In 2000, Drs Hill and Kerber, then of the University of Iowa Fellowship Training Program, published a manuscript exploring the state of specialty training in cardiology both nationally and at their institution.1 They described a comprehensive self-evaluation of their program and their approach to managing the many challenges faced by academic medical centers and cardiology faculties when training fellows. In an environment of diminishing economic resources, more demands on faculty time, and an exponential increase in medicine’s complexity, they identified eight key areas of focus and outlined their strategy for each:
- clarify the mission of the training program;
- inspire fellows to consider academics;
- revitalize interactions between faculty and fellows;
- maximize effectiveness of didactic teaching;
- optimize on-the-job training of fellows;
- provide adequate orientation of incoming fellows;
- take advantage of the Internet as a learning tool; and
- maintain a critical mass of fellows in an era of decreasing trainee positions.
While the pressures on academic medicine remain relatively consistent 10 years later, the overall healthcare landscape has undergone additional major changes since 2000.
Within the cardiovascular medicine fellowship at The Ohio State University Medical Center, we now find ourselves in a similar position to that of Drs Hill and Kerber 10 years ago. In reflecting on the state of our educational programs in 2010, we noted that it would be necessary to reiterate the context in which cardiovascular medicine training now exists. Plus, while still works-in-progress, our strategies for navigating these challenges might prove instructional for other institutions. Our review centered on four major topic areas: revisions in cardiology training requirements issued by the American College of Cardiology (ACC); the Accreditation Council for Graduate Medical Education (ACGME) Outcome Project; trainee duty hour restrictions; and the era of performance measures and quality metrics.
Changes in Cardiology-specific Training Requirements
The first ACC Guidelines for Training in Adult Cardiovascular Medicine (COCATS) were published in 1995 with task forces and requirements in 10 specific areas of cardiology.2 Since that initial document, general training in cardiology has become increasingly dense and complex. Following Drs Hill and Kerber’s initial review in 2000, there have been two revisions required to update competency requirements for procedures and rotational exposure. The 2002 COCATS 2 statement included the formation of two new task forces, in vascular medicine and peripheral catheter-based intervention, in addition to cardiovascular magnetic resonance (CMR). With the anticipated publication of the Multicenter automatic defibrillator implantation trial II (MADIT II) that same year,3 a new emphasis was placed on management of implantable cardioverter-defibrillators (ICD), as the post-myocardial infarction indication was likely to substantially increase the number of implanted devices. Recommendations were also made to spend at least one of a fellow’s nine required inpatient ward rotations on a dedicated heart failure/transplant service. Vascular medicine emerged as a new two-month core requirement, while the length of required clinical training remained fixed at 24 months.
A focused update was released in 2006 with an emphasis on advanced cardiovascular imaging and electrophysiology (EP).4 The first task force addressing cardiovascular computed tomography (CCT) was included, reflecting the emergence of this technology as a valuable tool for diagnosing obstructive coronary artery disease. Two years later, the most recent COCATS 3 document was released, with a new total of 13 task forces and the first mention of a new term—multimodality imaging—acknowledging the rapid growth in modalities beyond the original tools of electrocardiography, angiography, nuclear cardiology, and echocardiography.5 This concept’s importance was solidified by the subsequent publication of the ACC Foundation Training Statement on Multimodality Noninvasive Cardiovascular Imaging in 2008.6 Major revisions in the traditional siloed single-modality imaging (nuclear cardiology, echocardiography, CMR, and CCT) curricula familiar to most fellows-in-training were suggested, with an emphasis on patient-centered integration and flexibility in the number of procedures and months required. Programs were encouraged to combine curricula from each modality into three topics—imaging physics, general imaging aspects, and specific imaging applications. However, both three-year and advanced imaging subspecialty fellowship graduates with multimodality imaging expertise still face a fragmented system of board certification and an uncertain path to academic practice.7
Overall, these changes reflect both the evolution of technology and the complexity of medicine in general As such, they are clearly necessary. However, they present challenges to many programs with fewer resources or less infrastructure to adequately train fellows in some of the less common areas of cardiology—CMR, CCT, peripheral imaging and interventions, and cardiac transplantation. Programs are encouraged to collaborate with experts at their own or other institutions to ensure at least Level 1 exposure. The definition of a ‘general cardiologist’ has also become a moving target; with increasing requirements for minimum comprehensive training expanding to fill more of the 36 months of fellowship, there is less time for elective rotations. Careful consideration has to be given when structuring training programs to be sure that fellows achieve adequate clinical competency in basic cardiology in an environment where technology and mandated procedural volumes are a constant distraction.
