However, this number would be much greater if the costs for LOS, additional overnight stay, HD staff, and patient safety were measured and included. It is difficult to compare LOS before and during COVID-19 for the HD population because many HD/COVID-19-positive patients required a longer LOS. HD staff has expressed overwhelming satisfaction with the improved efficiency in identifying HD patients and accurately determining the best chair time. Likewise, bedside nurses have reported increased efficiency and satisfaction with knowing the planned chair times early in their shift. This allows for improved patient readiness planning. Patient satisfaction has also improved with the recent ability to order hot breakfast items earlier.

Systemwide Implementation of Comprehensive Behavioral Risk Screening

Julia Bossie, MSN, RN, CEN, CNL; Freda Lyon, DNP, MSN, RN, NE-BC, FAEN

Wellstar Health System

Background. The evidence is vast and ever-evolving for the behavioral health (BH) population and the documentation that is required. For example, there are currently only 5 evidence-based tools approved by The Joint Commission and the Centers for Medicare & Medicaid Services to screen for suicide. In our behavioral risk screen we included 3 of the 5 evidence-based tools required by The Joint Commission, which were the Columbia-Suicide Severity Rating Scale (C-SSRS), Ask Suicide Questions (ASQ), and Patient Health Questionnaire (PHQ)2-Adolescents and PHQ-9. According to the Columbia Lighthouse Project (2016), the C-SSRS with Suicide Assessment Five-step Evaluation and Triage are validated tools to assess for suicidality in patients that present to the emergency department (ED).¹ The Emergency Nurses Association also has a clinical practice guideline that validates the ASQ and the C-SSRS as the suicide rating tools with the highest validity in the emergency department setting.² Additionally, the Joint Commission standard (NPSG15.01.01 EP 2, EP3 & 4, 2019) requires that all emergency departments screen patients for suicide, using a validated tool if presenting with a behavioral health complaint. Annual staff training is required within the standard, as well as reassessment. We were using 3 of the C-SSRS questions, but not the full screen. All patients were either high risk or no risk. High risk required 1:1 patient-to-sitter ratio. This increased sitter utilization was unrealistic in the ED and inpatient settings. Staff was pulled to fulfill the sitter needs, which led to an increase in safety concerns for all patients. Lack of assessment opened the potential for patients to be discharged home without proper screening. Serious safety events occurred due to lack of a comprehensive behavioral risk score, including full suicidal, homicidal, or other concerning behavior screenings. Documentation that was available was confusing and lacked a defined reassessment, and ED and infection prevention nursing were unable to see the behavioral assessor’s documentation. Risk was not always easily identified in the chart.

Intervention Detail. Our innovative build is the use of a single BH risk-screening tool that encompasses evidence-based tools tailored for multiple subsets of users and areas of care across 10 hospitals. The tool is customized based on the population and is completed by nursing upon the patient’s arrival to the hospital. Pediatric adolescents are universally screened for depression with the PHQ-A and for suicide with the ASQ. The adult population is universally screened for suicide, utilizing the C-SSRS. In addition to suicide, the BH risk score encompasses risk of harm to others; self-injurious, nonsuicidal behaviors; and other concerning behaviors, such as paranoia and psychosis, for both populations. Every answer generates a numeric score in the background that tallies into a final risk level. If the patient has a low, moderate, or high BH risk, a member of the BH team then sees the patient and utilizes the Suicide Assessment Five-step Evaluation and Triage. This assessment updates the BH risk level and staff will then adjust the sitter ratio if the risk level changes. The risk level is highly visible to all users in Epic Storyboard, Track Board (ED), and Patient List (IP). We added tools for safety, assessment, and reassessment that included required documentation, clinical decision support, and reassessment reminders (such as shift documentation or ED alerts).

Outcomes and Impact. This initiative improved workflow and patient care for a disparate population. Automating the workflow for overwhelmed nursing staff was a huge win. All the calculations work in the background helped to avoid operator error. In the ED Narrator, we leveraged toolboxes to separate out into initial and reassessment. Required documentation prevents missed documentation of the initial screens and reassessments. Based on suggestions from The Joint Commission, physiological and psychological reassessments were combined into a shift assessment to simplify documentation for the frontline nurse. Further improvements included ED alerts to give nursing a visual reminder of when shift assessment documentation is due. This was an effort to streamline care and improve reassessment documentation for patients awaiting placement. The patient safety tool used by sitters was added in Epic as a sidebar report used to automatically pull information into the report, which can be printed on each shift. We have seen an increase in patient risk identification. Accurate identification is earlier in the patient encounter, which shows a better use of resources by decreasing sitter utilization. This was measured by sitter full-time equivalent usage and patient risk-level stratification. Documentation improvements and built-in prompting of assessments and reassessments has improved care of the behavioral health patient. This was measured by our serious safety event data that measures patient harm events.

