Multisite studies are often conducted in the practice setting focusing on improving processes of care, adopting evidence-based practices (EBPs), and improving patient outcomes. The research is intended to generate new knowledge-answering specific questions or testing hypotheses needed to inform practice decisions. Yet, in our team’s experience, we observed that embedding research in health systems also results in changes in the health system and participants. If an investigator attends only to the answers to the primary questions of the study, these changes remain unknown and uncaptured. It is the “story behind the story” in the implementation and conduct of studies that reveals how individuals and organizations are profoundly transformed. These stories also suggest a stronger capacity for future innovations required of a learning healthcare system.1,2
In this article, we describe our 40-hospital multisite study to give readers the context of the research activities. The study’s conceptual framework based on a model for diffusion of innovations in organizations3 is presented to illustrate the study design and the outcomes measured. The Knowledge-to-action (KTA) framework4,5 is introduced and used to illustrate how hospitals changed heart failure (HF) care processes in striving to implement the best evidence for care. Cases that were voluntarily submitted by participating hospitals are cited. The changes reported relate to contextual factors, implementation strategies, and the reported consequences.
Improving Heart Failure Outcomes Study
Heart failure affects 5.7 million people in the United States at a cost of nearly $34.4 billion each year.6 It is one of the most common reasons for hospital admission for people older than 65 years and has recently been a focus of intense initiatives to reduce the high rates of readmission. Although the measures reflecting medical management have steadily improved and show a ceiling effect, other national guidelines exist for the care of patients with HF in acute care settings and their transitions to home.7 Nurses often assume the primary responsibility for these important evidence-based recommendations.
The Improving Heart Failure Outcomes (IHO) study was a quasi-experimental study to evaluate the effect of 3 nurse-centric recommendations: teaching standardized HF content, making a follow-up appointment before discharge, and calling patients after discharge to clarify instructions and promote self-care.7 Forty hospitals with Magnet® status were recruited to participate. Outcomes examined included HF core measures, 30-day readmission to the same hospital, readiness for discharge, self-care, and HF knowledge. The IHO study received expedited approval by the University of Maryland, Baltimore, institutional review board. Each hospital obtained approval from their local board.
Participating hospitals identified 1 non-ICU that cares for HF patients for the study and a site coordinator to champion the initiative. Site coordinators developed a team to work on the study. During a 2-day collaborative meeting at the beginning of the intervention, the hospital teams received education to implement the protocol at their hospital including a review of HF physiology, best practices for HF teaching, how to obtain informed consent, and motivational interviewing. All materials, including videos of the sessions, were provided to hospitals via an electronic portal. During the study, 8 monthly synchronous meetings were conducted using distance technology. During these meetings, study coordinators asked clarifying questions and shared their strategies for patient recruitment and implementation. A 2nd collaborative meeting was held at the end to provide an opportunity for debriefing and sharing their experiences. Poststudy conversations at the final collaborative revealed the lessons learned and added knowledge about implementation strategies that resulted from participation.
Diffusion of Innovations Model
To guide the development of the IHO study, an adaptation of the Conceptual Model for Considering the Determinants of Diffusion, Dissemination, and Implementation of Innovations in Health Services Delivery and Organization was used.3 This model explains how adoption influences implementation or use of the evidence and subsequent consequences. Figure 1 frames the concepts of adoption/assimilation, implementation, and consequence as used in the IHO study. A direct linear relationship is proposed between these 3 concepts. Adoption and/or assimilation represent the process of incorporation of new evidence into practice. In this study, it is adoption of evidence-based HF care. Implementation is the actual use of the new evidence. Consequences are the result or outcome of adoption of new evidence. Although the patient outcomes measured in IHO (patient self-management, knowledge, and readmission) were important, they failed to capture the story behind the implementation challenges.
The purpose of this article was to describe how participating hospitals innovatively implemented the study in their hospitals—and to document some of the unforeseen benefits to nurses, hospitals, and patients. To capture these changes that were recognized more clearly after the study concluded, all IHO participating hospitals were invited to submit a short description of their experiences and the impacts realized. Twelve study hospitals submitted a case. Quotations abstracted from these cases are used to illustrate the experiences and impacts (see Document, Supplemental Digital Content 1, for the full cases, http://links.lww.com/JONA/A266).
