Venous thromboembolism (VTE) in hospitalized patients has significant adverse effects on morbidity and mortality as well as excess health care costs.1 VTE prophylaxis is the number one preventive strategy to improve patient safety, and the Centers for Medicare and Medicaid Services has declared VTE to be a “never event.”2,3
Extensive, evidence-based guidelines exist for the prophylaxis of VTE, especially the American College of Chest Physicians guidelines, which have been updated on a regular basis since 1998.4 However, despite the well-described risks of VTE and the availability of evidence-based guidelines, adherence to these guidelines remains low.5
Multiple approaches to improve VTE prophylaxis exist, including both education and noneducation (system or operational) aids and strategies.6 Simple didactic education or passive dissemination of evidence-based guidelines alone is ineffective.7 Data clearly support the efficacy of multiple educational content delivery techniques.8 They show that the most effective strategies to improve physician performance are multiple and repetitive in nature.
Few data are available demonstrating an alignment of continuing education (CE) and quality improvement expertise in an academic teaching hospital to improve performance of the health care team. Even fewer data are available describing specific and multiple educational strategies developed for the entire health care team, including physicians, nurses, and pharmacists by CE and quality improvement.
The authors describe multiple educational and operational strategies, developed and implemented by an interprofessional and multidisciplinary team, aligning education with quality improvement, to decrease the incidence of VTE prophylaxis in hospitalized patients in an academic medical center. The authors also discuss challenges encountered, lessons learned, and recommendations to guide and assist other facilities and teams faced with similar issues.
Description of problem and strategy to address
The University of Kansas Hospital (KUH) is a 750-bed academic teaching hospital. In response to national quality and potential future financial imperatives, VTE data collected from KUH discharge data from January 2009 to November 2010 showed an increasing number and incidence of VTE as a secondary diagnosis in hospitalized patients. In winter 2010, senior hospital administrators formed an interprofessional, multidisciplinary team with the goal of decreasing the incidence of VTE by aligning CE with quality improvement to develop strategic, educational and system-based plans to ensure evidence-based management of VTE prophylaxis.
The 24 team members included physicians, nurses, and pharmacists (see Supplemental Digital Appendix 1, http://links.lww.com/ACADMED/A151) but because of the biweekly meeting schedule, attendance varied. Physician disciplines included hospitalists; intensivists; and vascular, trauma, and orthopedic surgeons. Resident physicians also participated. Nursing disciplines included nursing administration; education; quality and safety; clinical projects coordinator; and bedside nurses. Pharmacy was represented by the inpatient pharmacy director and a clinical pharmacist. Hospital staff representation included the quality outcomes data manager, quality improvement coordinator, and information technology. The Office of Continuing Medical Education (CME) was represented by the associate dean of CME and a senior project manager.
The work of the committee was conceptualized into three areas: assessment, implementation through education, and implementation of system strategies and tools (see Figure 1).
Assessment of the problem
To gain in-depth understanding of the causes or themes associated with VTE in hospitalized patients, we reviewed 261 charts of patients with a secondary diagnosis of VTE (i.e., those hospitalized patients who developed VTE during hospital admission) over approximately 12 months. Of those 261 charts, 66 (25%) were found to have appropriate prophylaxis in place with VTE related to the disease process, most often a malignancy. Of the remaining 195 charts, the most common theme associated with the diagnosis of secondary VTE was the presence of a peripherally inserted central catheter (PICC) (38%), coding issues (miscoded) (23%), no prophylaxis (11%), and lack of documentation to support that VTE was present on admission (POA) (8%) (see Supplemental Digital Appendix 2, http://links.lww.com/ACADMED/A151). Coding issues included those patients who were coded for VTE, which on chart review was not present, or patients with deep venous thrombosis POA but not properly coded as POA. These four causes resulted in 80% of all VTE, supporting the Pareto principle that a majority (80%) of problems are produced by a few key causes.9 Less common causes identified included subtherapeutic dosing, interruption, and delay in prophylaxis. We also assessed our PICC and mechanical compression device use as well as nursing unit workflows, including standard practices around order implementation and dosage timing.
Implementation of strategies
Education and system.