Our Approach to Managing this Challenge— Leveraging Expertise of Faculty and Staff
With these changes in training requirements specific to peripheral vascular disease, ICD management, heart transplantation, and multimodality imaging, we have had to make several adjustments to our fellowship program. Overall, we have the luxury of having faculty and clinical volume that we have been able to leverage in each of these areas. Together with their input, we have made changes that permeate fellow education at multiple levels.
During their vascular medicine rotations, trainees interact with fellowship-trained, board-certified practitioners, while seeing patients and interpreting diagnostic testing ranging from ultrasound to magnetic resonance imaging. This is paired with a didactic curriculum that alternates case-based and topic-orientated vascular medicine topics. We have also worked to alter our EP rotation from laboratory-based to a more consult-driven service, realizing that most of our graduates will not be pursuing additional EP training. Time in the laboratory is then more focused on interpretation of intracardiac electrocardiograms (ECGs) for patients who have been seen the previous day when rounding on the floor. Trainees that indicate an advanced interest in EP may spend additional elective time with a heavy emphasis on laboratory-based diagnosis, therapy, and device implantation while also learning from pacemaker clinic staff as they focus on post-implantation device management. We feel that this approach is more suited to today’s general cardiology practice, in contrast to the previous paradigm which placed a high priority on the placement of temporary pacemakers, for example.8 The expansion of implantable devices in heart failure also provides additional opportunities for training9 and education of heart failure/transplant fellows, in addition to EP fellows alone. The necessity and practicality of such training is exemplified by some of our faculty within the section of heart failure/transplantation who have joint training in ICD and cardiac resynchronization therapy-defibrillator (CRT-D) implantation and management. Although this type of integration has been beneficial for trainees, it simultaneously presents some paradoxical challenges for our business model and faculty coverage, which tend to be organized by subspecialty ‘silos.’ In regard to training in multimodality imaging, we are again fortunate in having all four imaging modalities managed by our division, thus making faculty with multimodality expertise available to fellows. In previous years, morning conferences focused on individual modalities in isolation. As faculty members with credentials and board certification in multiple areas of imaging have joined the faculty, we positioned ourselves well to make adjustments to our conference curriculum even before the release of the 2008 ACCF Training Statement. These adjustments resulted in more patient-centered, integrated case presentations that allow discussion of the various strengths and weaknesses of each imaging method. Faculty responsibilities have also been revised, centering those with multimodality imaging backgrounds as primary rotators on the inpatient cardiology service and in the imaging laboratories, maximizing their abilities to triage imaging and educate both faculty and fellows on the most appropriate and efficient non-invasive testing for a specific clinical question.
We have also gone through a period of intense self-reflection regarding our programmatic mission and strengths. Our fellowship program has traditionally provided strong clinical training with less of an emphasis on producing independent investigators. This approach has been influenced by many factors within the medical center and department dating as far back as the late 1960s. Since then, the size and subspecialization of our division, fellowship structure, and number of available faculty mentors have been continually evolving. The expertise and varied backgrounds of our fellowship committee have become more important than ever, as we have struggled to manage the increasing complexity of training while making sure fundamental skills of ECG interpretation and physical examination remain a priority.