The KTA Framework
The KTA framework was chosen to illustrate these changes because it is grounded in planned action theories.4,5 Figure 2 illustrates the KTA cycle. The framework integrates the concepts of knowledge creation and action, recognizes the importance of local context, and accepts that phases may occur sequentially or simultaneously. The knowledge creation funnel in the center is seen in 3 stages: knowledge inquiry, knowledge synthesis, and knowledge tools/productions. The IHO protocol, which included standardized HF teaching, confirming follow-up appointments before discharge, and making phone follow-up calls at 48 hours, reflects the knowledge. The 7 action phases reflect how knowledge can be implemented in systems and groups. The following summary uses the KTA cycle to illustrate how the study hospitals implemented the protocol, learned from implementation and participation in a multisite study, and the resulting consequences.
Illustrations of Change in IHO Hospitals
Adopting Knowledge to Local Context
As described by the developers of the KTA framework,2 adopting knowledge to the local context reflects the processes that individuals or groups go through as they make decisions about the value, usefulness, and appropriateness of particular knowledge to their setting and circumstances. In deciding how to best implement the IHO protocol in their hospitals, each site examined their current practices to develop the best approach. In addition to the 2-day collaborative meeting where training was done, an investigator-created toolkit was provided to sites including PowerPoint presentations, videos, and written materials. All were made available through a Web-based portal. Many study sites used these materials to educate the staff nurses, to ensure the completeness of their nurse and patient teaching materials, and to review their procedures when a patient screens positive for depressive symptoms.
One example of this adoption to a local context was how study hospitals scrutinized their processes for teaching HF patients. The content taught by pharmacists, social workers, physicians, and advanced practice nurses was examined, and this often led to a more comprehensive and efficient approach to teaching. As noted in hospital case 1, the importance of a multidisciplinary approach was recognized and incorporated into care management:
A daily unit huddle is held with numerous disciplines including the charge RN, pharmacy, social services, case management, palliative care, home health, dietary, the cardiovascular clinical nurse specialist (CNS), the nurse manager, cardiac rehab, and a hospitalist.
The need for coordination with existing initiatives was also recognized:
…the IHO study team determined that consistency in the delivery of patient education would help to ensure that all components of education were addressed. The cardiac rehabilitation nurse team is composed of 4 RNs. With the initiation of the Hospital to Home project, their patient population was expanded to include patients with a diagnosis of HF. A small core group of staff RNs on the HF unit reinforced the primary education that was completed by the cardiac rehabilitation team.
The hospital in case 2 recognized the need to change their teaching materials and solicited support from other hospital departments to make this happen:
From the knowledge gained from this research, we realized we were lacking an evidence-based HF education booklet for patients and families. With the support of cardiology administration and the university marketing department, a 23-page education booklet was designed and written and made available in November 2011. This discharge booklet was a direct result of new information gained from the IHO study. Each HF patient is discharged with written information on understanding HF. The education booklet supported standardizing discharge teaching to include the definition of HF, symptoms of HF, diet, exercise, medications, weight control, sex and HF, when to call the healthcare provider, how to deal with stress and depression, smoking and alcohol use, and lifestyle changes. Greater than 95% of the patients love the new education booklet. Patient comments include “very well written,” “easy to understand,” and “can keep notes in it.”
Adapting knowledge to the local context was also illustrated in case 3, where the hospital used the simulation laboratory to prepare nurses for the IHO study:
The nursing leadership team collaborated… with the HF care manager to develop creative ways in order to prepare nurses for their participation in this study. Many innovative strategies were used. Our educator utilized the department of education simulation laboratory to conduct several cases aimed at utilizing teach-back as an evidence-based method for teaching patients about HF. RNs completed this case in the simulation environment in order to better utilize the method of teach-back and increase their confidence and competence with educating their patients about key HF self-management information such as diet, medication knowledge, weight monitoring, when to call the doctor, and when to follow up. In addition to simulation cases and information about teach-back, the study protocol and HF education were posted in convenient locations on the unit so nurses would have access to it during their shift for review; that is, bulletin areas, medication room, and rest rooms (anywhere an RN may have a moment to read) were utilized. E-mail communication is a major form of information sharing at our hospital, and RNs received key study information electronically as well as a link to a video where teach-back is demonstrated.