1. Course Hero. SUICIDE assessment CCSRS the Columbia Lighthouse Project. Identify Risk. Instructions: Instrument/Tool Criteria. Accessed January 31, 2022. https://www.coursehero.com/tutors-problems/Nursing/32123693-uicide-Assessment-CCSRS-The-Columbia-Lighthouse-Project2016-Identi/.

2. Emergency nurses association. Emergency Nurse. 2017;25:15–15.

Safety Event Management: The Journey to Zero Harm

Crystal Veal, RN, BSN, MSPSL

Wellstar Health System

Background. A multidisciplinary team mapped the current safety event management process using value stream analysis. In parallel, event reporting and root cause analysis (RCA) data were analyzed. Based on these data, it was determined that process redesign was needed and an evidence-based, standardized, safety event management process was implemented. In response, a kaizen was held, and a multidisciplinary team established a consistent process with accountability metrics. The redesign eliminated nonvalue-added steps and created standard work documents. Each step of the process was incorporated into the new electronic event reporting system, including:

• Frontline staff members were involved in the selection and build of the new electronic event reporting system; frontline quality/safety team members were involved in the safety event management redesign via value stream mapping and a 4-day kaizen.
• Team members involved in reporting safety events and conducting RCA were involved in this project.
• The board of trustees supported this project through overall patient safety program oversight.

Senior leadership supported the project through allocation of resources, fostering a culture that thanks and protects those who report safety concerns, and prioritizing and committing to serve as RCA executive sponsors. All team members, including physicians and nursing, committed to serving on RCA teams. Patient advocates are included in several hospital-led committees and their feedback is considered in specific initiatives. Plans involve inclusion of patient advocates in patient safety committees and RCAs. This work is important to share because many organizations struggle with reliable RCAs, as they are very complex and resource intensive.

Intervention Detail. Each step of the process was incorporated into the new electronic event reporting system. Best-practice RCA methodology was implemented, including: (1) alignment with National Patient Safety Foundation recommendations in “RCA2: Improving RCA and Actions to Prevent Harm”; (2) developed standard work and incorporated the standard work for safety event management in the electronic event reporting system; (3) conducted RCA training for 84 team members, leaders, and physicians; (4) established baseline metrics with ongoing measurement; and (5) established a multipronged approach to share lessons learned across the system. Baseline metrics were collected prior to the redesign and training, and metrics are tracked each month for monitoring and prioritization of improvement efforts.

Outcomes and Impact. The safety event management process redesign was successful. The team was engaged and committed to implementation of the best-practice RCA methodology. When Culture of Safety Survey results were reviewed, performance signified a robust culture of reporting, as evidenced by the “feedback and communication about error”—which exceeded the national average by 5.3%.

Compared to baseline, performance includes:

• A 23% decrease in serious safety events.
• A 19% increase in safety event reporting.
• A 164% increase in great catch/near-miss reporting.
• A 164% increase in detection of reportable safety events.
• Percentage of RCAs that include identification of strong* action items is 86%, compared to baseline performance of 40%.
• Percentage of RCAs that adhere to best-practice meeting methodology is 90%.
• Percentage of corrective action items implemented or in-process (not overdue) is 88%, compared to baseline performance of 24%.
• A 122% increase in senior leaders serving as executive sponsors for each RCA.
• A 22% increase in the identification of measures of success.

With our safety and reliability culture transformation, including safety event management process redesign, we expect an 80% reduction in preventable harm when the programs are fully implemented and sustained. Thus far, peak performance demonstrates a 23% decline in our serious safety event rate. Zero harm is the goal and true measure of our success. To foster consistency of safety event classification, transparency, and shared learning, a weekly call was established that includes key stakeholders. Safety events from the previous week are discussed and consensus reached regarding classification and immediate actions to prevent recurrence. A debrief is conducted after each call and improvements implemented, utilizing the Plan-Do-Study-Act improvement cycle. In addition, senior leadership at the system executive vice president and hospital president levels conduct weekly safety huddles to discuss events or trends that require proactive attention to mitigate risk of occurrence elsewhere in the system.