Changes in Graduate Medical Education Requirements
“What we measure, we tend to improve.” David C Leach, MD, Former Chief Executive Officer, Accreditation Council for Graduate Medical Education
The ACGME, which oversees trainees in over 8,300 training programs in the US—including nearly 3,000 general cardiology and subspecialty positions, altered the landscape of graduate medical training with the endorsement of release of the ACGME Outcome Project in 1999.10 The core principle was to change the emphasis of training programs to assessing actual educational accomplishments versus a program’s ‘potential’ to educate. The project also marked the launch of the now ubiquitous six general competencies: patient care, medical knowledge, professionalism, systems-based practice, practice-based learning and improvement, and interpersonal and communication skills. Programs were expected to be compliant with integrating these changes by July, 2002. Aside from the obvious time required to implement this required reorganization, there was an additional mandate for data acquisition and quality improvement. Program directors were now responsible for obtaining trainee performance evaluations from multiple sources beyond faculty alone—nurses, technologists, patients, and themselves (in the form of self-assessments)—as well as structured assessment of clinical knowledge and procedural competency in the form of in-training examinations and proctored practical testing, respectively. Demonstrable efforts to analyze these data and show action designed to improve educational goals are now part of each program’s expected portfolio when visited by the designated residency review committee (RRC) for a site review. The Outcome Project was then followed by new requirements for trainee duty hours, restricting them to 80 hours per week.11 This change has had a massive ripple effect, forever changing the way all inpatient call schedules have been organized since its introduction. Further adjustments were made after the Institute of Medicine (IOM) released a five-year follow-up report,12 containing recommendations that were endorsed to a large degree by the ACGME.13 These both acknowledged the burden of implementing the 2002 recommendations, while also further restricting trainees in terms of consecutive hours worked.
The majority of these changes in graduate medical education have weighed most heavily on residency programs, where inpatient call coverage and team handoffs have become infinitely more complex. Cardiology fellowships, most of which use an at-home call system, have only had to manage the 80-hour work week. Mandates stemming from the ACGME Outcome Project, however, have had a large impact on cardiology training. The controversy was addressed in detail by current and former fellowship program directors at Tufts Medical Center and Duke University Medical Center in two separate commentary papers published in 2008.14,15 Touted positive effects included the project’s guidelines which allow for standardization among programs, multi-source feedback (MSF), providing a more comprehensive analysis of fellow performance, and the requirement for academic productivity from identified key faculty. Negative effects were dominated by demands placed on the time of trainees and faculty. With more restricted hours, shift coverage, and increased handoffs for interns and residents, fellows and faculty have had to fill in the gaps in patient care. Perhaps most important was the acknowledgment that program directors have begun to ‘drown in data,’ where time spent acquiring and analyzing information is time spent away from the actual work of educating trainees. Revisions to all curricula, organization and implementation of MSF across the program, documentation of continuous professional development for key faculty, and exploring methods of assessing clinical competency beyond procedure numbers alone have resulted in an explosion of paperwork and corresponding time commitment, stressing the capacity of program directors and coordinators to manage the information.
Our Approach to Managing this Challenge— Targeted, Collaborative, Data-driven Educational Reform
Aside from the commentary pieces referenced, there are very few data describing educational methods among cardiology fellowship training programs. A PubMed search using the terms ‘cardiology fellowship’ and ‘education’ yields 104 articles, the vast majority of which are either similar commentary papers or panel consensus statements on number of months/procedures required for clinical competence in a specific area of cardiology. None of them describe an educational intervention studied in rigorous fashion. The key for our fellowship program when weathering the educational improvement mandates of the Outcome Project has been collaboration. Our divisional faculty and program leadership represent somewhat of a paradox, probably similar to other academic divisions in the US—highly skilled clinicians and nationally recognized content experts with largely minimal training in adult learning, curriculum development, and educational research. Simply put, while we are in support of data tracking and analysis to make targeted, efficient changes to improve how we teach, we need help from educational experts.
With the institutional license for the TurningPoint® Audience Response System (Turning Technologies, Youngstown, OH) purchased by the medical center, our fellowship program has full access to the hardware and software for interactive question-and-answer activities during conferences. We have employed this every morning during conferences, where the first 10 minutes of the hour have been designated for board review, with a single ECG and multiple-choice question (MCQ) posed by the chief fellows and answered by the core fellows. Anecdotally, the practice of using TurningPoint has improved conference attendance and fellow satisfaction with board review. We also host visiting residents and fellows during morning conferences and this provides an anonymous forum for them to participate without concern for being called upon in an unfamiliar environment.