Assessing Barriers to Knowledge Use
In the next phase of the KTA cycle, the implementers assess for barriers that may impede or limit uptake of knowledge so that strategies can be implemented to overcome or diminish their effect. One challenge experienced by many hospitals was how to engage and prepare staff for participation in the IHO study. Each hospital could determine whether all nurses on the unit were trained or just a targeted IHO team. All had to be made aware of the study and the expectations of patients participating. Some hospitals required attendance at an orientation session or required that certain videos be viewed.
Many hospitals examined the care processes and realized that they needed some new roles to implement the protocol. The IHO study team represented in case 4 developed a role for nurse navigator:
The HF navigator is an RN with both a strong clinical background and care management expertise. This vital role bridges the gap between inpatient management and self/family care at home. Establishment of a professional relationship between the nurse navigator and the patient/family is a key factor in success of the program.
A challenge to several IHO study hospitals was making a follow-up appointment within 2 weeks of discharge. Some hospitals assigned a specific nurse to make the appointments. A few hospitals shared that they were starting nurse practitioner–led HF clinics to ensure that patients were seen within 2 weeks. As revealed in case 5, a new role was developed for a flow nurse who is accountable to make the appointments:
A flow nurse is an RN who starts their shift at 11 AM with the sole intention of completing HF discharges. This unique staffing pattern allows a nurse to spend undivided time at the bedside to coordinate all aspects of care at time of discharge. The flow nurse addresses key points that have been discussed earlier but also makes cardiology follow-up appointments, assesses if the patient has a home scale in working order, and discusses the patients plan to obtain new prescriptions. This is a patient as well as a direct care staff nurse satisfier on many levels.
Select, Tailor, Implement Interventions
During this phase of the KTA cycle, the implementers plan and execute the interventions. This involves tailoring the intervention, addressing identified barriers and specific audiences. The previous section illustrates some of these tailoring approaches once barriers were identified. Several hospitals reported that there was not a consistent approach to teaching patients how to weigh themselves and how much of a weight gain is a sign of increasing fluid retention. Moreover, as nurses start asking patients how they weighed themselves, it became clear that some patients did not own scales. Even though they planned to purchase them after discharge, nurses learned in the 48-hour phone call that they had not obtained one yet. The hospital represented in case 6 addressed this barrier by working with the hospital foundation to provide scales before discharge:
A decision was made to seek funding from our hospital foundation to provide a scale to each patient who said they did not have one at home. We do this regardless of the patient’s ability to purchase their own scale. We believe it is essential for the patient to leave the hospital with a scale in hand.
One year after the completion of data collection at our hospital, we continue to give a scale to each HF patient who don’t have one at home.
Monitor Knowledge Use
During this phase, the implementers identify what constitutes knowledge use and then analyze the data to determine the extent of implementation. Both the individual patient-level data and the aggregate data collected for the IHO study provided relevant benchmarks as to the quality of HF care and the success of the implementation. Perhaps the most relevant data for the study sites were not the final patient outcomes, rather what study sites learned about their patients and their procedures while collecting patient data. During the patient screening phase, the case 7 IHO study team realized that a greater percentage of patients had cognitive deficits and that this had to be taken into account in their teaching:
First, we were surprised to find out during our patient screening process that almost one-third of our HF patients (28%) had to be excluded from the study because of documentation of a cognitive deficit in the medical record. Second, we were startled when some of the patients who were enrolled in the IHO study answered the same HF knowledge question incorrectly on the pretest and posttest, even after 1-on-1 instruction by cardiac rehab and dietitian experts, utilizing videos, discussion, and written materials.
Armed with data showing that patients were not learning, teach-back was initiated:
Before the IHO study, we had considered the teach-back method for patient instruction, but decided against it. After the IHO study, the nurses on the study team felt that it was imperative that we begin immediately in order to provide consistent, frequent reinforcement of the expert education sessions. They felt that this would benefit patients with cognitive deficits, those with low health literacy, and all patients who needed to hear our critical self-care messages more than once. An additional potential benefit of teach-back would be the process of identifying, documenting, and including the key learner. Once we recognized the volume of our HF patients with cognitive deficits, it became essential to figure out who the key learner and primary caregiver were and include them in our educational efforts.