*Strong action item is defined in the white paper “RCA2: Improving Root Cause Analyses and Actions to Prevent Harm.”

We Deliver Babies, Not Infections

Allison Blank, MSN, RN; Emily S. Shrontz, BSN; Amanda H. Hargis, BSN, RN

Wellstar Health System, Douglas Hospital

Background. In 2019, the new women’s center was opened that housed 10 labor/delivery/recovery/postpartum rooms, 1 operating room, and 4 neonatal intensive care unit rooms. Historically, staff would turn over care of the patient to the main operating staff to complete all cesarean sections until the new center opened. Months of training went into preparing obstetric staff on the prepping and management of cesarean patients. Of note, the main operating room had gone over 4 years without a surgical site infection (SSI); however, from May 2019 to September 2019, this unit experienced 3 SSIs—equating to a 15% SSI rate. One of the SSI cases involved a readmission that separated mom and baby. During a unit staff meeting where quality metrics were reviewed, it was reported the unit had experienced another SSI. Two registered nurses within the department accepted the challenge of identifying gaps in the current process. The overall goal of this project was to eliminate SSIs in the postpartum patient.

Intervention Detail. The team collaborated with infection prevention nurses for guidance and to identify a benchmark for the obstetric population related to SSI rates. It was difficult to pinpoint this specific criterion, so ultimately it was decided that if an infection could cause harm to the patient, then the goal or benchmark needed to be zero. We utilized data for all postpartum C-sections from the previous year and determined the need to help prevent SSIs. We utilized our internal incident reporting system and chart reviews to find the necessary data. A literature search was conducted, and articles identified the best evidence-based practice for preventing SSIs in the post-cesarean patient population. Specific findings were appropriate prophylactic antibiotics, proper skin preparation, hand hygiene, and maintaining proper temperature during the procedure. Additionally, clinical experts on the products utilized for these practices were consulted to ensure that we were using proper techniques when performing the abdominal prep. The team joined forces with the obstetricians and anesthesia providers to share research data and work on a communication plan for the unit. Multiple changes occurred to improve quality and safety. All patients admitted to the unit who will be delivering a baby, whether by scheduled cesarean or for labor, are instructed to shower with chlorhexidine gluconate (CHG) upon arrival. The process for prepping the abdomen prior to cesarean was improved and now adheres to manufacturer guidelines. A second prophylactic antibiotic, Zithromax, became standard for any unscheduled cesarean. Intraoperative temperature is monitored on all patients undergoing a cesarean. Providers, assistants, and surgical techs involved in the cesarean change their sterile gloves prior to skin closure, and lastly, all staff members perform a 3-minute surgical scrub upon arrival for their shift.

Outcomes and Impact. Monthly tracking of the SSI rate was key in monitoring the initiative’s progress. Chart reviews were conducted capturing both CHG bath documentation on admission and postoperative patient education. This process has sustained a positive trajectory and can be transferred to other surgical departments. The successes of this initiative were reviewed by the main operating room team, which adopted the CHG bath on admission just prior to surgery for all surgical patients. Since implementation of this project, the obstetric department has been able to sustain zero SSIs in the last 12 months.

Workflow Optimization: Enhancing Quality of Care for Patients With Sepsis

Alex Forman, BS; Hannah Trickett, PharmD, BCPS

WVU Medicine

Background. The Vizient Clinical Data Base includes a risk model to precisely calculate key performance indicators (KPIs). It is important for organizations to monitor these metrics and maintain them below a designated threshold to sustain the quality of care provided to patients—specifically those with diagnoses that carry high morbidity and mortality risks, such as sepsis. The Vizient Clinical Data Base patient outcomes and readmissions reports were utilized to obtain and organize data into a systemwide sepsis dashboard to monitor these KPIs. Sepsis is considered to be a medical emergency in which time-sensitive interventions must be made. The sepsis dashboard was originally created for health care providers at a large academic medical center to incorporate the Centers for Medicare & Medicaid Services SEP-1 core measure into the sepsis workflow. The SEP-1 core measure consists of a 3-hour and 6-hour bundle of interventions that, when completed in a timely manner, can improve outcomes for the sepsis patient. The sepsis dashboard was monitored regularly for systemwide performance in an effort to identify optimization strategies for improving workflow and enhancing quality of care. A sepsis committee was created in conjunction with the dashboard to evaluate and trend specific sepsis metrics and assist in decision-making and implementation of new strategies to reduce overall mortality, length of stay, and readmission rates.