To develop the educational plan for the health care team, the committee first viewed a “menu” of evidence-based content delivery strategies and techniques as described by the American College of Chest Physicians and abstracted by the University of Kansas Office of CME (see Supplemental Digital Appendix 3, http://links.lww.com/ACADMED/A151).8 This menu described evidence-based instruction methods (e.g., particular educational procedures, educational materials, and strategies) that could be used to develop an educational plan.
Next, the committee completed a matrix of methods to change provider performance as described by the Pathman/PRECEDE (Predisposing, Reinforcing, and Enabling Constructs in Educational Diagnosis and Evaluation) stages (see Table 1).10,11 The matrix describes the types of interventions (predisposing, enabling, reinforcing) based on the learners’ stages of acceptance (awareness, agreement, adoption, and adherence). Interventions were assigned to a particular cell in the matrix by committee consensus. The broad educational plan, informed by the causes of VTE as delineated by the VTE chart review, was developed through this process of completing the matrix.
As both the presence of a PICC catheter as well as cases with no or subtherapeutic prophylaxis were common, educational strategies to improve awareness of the need for VTE prophylaxis as well as the risks of a PICC catheter were developed. A PICC/VTE subcommittee was formed to develop an algorithm for evidence-based PICC indications (see Supplemental Digital Appendix 4, http://links.lww.com/ACADMED/A151). A short (<5 minutes) podcast describing an overview of VTE risk, consequence, and prophylaxis was placed on the hospital intranet for viewing. An interprofessional, multidisciplinary, case-based patient safety conference with CE credit was given for physicians, nurses, and pharmacists. Over 300 professionals attended. The ability to offer CE credit was integral to committee decisions of all other broad educational and system strategies.
“Badge buddies” (pocket-sized reference tools tucked into staff badge holders) focused on appropriate VTE risk assessment, prophylaxis guidelines, and contraindications. These were developed from KUH guidelines and distributed to the entire medical and nursing staff, pharmacists, and all trainees (see Supplemental Digital Appendix 5, http://links.lww.com/ACADMED/A151). A 30-minute VTE education presentation was delivered during resident orientation by one of the authors (T.K.) to the entire new resident class.
A VTE link was created on the KUH Web site providing a “one-stop shop” for access to evidence-based, service-specific guidelines for VTE prophylaxis and treatment. The site included a variety of topics and guidelines that are beneficial to the multidisciplinary patient care team. Finally, the Web site housed other organization-specific resources, such as electronic health record ordering tips and patient education materials. Web-based materials were promoted in several ways, including a VTE-themed screen saver used throughout the hospital, a link on the KUH intranet, and all intra-KUH staff communications.
To improve and enable agreement as a method for improving performance, individual departmental VTE data were presented to clinical departments for discussion and review as listed (see Table 1). Chairs received department-specific VTE data monthly. Individual department meetings were scheduled depending on the department’s culture/interest and their specific data. The entire department, targeted physicians, and/or interprofessional teams (e.g., Trauma) were present for the meeting.
The committee developed “My KU VTE Prophylaxis,” an agreed-on approach to VTE prophylaxis for the entire medical staff which was approved by the hospital executive committee of the medical staff (see Supplemental Digital Appendix 6, http://links.lww.com/ACADMED/A151). Nursing unit data reviews were also conducted on each nursing unit.
Adoption and adherence.
To improve adoption and adherence, the pharmacy department implemented active patient VTE prophylaxis surveillance. Active surveillance included daily, patient-specific risk stratification and reprioritization of pharmacist workflow to focus on VTE prevention. This was accomplished by best practice alerts (BPAs) in the electronic health record, order set revisions, standards of practice, and VTE education meetings. The BPAs were activated by the user breaking the “rules” of the BPA (e.g., no prophylaxis ordered). However, the most active oversight strategy was hardwiring VTE risk stratification and documentation into the pharmacists’ daily workflow by implementation of an electronic clinical decision support tool (see Supplemental Digital Appendix 7, http://links.lww.com/ACADMED/A151). Pharmacists reviewed patient-specific information and discussed VTE prevention therapies with the multidisciplinary team.