When re-evaluating our ECG interpretation curriculum, it became clear that there were several approaches to managing the challenges posed by fellows having to acquire two separate sets of skills: patient-centered interpretation, which often involves subtleties and ambiguity; and facility with the American Board of Internal Medicine (ABIM) answer sheet used during the board certification examination, which comprises very straightforward coded options but is non-intuitive and may cloud measurement of content knowledge. As the board examination for cardiovascular disease is split between MCQs and ECGs and is non-compensatory—a candidate can achieve a 100 % score on the MCQs and still fail the overall examination if they fail the ECG section—there is a major emphasis placed on competent ECG interpretation skill. Because there was a lack of data to guide our revisions, we partnered with the ACC to survey fellowship program directors in the US about how they taught ECG interpretation and used the ABIM answer sheet in their curriculum. The results showed that most programs (92 %) used the ABIM answer sheet when teaching their fellows, but only 42 % performed formal testing of skills.16 The information gained from this joint venture with the College led to our use of a proctored ECG test using the ABIM answer sheet under timed conditions. This is given twice annually and has been received positively by the fellows, who note that the board simulation conditions give them better insight into time management during test taking. We hope to review data on certification examination performance after the intervention, to see if scores demonstrate statistical improvement.
A relatively recent innovative effort launched within The Ohio State University College of Medicine, the Center for Education and Scholarship (CES), has also served as a valuable resource for educators within the medical center (http://medicine.osu.edu/faculty/ces/pages/index.aspx). Its stated goals are to:
- provide consultation on all aspects of medical education;
- promote and develop skills that enhance the quality of health sciences education:
- assist in the development of outstanding teachers, curriculum developers, program leaders, and educational scholars;
- provide guidance to medical educators in turning their teaching activity into scholarly presentations and products;
- provide guidance for educational recognition and rewards; and
- assist in the identification and/or development of guidelines to recognize excellence in educational scholarship and scholarly activity.
On a yearly basis, they select participants in the Faculty Teaching Scholars Program, an 18-month course designed “to address the professional development needs of faculty who are committed to a significant career investment in health sciences education.” Graduates have included clinicians from all medical disciplines as well as PhDs from within the College of Medicine.
With the help of CES experts, we have nearly completed an analysis of predictors of fellowship graduate board certification examination performance when weighing the contribution of morning conference attendance, faculty evaluations during clinical rotations, and in-training ECG interpretation test scores. They have collaborated to aid in efficiently tracking and labeling data in accordance with Institutional Review Board certifications and have helped facilitate conversations with psychometricians at the ABIM about how best to proceed with proper statistical analysis. This process has already identified additional areas for future investigation to improve our faculty evaluations of fellow performance to save valuable time while hopefully providing better information. Benefits from this type of collaboration extend beyond the relationship and data-driven educational reform. Co-authors of anticipated manuscripts are able to demonstrate academic productivity that engages them in the peer-review process while satisfying promotion criteria.
The Dawn of the Quality Improvement and Safety Era
“At the very least, quality improvement has little chance of success in health care organizations without the understanding, the participation, and in many cases the leadership of individual doctors.” Donald M Berwick, MD, MPP, Former President and Chief Executive Officer of the Institute for Healthcare Improvement
The 1999 release of the landmark IOM report To Err is Human: Building a Safer Health System arguably launched the current era of quality healthcare and patient safety. In it, authors estimated that as many as 98,000 people die in hospitals each year as the result of preventable medical errors, resulting in total costs between $17 and $29 billion nationwide. These were considered results of faulty systems, processes, and conditions throughout healthcare institutions and facilities. Proposed actions included increasing the national focus on safety, encouraging voluntary reporting of statistics, raising performance standards, and creating a culture of safety at the level of healthcare delivery.