Intervention Detail. Sepsis data was imported from the Vizient Clinical Data Base Report Builder, which is based on the 2020 risk model (academic medical center). A Tableau dashboard was developed for continuous monitoring of specific KPIs, including sepsis patient counts, inpatient mortality, length of stay, 30-day readmission rates, and 30-day mortality. Centers for Medicare & Medicaid Services compliance was also included in the dashboard by incorporating the percentage of patients with initial lactate level, blood cultures drawn prior to antibiotic administration, and broad-spectrum antibiotic administration within the first 3 hours of time zero. A repeat lactate within 6 hours of an elevated initial lactate was also tracked on the dashboard.

An interdisciplinary sepsis committee was formed comprised of all levels of hospital staff (ie, coding, quality, clinical documentation, information technology, strategic analytics, pharmacy, nursing, and multiple service line providers). Department leaders were chosen to represent their respective service lines to evaluate how patient care could be optimized to ensure smooth transition of care. Committee members reviewed patient-level sepsis data from the dashboard to determine cause and effect, leading to observation of certain trends in the various sepsis metrics.

Outcomes and Impact. Development of the sepsis dashboard and the multidisciplinary sepsis committee allowed our organization to investigate various areas of opportunity to improve care for sepsis patients with a high level of patient detail included. Compliance with specific SEP-1 quality metrics was tracked and those patients for which appropriate care was not provided within the specific time interval were reviewed and discussed. Cause and effect were determined, and a plan was developed to address the reason(s) for noncompliance and prioritized on a project list for implementation.

Various projects that resulted from access to sepsis data included.

• Best practice advisory optimization. The best practice advisory was triggering time zero on the sepsis dashboard in the majority of patient cases. The predictive model and the corresponding positive and negative predictive values at each threshold were further evaluated to improve the sensitivity of accurately identifying sepsis.
• Order set updates. Order set names were updated to eliminate confusion and simplified to contain only the pertinent orders a provider would need to quickly activate appropriate labs and medications.
• Auto-reflexed lactate labs. Lactate lab ordering options were streamlined and optimized to automatically reflex if the initial value was elevated, eliminating the manual process for providers.
• Nursing protocol development. These protocols were discussed and a plan established to eliminate confusion surrounding the existing protocol and simplify the nurse workload by removing the task of contacting the physician prior to placing specific orders.
• Staff education. Education was delivered first to the sepsis committee to make sure each department representative was aware of current workflows across the hospital. A focus on early detection of sepsis, current evidence-based medicine, SEP-1 requirements, and appropriate clinical documentation within the electronic health record were the initial points of staff education.

Ongoing implementation of a systemwide sepsis campaign will continue increasing awareness of the time-sensitive patient care required when treating sepsis patients. The importance of the Vizient Clinical Data Base and electronic health record system data to track performance at individual hospitals, as well as throughout the health system, has been instrumental in identifying areas of opportunity.

Volunteers Helping Patients Virtually in Times of COVID-19

Fernanda Clariana, MA

Yale New Haven Hospital

Background. To transition the Music for Healing volunteer program to virtual operation, we needed to identify if the hospital’s patient television system would allow external videos made by our volunteers. Our TV system vendor specialist provided us with basic videomaking guidelines and we asked our music volunteers to send us videos of themselves playing music. Volunteers were eager to help the hospitalized patients remotely, particularly during these unprecedented times. The music volunteers started working on their music performance videos and transferred them to us in volunteer services. We edited most of the videos, making sure they conformed to hospital specifications. We then transferred each video to our hospital television system. In collaboration with the TV system vendor specialist, we recreated and rearranged the TV menus for patients to easily reach and view the videos. We created libraries of volunteer music videos and uploaded them into the patient TVs, which are available systemwide. We were able to reach many isolated patients and provide them music videos that include the volunteers heartwarmingly introducing themselves before performing. During the 2020 holiday season, the volunteers sent us holiday-themed music videos, with wishes of hope and healing dedicated to the patients. Volunteers continue to submit their dedicated music videos to be viewed by inpatients throughout our health system delivery networks.