Quality improvement and CE elements for VTE risk reduction were defined as follows, with multiple overlapping responsibilities for each profession:
* Nurses: patient risk assessment, prophylaxis administration, mobility, prophylaxis interruption avoidance, mechanical compression device compliance, appropriate PICC maintenance, continuing nursing education, and patient education
* Physicians: patient risk assessment, prophylaxis order, prophylaxis interruption avoidance, appropriate PICC needs assessment and use, CME, and patient education
* Pharmacists: risk stratification via pharmacy score card electronic medical record (see Supplemental Digital Appendix 7, http://links.lww.com/ACADMED/A151) medication and dose monitoring, prophylaxis interruption avoidance, pharmacy VTE education, continuing pharmacy education, and patient education
* Ancillary support: concurrent surveillance to ensure appropriate prophylaxis, physical therapy mobility, occupational therapy mobility, coding review, and organizational improvement
Insertion of PICCs dropped dramatically during the 16 months of study. Almost 360 PICCs were inserted in December 2010, dropping by almost half in April 2012 to less than 200 insertions. Hospital VTE rates dropped from 12.68 per 1,000 patients in December 2010 to 6.10 per 1,000 patients in June 2012. As demonstrated in Figure 2, the changes are greater than would be expected by chance alone and more than the random or background noise of the system.12
Our results show a significant decrease in the incidence of VTE in a large academic medical center after alignment of education with quality improvement through an interprofessional, multidisciplinary team effort. Both educational and system strategies associated with that improvement were developed and implemented by the team.
Previous work also describes improvement in VTE incidence with education and system changes.6 Some data are limited to pilot interventions; many are limited to specific units or departments, and many to only one or two interventions.13 We believe our work is important and unique in that we describe detailed, multiple, and extensive quality improvement interventions, shaped by evidence-based CE guidelines, and implemented over a prolonged period of time throughout the entire academic teaching hospital with the involvement of multiple professions and disciplines.
Our results are in agreement with evidence-based CME guidelines demonstrating that multiple instructional media and multiple exposures are most effective in implementing change.8 Our “menu” of potential educational venues was helpful to the committee not only to describe in a visual way the potential educational choices but also to align those choices with our chart review data to demonstrate causes or themes associated with VTE. Using the Pathman/PRECEDE matrix to align type of interventions with stages of learners’ acceptance was key in developing specific educational and system strategies.9,10
Aligning education with quality improvement has not been a traditional methodology in academic medicine.14 In 2006, however, the Accreditation Council of Continuing Medical Education developed new standards for accreditation of any CME program, requiring not just knowledge transfer but also performance improvement.15 In 2009, the Institute of Medicine delivered a stinging indictment of current CE efforts and proposed a new comprehensive vision of continuous interprofessional education.16 These drivers have resulted in new examination and reform of CME, including alternative educational content delivery techniques and new partnerships to facilitate performance improvement.17
Aligning education with quality improvement is essential. The Association of American Medical Colleges developed a national initiative, Aligning and Educating for Quality, to encourage collaboration between CE and quality improvement.18 Through consultation with senior clinical and university leaders, this initiative provided a stimulus and an invitation for collaboration between these two entities in our institution.
Our approach is also unique in that we provided CE credits for physicians, nurses, and pharmacists who participated in our educational components. We believe this is an important aspect of CE and acknowledge the importance of education for the entire interprofessional health care team.
Our team has met almost biweekly over the last 20 months. Our approach was complex and intensive, which may not be applicable to other academic institutions or community-based health care environments. However, we feel that the multiplicity of our approaches could allow for smaller systems to “pick and choose” what might be most applicable to them. Also, our almost two-year experience gives credibility to the validity of the approach.
During this two-year period, we have clearly identified challenges as well as lessons learned that allow us to suggest recommendations for others’ work. Challenges have included the huge amount of time and resources necessary for such a detailed and lengthy process. As an academic medical center, we have multiple clinical departments, divisions, and service lines whose level of confidence in the data has been challenging. Reaching out through various education and system initiatives is vital.
We have learned that this process is challenging and labor-intensive. Leadership commitment is essential. Individual hospital/system data are mandatory in our view, as one simply cannot make assumptions about causes of VTE without data to support theories. Overall champions are necessary, especially physician champions. Interprofessional education cannot take place without an interprofessional team in place. And lastly, the expertise of CE providers in describing, defining, and supporting educational content delivery is essential. We felt that offering CE credits for the entire health care team was also a driver in improvement.
In summary, we demonstrated a significant decrease in the incidence of VTE in a large academic medical center through extensive educational and system initiatives implemented by an interprofessional health care team. Aligning education with quality improvement driven by an interprofessional health care team using a comprehensive, multifaceted strategy produces positive results.