From a cardiovascular standpoint, national organizations have recognized the need to define standards for quality and safety, while creating data registries and templates for medical centers and practices to participate. The ACC has established several specific quality and safety programs: Imaging in Focus, the Practice Innovation and Clinical Excellence (PINNACLE) Network, Hospital to Home, the Door-to-Balloon Alliance, and the Guidelines in Practice programs, while the American Heart Association (AHA) offers their series of Get With the Guidelines toolkits and the newly established journal Circulation: Cardiovascular Quality and Outcomes. These two entities have also joined forces to release statements regarding performance measures and quality metrics, data elements, and definitions to track outcomes and performance measures for common cardiovascular disease states.17 It is clear that these efforts are necessary, as the cardiology community continues to face challenges to providing safe, evidence-based care. A recent review of the scientific evidence underlying the 53 guidelines issued by the ACC/AHA from 1984 to 2008 found that only 19 % of Class I recommendations are based on evidence from multiple randomized trials or meta-analyses.18 The release of the ACC Foundation’s Appropriateness Criteria documents (www.cardiosource.com) has provided a framework for the utilization of diagnostic (transthoracic, transesophageal, and stress echocardiography, radionuclide imaging, CCT, and CMR) and therapeutic (coronary revascularization) modalities, but patients are still exposed to large amounts of radiation during diagnosis19–21 and management.22 Even with the many diagnostic tools at our disposal, our clinical triage needs improvement. Data from the National Cardiovascular Data Registry (NCDR) reveal the low diagnostic yield of elective coronary angiography when patients are referred for suspicion of angina, with only 39.2 % of referred patients exhibiting significant obstructive coronary disease.23 Finally, in the subset of patients undergoing percutaneous coronary intervention who are at the highest risk of bleeding complications, registry data demonstrate that only 14.4 % received bleeding avoidance strategies such as vascular closure devices and use of bivalirudin24—a clear opportunity for improvement.
Our Approach to Managing this Challenge— Teaching the Language and Concepts of Quality Improvement and Patient Safety
As directors of cardiology training programs, we are obligated to position our graduates for clinical/academic practice and board eligibility. Many of us probably lack sufficient expertise to provide sophisticated training in the specific area of quality improvement and patient safety. There are no clear guidelines other than the existing ACC statements described above, and a review of publications describing effective teaching of these concepts to trainees shows that data are limited to medical students and resident-level physicians.25 As such, many fellowship programs may have concluded, as we have, that these efforts to educate and improve clinical outcomes are best addressed at a department,26 hospital, or medical center level. Under the auspices of the University Medical Center’s Quality and Patient Safety Office, the Evidence-based Practice Policy Group has produced practice resources for common cardiovascular diseases (acute coronary syndromes, heart failure, atrial fibrillation, infective endocarditis prophylaxis). These incorporate the ACC/AHA guidelines and ACC quality measures into a single document, available on the hospital intranet. When referenced by fellows on the wards or in clinics, they serve as a quick reference for both treatment and documentation. As a complement, fellowship leadership have partnered with the administration of the Richard M Ross Heart Hospital, our free-standing 150-bed facility containing all of our diagnostic and treatment laboratories, to provide a monthly morbidity and mortality conference focused on quality and safety. This is a modification of the traditional ‘M and M’ structure where biopsy or post-mortem data are used to refine diagnostic skill, focusing instead on guideline-based decision-making and opportunities to learn from complex clinical situations that led to suboptimal outcomes. Fellows also receive a yearly lecture, as part of their core curriculum, that is focused on the basic structure and current state of the healthcare system in the US.
For those with specific interest, our Office of Graduate Medical Education has recently created a pilot four-week rotation—a ‘Quality Elective’—where house staff are immersed in a combination of self-driven reading, participation in medical center committee work, and completion of a targeted quality improvement project focusing on error investigation and root-cause analysis.
These are only rudimentary first steps, but the field itself is continually evolving and the attainment of ‘quality’ care is a moving target. Most important is to provide our fellows with the skills to become lifelong learners in their clinical practice with quality and safety in context—“to do their work and to improve it.”27 In the coming academic year we hope to develop a curricular framework for fellows to undertake their own small-scale quality improvement projects as a learning tool, probably enrolling in and using components of the ACC’s NCDR—Acute Coronary Treatment and Intervention Outcomes Network (ACTION), Carotid Artery Revascularization and Endarterectomy (CARE), CathPCI (diagnostic catheterizations and percutaneous coronary interventions), ICD, Improving Pediatric and Adult Congenital Treatment (IMPACT), and PINNACLE (www.ncdr.com). Conversely, a parallel priority is to educate them about the disadvantages and shortcomings of quality/performance measures and future implications in their practice. While well-intended, the phenomenon of ‘pay for performance’ has also been found to have unintended negative consequences in the state of Massachusetts, as an example. At the least, physicians are finding themselves being relegated to lower tiers of insurance reimbursements due to an opaque system rating their care by performance measures and, what is most concerning, creating an environment where physicians shy away from complex cases to avoid being associated with poor outcomes.28
All of these issues will continue to evolve since the passage of the Patient Protection and Affordable Care Act in March, 2010, making familiarity with terms and concepts highly important to graduating cardiology fellows preparing to enter practice.