As illustrated in case 8, quantifying patients’ knowledge deficits led nurses to realize how important a role they had in teaching patients:
The responses that some of our patients provided to these baseline questions clearly demonstrated the knowledge gap that exists not only in newly diagnosed patients with HF, but also in patients with longstanding HF. Just after this initial point of patient consent into the IHO study, these nurses observed the continued great need for patient education in tackling HF as a chronic condition. It was also at this point where these nurses saw the difference they can and do make in the level of success these patients experience after discharge from the hospital. Beyond the specifics of the particular study being conducted, nurses were charged with the great responsibility of reducing and correcting the knowledge deficit that exists in the HF patient population.
The relevance of hospital-specific and benchmark data to driving practice change was acknowledged in case 9:
After receiving preliminary IHO study site results, our study group met to discuss how our program could be further enhanced based on the findings. Because the prevalence of depression was so high, we identified the need for improved processes related to depression in this population, such as routine screening and a protocol for referrals. In addition, the team intends to incorporate a standard educational needs assessment so they can continue to provide individualized patient-centered education.
One aspect of the IHO protocol required nurses to estimate the actual minutes of teaching done by nurses and other disciplines. After reviewing the data, some hospitals readily recognized that there was overlap and gaps in the efforts across nurses, pharmacists, physicians, and other care providers. Nurses also appreciated the quantification of the teaching minutes because the amount of time it takes to teach is often not recognized.
In the evaluation phase of the KTA cycle, the implementers determine the impact of using the knowledge in terms of health, practitioner, and system outcomes. Although the primary study outcomes for IHO patients were readmission and self-care ability, one of the most relevant findings of the study was related to the frequency of depressive symptoms in the HF population. The study protocol required the administration of the Beck Depression Inventory8 because of the high prevalence of depressive symptoms in the HF population. Each hospital had to develop a procedure to evaluate patients when they had a score indicative of depression, so the nurses became increasingly aware of the frequency and severity of symptoms in their patients. Several IHO study hospitals noted that they have extended their depression screening practices to include patients with other chronic illnesses, whereas others incorporated a depression screen in their electronic health information system.
The IHO protocol required that the hospital database be checked for readmission 30 days after discharge. At times, the study nurses were surprised to learn that a patient had been readmitted to their hospital. One hospital set up a procedure to inform the discharge unit if a patient was readmitted within 30 days so a follow-up could be done to examine potential areas for improvement.
Several hospitals noted the benefits of the phone calls 48 hours after discharge as an effective intervention. They learned that some patients had not purchased scales, had medication discrepancies, and had misconceptions about their care. These calls were so revealing that some hospitals devoted more resources to institute routine follow-up calls for all HF patients.
Sustain Knowledge Use
The final stage in the KTA cycle, sustaining the knowledge use, is particularly challenging. Sustained knowledge use “refers to the continued implementation of innovations over time and depends on the ability of workers and organizations to adapt to change.”5(p165) This is likely through the use of feedback looping that cycle through the action phases. Although the IHO project was time limited, the case studies and shared stories provide evidence of the sustainability efforts. The case 10 IHO study team summarized next steps:
1. recruitment of a HF CNS to coordinate a team to focus on HF patient care in the hospital, preparation for discharge, and follow-up monitoring via discharge phone calls; this was accomplished as the need was validated during the study period;
2. ongoing education of all nurses caring for HF patients on the evaluation of postdischarge needs;
3. educating nursing staff to utilize a teach-back method of key HF content;
4. establishing an intensive coordination between our facility and skilled nursing facilities to assist with continuity of care; and
5. monitoring readmissions in order to identify further opportunities for improved care transitions.
Case 11 illustrates how aspects of the IHO study protocol have been or will be incorporated into 1 hospital’s routine care processes to sustain the knowledge use:
Besides an electronic physician discharge checklist that targets the “Get With the Guidelines for HF” variables, some of the HF predischarge questions utilized in the IHO study will likely be considered as part of the routine discharge for HF patients. The predischarge checklist creates a consciousness that provokes action by the patient and the provider in securing a safe transition to home. The collective participation in the education of the HF patient is showing glimmers of recognition for what it is, more than a core measure for reimbursement, but as integral as evidence-based medications for improving outcomes for our patients.