Intervention Detail. We most often focus on the impact that educational content has on patient outcomes and improving engagement in patient care and recovery, but positive messaging has had a larger impact while dealing with the COVID-19 pandemic. Education plays a key role in recovery, but providing patients with distraction and relaxation has become even more important as they deal with uncertainty, doubt, fear, and a lack of control over their situation. Music is a therapeutic tool that brought patients comfort during a time of loneliness as families and friends were not allowed to visit. In addition to the Music for Healing volunteer videos, we added to our television menus—including guided relaxation videos recorded by reiki volunteers, therapy dog videos made by our dog volunteer teams, and get well cards from the Yale New Haven Health System community that showcase electronic versions of handmade cards with background music played by volunteers. After observing how well patients responded to these types of relaxation/entertainment content, we decided to offer virtual-live music and virtual-live relaxation sessions provided by volunteers over the phone or through Zoom to patients interested in a more personal approach. Music for Healing volunteers now offer “special musical wishes” in the form of live-virtual musical sessions with inpatients by phone or using FaceTime/Zoom. We make sure that clinical staff assist us in making the virtual-live sessions happen, either by providing a hospital cell phone, iPad, or laptop to connect with the patients. We have provided “musical birthday wishes” to palliative care patients via Zoom, with the Palliative Care Team and patient’s family present—all remotely. Sometimes these musical wishes can be a palliative care patient’s last wish.

Outcomes and Impact. The latest data collected indicates that we have 969 “volunteer-made video views” per month across the health system between the start of the Volunteer Relaxation Channel on May 29, 2020, and January 27, 2022. We continue to record data of video views per month on an ongoing basis. The total number of video views by patients throughout the health system between the start of the Relaxation Channel until January 27, 2022 is 19 376. These are very solid and impressive utilization numbers that confirm the importance of this type of content in patient well-being.

Lead Poisoning. Reducing Risk in the Most Preventable Childhood Illness

Marta Kostecki, LCSW

Yale Regional Lead Treatment Center

Background. In Connecticut, over 70% of the housing stock was built prior to the ban of lead-based paint in 1978. There are about 1600 known childhood lead cases in Connecticut—a number likely underreported due to the 57.4% completion rate of both mandated blood lead screenings at ages 1 and 2 years old.¹ Ideally, any child burdened by lead toxicity should be able to move into a lead-safe home or have their current dwelling abated, but for many, this is not the case. In spite of grand efforts by local health officials to hold homeowners accountable, only a small percentage of abatement orders are remediated each year, taking an average of 22 months to complete.¹ While the general public understands that lead exposure is problematic, there appears to be a gap in knowledge as to what it means to be poisoned by lead. There is an even greater discrepancy in recognizing potential lead hazards in the home and how to stay safe when relocation or abatement is not an immediate option. In 2016, the Regional Lead Treatment Center at Yale New Haven Hospital (YRLTC) shifted its treatment model from a predominantly medical approach to one that incorporates public health and social work competencies in its evaluation of lead-poisoned children. A joint approach, developed by a healthy homes specialist and a licensed clinical social worker, established a home visiting routine that ranks the accessibility of potential lead hazards by children within their home. Ongoing case management ensures that children obtain timely follow-up testing and are referred to community resources, such as early childhood intervention services, medical-legal partnership, and funding sources for lead abatement (when available). The YRLTC maintains a presence at state and local levels, connecting with a variety of partners to expand the reach of lead literacy and reach community agents that often work in silos.