Conclusions from Our Fellowship Training Program
Clearly, the training of future specialists in cardiology, as with all of medicine, becomes increasingly complex with each passing year. At The Ohio State University Medical Center, much of our recent focus has been adapting to increasingly complex training requirements from the ACC, managing the demands imposed by regulations issued by the ACGME, and providing fellows with an adequate introduction to quality improvement and patient safety concepts. For program leadership, fellowship committee members, and key faculty, these efforts have been an education among themselves through collaborative efforts and continual dialog with our Department of Internal Medicine, Office of Graduate Medical Education, and College of Medicine educational experts. As we have worked to rise to these challenges within our own program, however, there remain other, larger concerns on the horizon.
Future Challenges and Opportunities for Cardiovascular Educators—‘Medical Education as Translational Science’
There exists a great opportunity to combine medical education research and clinical training to improve patient safety.29 The use of pre-procedural checklists has already been shown to reduce catheter-related bloodstream infections in the medical intensive care unit setting30 and death and surgical complications in international operating rooms.31 These examples, and others like them, were implemented by medical centers and carried out by multidisciplinary teams of practicing physicians, nurses, and technicians in the specific clinical settings. Checklists tend to focus on technique and equipment, but also place just as large an emphasis on a team-based approach to staff communication. In fact, simple training in the use of ‘crew resource management’ techniques—rules of conduct derived from the aviation industry which are designed to create an environment where communication is prioritized above positional hierarchy—was recently shown to improve surgical mortality within 74 hospitals in the Veterans Health Administration.32 Expanded considerations of checklists have also included utilization as an alternative to relying on intuition and memory in clinical problem-solving, with the hope of reducing diagnostic error.33 Published data have described methods of ‘deliberate practice’—focused, repetitive practice with testing to ensure achievement of a mastery standard34—when instructing medical residents during simulation workshops in advanced cardiac life support and central venous catheter insertion.29 Translated to cardiology fellowship, there are ample opportunities to model educational interventions to improve patient satisfaction and sedation/anesthetic-related morbidity during transesophageal echocardiography and reduce fluoroscopy times, contrast volumes and complications during cardiac catheterization. Fellows perform hundreds of these procedures over the course of their training with unknown variability in realtime supervision and teaching. As such, a fellow’s ‘learning edge’ of procedural or medical knowledge may or may not be routinely explored. Therefore, despite the volumes listed in fellow logbooks, faculty preceptors and fellowship directors mark trainee progress with a combination of partial data and assumptions.
The Elusive Definition of Clinical Competence
The next frontier of medical education in cardiology will probably be the redefinition of clinical competence and support of lifelong learning. A review of studies relating medical knowledge and healthcare quality to years in practice suggests that physicians who have been in practice longer may be at risk of providing lower-quality care.35 To address this issue during fellowship training, program directors will have to be able to help trainees achieve core knowledge (know what they should know), identify knowledge gaps (knowing what they don’t know) encountered during a typical day, while providing the framework for them to ‘close’ them, and translate published evidence into practice.36 For fellows, the first step toward this goal will be the upcoming ACC In-Training Exam (ITE) in fall 2011. Written by educational experts within the college with the collaboration of the National Board of Medical Examiners, the ACC ITE will be a multiple-choice format examination testing comprehensive medical knowledge in cardiology. Results will help program directors provide formative assessments for each specific fellow and the program as a whole in comparison with others. In parallel to an improved assessment of medical knowledge, the ACC hasbegun to explore medical simulation in partnership with the ABIM as a potential means of testing procedural skills in interventional cardiology.37 Such simulators already exist for transesophageal echocardiography and pacemaker implantation, and may serve as methods to learn without patient exposure to risk at some point in the future. Whether they will reach the point of near-universal use (they are quite expensive) during fellowship training or become adopted as a final arbiter of procedural competence during a board certification exercise is certainly possible. For now, fellowship directors must continue to adapt to dynamic requirements, seek opportunities to collaborate, and persevere among a storm of data—all while never forgetting to trust their gut.