Many IHO study coordinators noted how the nurses’ perceptions were changed as a result of their participation in the IHO study. At the case 12 hospital, the site coordinators were profusely thanked for the 1-on-1 time that was required to implement the IHO protocol. This prompted the nursing staff to consider how patients perceive their interactions with nurses and the negative effects of multiple interruptions when providing care:
We found that the connection we made with the study subjects influenced their perceptions. Based on these initial results, we will look to our shared governance unit councils for ways to positively impact the connection made through the nurse-patient relationship. Working collaboratively, we could establish the patient’s room as a protective patient zone.
Furthermore, they reported a subsequent study to better explore the effects of disruptions:
To provide even greater evidence on the impact on safety of “projected patient time,” our hospital has engaged in a study of distractions and interruptions. With the addition of this knowledge, we will be in an ideal position to seek ways to enable the patient and nurse to make the vital connection and be able to communicate freely that will assist them in improving patient safety and well-being for life. This is what it means to care.
This article describes the consequences and lessons learned directly from the voices of nurses who participated in a multisite study. The case studies illustrate how the protocol was implemented in their organizations, lessons learned, and how it affected outcomes and sustainability. The monthly IHO synchronous Web-based meetings allowed participants to share their experiences, which undoubtedly influenced the implementation itself as sites experienced the same barriers and shared their approaches to tailor their interventions. The sharing also influenced the success of implementation and likely improved hospitals’ capacity to do future studies.
The KTA conceptual framework4,5 provides a useful way to look at the distinction between knowledge creation and knowledge application. The action cycle further illustrates the dynamic nature of the phases, recognizing that phases can influence each other and that sustainability likely requires a feedback loop that cycles back through the action phases. During the IHO study, implementation methods common to quality improvement were used including working as a team, promoting a quality culture, and data monitoring and feedback. Data were provided back to sites in aggregate form so an immediate feedback loop was established. In addition, many sites closely monitored their patient data as the study progressed. These activities reflect the monitoring knowledge use and evaluating outcomes phases. It is expected that some knowledge would be sustained, but the extent of the knowledge application and spread was remarkable as evidenced by the continued conversations with nurses at the study hospitals. These examples of spread are often not captured in research. Yet, they may hold the true potential for improving care.
Research protocols need to include methods to capture and codify how the protocols are implemented, including identified barriers, tailoring that is done, learning from study data, and efforts to sustain the new knowledge. Innovations that influence quality of care are often reported, but the implementation strategies are not codified. The case studies illustrate how nurses and interdisciplinary teams can be mobilized to achieve immediate and lasting changes to patient care.
While the hospital experiences presented here are reflective of the IHO participant experiences, they cannot possibly capture the breadth of lessons learned from participating in the study. The cases represent a convenience sample of those who responded to our invitation, and they may be more likely to have had a successful implementation. Data from monthly meetings were not systematically captured, which also reflected how hospitals adopted the protocol to their local context and the lessons learned. It should also be acknowledged that the hospitals were all Magnet® hospitals and self-selected for participation in the IHO study.
The examples presented here illustrate a learning healthcare system, where the lessons learned from both research and clinical practice are captured, analyzed, and translated into improved care. These examples demonstrate how hospitals narrowed the research-practice divide through feedback loops and accelerated innovation within their organizations. New approaches were generated to solve problems in a learning system in progress. Clinical data were used to continuously improve care processes. The IHO research study served as a method of educating health professionals, demonstrating a new clinical research paradigm in which clinical settings and research teams are partnered in a common vision to apply evidence to improve care.
The authors thank all the hospitals who participated in the IHO study. They also acknowledge those that submitted testimonies that support this article: Advocate Christ Medical Center (Oak Lawn, Illinois); East Jefferson General Hospital (Metairie, Louisiana); El Camino Hospital (Mountain View, California); Hudson Valley Hospital Center (Cortlandt Manor, New York); Middlesex Hospital (Middletown, Connecticut); Northwest Community Hospital, (Arlington Heights, Illinois); OSF Saint Anthony Medical Center (Rockford, Illinois); Penn State Milton S. Hershey Medical Center (Hershey, Pennsylvania); Rex Healthcare (Raleigh, North Carolina); St Francis Hospital (Roslyn, New York); University of Alabama at Birmingham (Birmingham, Alabama); and University of Pittsburgh Medical Center St Margaret (Pittsburgh, Pennsylvania).