Intervention Detail. A parent’s first introduction to lead poisoning is often presented through their pediatrician. Limitations in meaningful conversations occur, however, as the provider’s ability to diagnose the exposure source is only as good as the parent’s ability to recognize and report the conditions of their home. The YRLTC seeks to expand beyond the pediatrician’s basic screening questions to incorporate knowledge of housing, developmental milestones, and activities of daily living to provoke suspicion toward otherwise normal childhood behaviors. Staff members explore the way family members gain entry into their home, utilize their space, and are impacted by climate and other circumstantial conditions. For example, a 3-year-old who climbs up a staircase on their hands and feet to access a second-floor apartment may be coming into contact with lead dust if it is being tracked into the common area from a lead-positive porch. Seemingly innocuous behavior is now identified to be a lead risk, but it can be remedied with simple behavioral changes. Custom care plans are focused on empowering parents to better understand how to recognize potential lead hazards in their current and future homes, and they receive coaching to implement low-to-no-cost interim controls until a long-term solution can be achieved. The YRLTC often collaborates with local health departments to simplify inspection findings and improve compliance. Outside of a monthly clinic, the YRLTC accommodates families by utilizing satellite blood draw stations across the entire Connecticut coastline. The YRLTC has partnered with The Child Health and Development Institute of Connecticut to provide continuing education credits regarding lead screening practices to pediatric offices in its catchment zone. Since 2018, 19 practices have participated, with more than 200 medical professionals in attendance. A clinical lead algorithm is shared to help practices manage lead cases independently and know when to refer to the treatment center for additional guidance.

Outcomes and Impact. Since its model shift in 2016, the YRLTC has completed 430 home visits across southern Connecticut and closed nearly 1000 cases of childhood lead poisoning (cases that have dropped below 5 micrograms of lead per deciliter of whole blood [μg/dL] as set by the Centers for Disease Control and Prevention). The YRLTC has been successful in reducing the number of children requiring hospitalization at levels of 45 μg/dL and higher. Aggressive interventions at the onset of a critical blood lead level prevents the trauma of a family being removed from their home, often permanently, and protects valuable hospital resources while avoiding the exorbitant cost of social admissions. The majority of children requiring inpatient chelation are patients outside of the YRLTC with no prior history of elevated blood lead levels. Complementary to its presence in the community, the YRLTC has served as a preceptor to multiple community-based research projects through the Yale School of Public Health. A recent team of students studying for their master’s degrees in public health analyzed a sample of 332 resolved lead cases to compare intervention outcomes among patients. It was calculated that enhancements made to the home visiting approach correlated to a decrease in the average length of time patients tested with a detectable blood lead level. Improvements were seen most prominently in children with blood lead levels between 10 μg/dL and 44 μg/dL—up to a 15-month decrease when compared to their counterparts who were not included in the home visiting model. Implications for these findings are significant, as a rapid resolution in blood lead level suggests a reduction in the total length of time lead is present to interrupt critical brain development in the first years of life. The Yale New Health System chemistry lab processes approximately 6000 blood lead level screenings per year. Ongoing outreach has increased the total number of referrals received from outside the health system and access to the Connecticut Lead Surveillance Database helps ensure that as many children as possible remain in screening compliance, regardless of location.

1. Hung T, Ortiz D, Davila J, et al. CT Department of Public Health 2017 Annual Disease Surveillance Report on Childhood Lead Poisoning Prevention and Control. Connecticut Department of Public Health. 2019. Accessed February 28, 2021. https.//portal.ct.gov/-/media/Departments-and-Agencies/DPH/dph/environmental_health/lead/Surveillance_reports/CY-2017-Annual-Lead-Surveillance-Report-_Updated-2-27-2020Final.pdf.

Additional Power Huddle Sessions

Strategies for Hospitals Investing in Community Health and Equity

Karley M. King, MPH, MBA; Robin Hacke, AB, MBA; Terri M. Scannell, BA, MBA, Law (UK); BJC HealthCare

Is There a (Supply Chain) Doctor in the House?

Jimmy Chung, MD, MBA, FACS, FABQAURP, CMRP; Providence; Stacy Brethauer, MD, MBA, FACS; The Ohio State University Wexner Medical Center

Implementing and Optimizing Biosimilar Medication Use at Mayo Clinic

Chelsee Jensen, PharmD, RPh; Scott A. Soefje, PharmD, MBA, BCOP, FCCP, FHOPA; Mayo Clinic

Allina Health Cloud ERP and Procurement Shared Services Transformation

Tom Lubotsky, FACHE; Stephanie Derichs, MBA; Allina Health

Your Health Deserves a Partner. Navigating COVID-19 for the Latinx Population

John Fletcher, BA; Michael Evans, MD, MPH; Jodi W. Lenko, MD; Lehigh Valley Hospital - Hazleton

Primary Care in a Digital-First World

Doug Marx, DO; Mike Anderes, FACHE; Erika Smith, PharmD, FACHE, FASHP; Froedtert & Medical College of Wisconsin

Establishing a Clinically Integrated Supply Chain Through Vizient Procedural Analytics

Alex Matov, MBA; David Van de Kamp; Jason A. Stein, MD; MedStar Health

Decreasing Health Inequities at Emergency Department Triage

Nadia Huancahuari, MD; Diane Miller, MSN, RN, ACNS, CEN; Brigham and Women’s Hospital

Virtual Onboarding Program Design for Physicians and APPs

Emily L. Thomas, SHRM-SCP; Jennifer Hunt, MBA; OSF HealthCare

Building Critical Testing Capacity to Support COVID-19 Pandemic Response

Andrew Maze, BSB, AS MS, AS H, AA SHB; Michelle E. Bosewicht, MBA-HCM, MLS(ASCP)SBB, DLM; Providence

Family Connect. Supporting Families With Daily Clinical Updates During COVID-19

Katherine Hochman, MD, MBA, FHM; Regina Grinblat, MSEd; Rachel Ramsey, MHA; NYU Langone Health

Hardwiring Our Path to High Reliability!

Janet A. Ragle, DNP, MSN, MBA, CPHQ; Anne Shriner, MD; Nash UNC Health Care

Sustainability Leadership Across the Supply Chain Enterprise

Aparna Dial, MS, MBA, LEED AP; Kai Peters, CPSM; The Ohio State University Wexner Medical Center

Bridging the Gap. Improving Postdischarge Care Transitions Telemedicine

John Amory, MPH; Kevin P. Eaton, MD; Anjana Sreedhar, MPA; NYU Langone Health

Preventing COVID-19 Hospitalizations by Implementing a Home Monitoring Program

David Milbrandt, MD, FACEP; Tracy Telander, RN, BSN, PHN, CCM, MBA; M Health Fairview

Virtual Urgent Care Victories

Meeta Shah, MD; Renata McKim, MHA; Rush University Medical Center

How Not to Strike a Nerve. Engaging Spine Surgeons in Vendor Reduction

David Wente, RN, MS; Alpesh A. Patel, MD, MBA; Northwestern Medicine

Transforming Ambulatory Supply Chain

Robert Martin, MBA; Gabe Toledo; University of Chicago Medical Center

Additional Poster Presentations

Supply, Demand, and the OR Internship

Vickie Ripkey, MSN, MBA/HCM, RN, CNOR; Lehigh Valley Health Network

Dynamic Duo. Partners in Combating Sepsis Bundle Compliance

Tusdi Rodriguez, MSN, RN, CCS; Keck Hospital of USC

The Game of Life. Care Across the Continuum

Kathryn Martinez, MSN, RN, Cari Johnson, BSN, RN; FHN

Reducing ICU Hospital-Acquired Pressure Injury Rate With Diaper-Free Initiative

Britney Whaley, BSN, MSN, RN-BC, CWON, CWS, CNL; Wellstar Paulding Hospital

Launching HOPE (Health Outreach Promoting Equity). Pandemic Response to Sustainable Model

Kelly Pigott, MSW, LCSW, ACM; Erica Saito, MHA; Northwestern Medicine

Who’s on First. A Team Approach to Goals of Care

Weston Jordan, BS IE; Kristen R. Lakis, MDiv, LCSW; Betsy Fricklas, MMSc, PA-C; Sidonie (Donie) Magee, RN BSN, CHPN; Randall Scott Pollard, MSW, BSN, RN; Jennie Petruney, MSN, ANP-BC, AOCNP; Carey Anders, MD; Natalie Ashley, BN, GradDipAdvN; David Casarett, MD, MA; Eris Worlds; Duke University Health System

Hospitalwide Risk for Pediatrics. Does the Vizient Model Stack Up?

Janelle Lee, DrPH; Katherine Pavlovich, MPH; UCSF Health

If You Build a (Vizient) Website, Will They Learn?

Lindsey Walicek, MPH; Justice Dahle, MHA; UCSF Health

Impact and Clinical Outcomes of the Troponin Optimal Care Pathway

Vedant A. Gupta, MD; Daniel Moore, MD, FACEP; Alison Woodworth, PhD, DABCC, FAACC; UK HealthCare

POC Blood Gas Testing in Critically Ill Patients With and Without SARS-CoV-2 Infections

Thai Osborne, MHA; Lisa Wright, MHA, RRT-NPS; Alison Woodworth, PhD, DABCC, FAACC; Erin Weber, JD, MT; Shari Fullenlove-Cook, BS, RRT; UK HealthCare

Emergency Department Navigation. No Longer Questionable

Bradley Leonhard, BSN, RN; Julie Nevers, MN, BSN, CCM; Angela Gottschalk, BSN, RN, CCM; St. Tammany Health System

Movement to an Accelerated Discharge Process for TAVR Patients

Cecilia Mortorano, MSN, RN, NEA-BC; Patricia Keegan, DNP, NP-C; Emory University Hospital Midtown

INDEX

A

Allina HealthS129

AtlantiCareS18

B

BJC HealthCareS129

Brigham and Women’s HospitalS19, S20, S84, S129

C

Carilion ClinicS20, S78

D

Denver Health & Hospital AuthorityS84, S85

Duke University Health SystemS22, S87, S130

Duke University HospitalS23

E

Emory HealthcareS24, S87

Emory University Hospital MidtownS25, S130

F

FHNS26, S130

Froedtert & Medical College of WisconsinS27, S28, S29, S30, S89, S129

Froedtert & the Medical College of WisconsinS32

Froedtert HealthS32

Froedtert HospitalS33

H

Hackensack Meridian NetworkS89

Houston MethodistS91

I

Indiana University HealthS34

Intermountain HealthcareS35, S36, S38, S92, S93

IU Health—Riley Hospital for ChildrenS39

J

Jefferson HealthS94

Johns Hopkins Health SystemS39

K

Keck Hospital of USCS130

L

LAC + USC Medical CenterS95

Langone HealthS130

Lehigh Valley Health NetworkS96, S97, S130

Lehigh Valley Hospital - HazletonS129

Lifespan—Rhode Island HospitalS40

M

M Health FairviewS130

Maimonides Medical CenterS41, S98

Mayo ClinicS99, S129

MD Anderson Cancer CenterS42, S74

MedStar HealthS129

Memorial Hermann-Texas Medical CenterS43, S100

MU Health CareS44

N

Nash UNC Health CareS129

Nebraska MedicineS101

Northwestern MedicineS102, S103, S104, S105, S130

Northwestern Medicine Lake Forest HospitalS102

NYU Langone HealthS130

O

Ochsner HealthS45

Oregon Health & Science UniversityS46

P

Penn MedicineS47

Phoebe Putney Memorial HospitalS48

Prisma HealthS49

ProvidenceS50, S129

R

Redlands Community HospitalS106

Rush University Medical CenterS51, S53, S54, S130

S

Saint Luke’s Hospital of Kansas CityS107

Sanford HealthS55

Seattle Children’sS108

SoutheastHEALTH109

St. Tammany Health SystemS130

Stanford Health CareS56, S58, S59, S60, S111, S112, S113

Stanford School of MedicineS59, S113

Stanford University School of MedicineS60

Stony Brook MedicineS61, S115, S116

Strong Memorial HospitalS71

T

Tampa General HospitalS62, S116

The Miriam Hospital, LifespanS63

The Ohio State University Wexner Medical CenterS129, S130

The University of Kansas Health SystemS119

The University of Toledo Medical CenterS75

U

UAMSS117

UC Davis HealthS64, S65,

UC Davis Medical CenterS66, S67

UCSF HealthS130

UK HealthCareS130

University of Chicago Medical CenterS130

University of Iowa Health CareS118

University of Kansas Cancer CenterS68

University of Kentucky HealthCareS69

University of Michigan Health—Michigan MedicineS70

University of RochesterS71

University of Rochester Medical CenterS72, S73, S120

University of Texas Health Science Center at HoustonS74

University of Utah HealthS76

University of Utah Medical GroupS122

UT Health McGovern Medical SchoolS100

UT Southwestern Simmons Comprehensive Cancer CenterS77

V

Virginia Tech Carilion School of MedicineS78

W

Wake Forest Baptist HealthS79

Wellstar Douglas HospitalS123

Wellstar Health SystemS80, S124, S125

Wellstar Health System, Douglas HospitalS126

Wellstar Paulding HospitalS81, S130

Y

Yale New Haven Health SystemS83

Yale New Haven HospitalS127

Yale Regional Lead Treatment CenterS128