Author + information
- aCAPITAL Research Group, Division of Cardiology, Department of Medicine, University of Ottawa Heart Institute, Ottawa, Ontario, Canada
- bDepartment of Radiology, University of Ottawa and The Ottawa Hospital, Ottawa, Ontario, Canada
- cDepartment of Cellular and Molecular Medicine, University of Ottawa, Ottawa, Ontario, Canada
- ↵∗Address for correspondence:
Dr. Jeffrey A. Marbach, Division of Cardiology, The University of Ottawa Heart Institute, 40 Ruskin Street, Ottawa, Ontario K1Y 4W7, Canada.
Scholastic development of fellows-in-training (FIT) has been a recurrent theme in the Journal’s Fellows-in-Training & Early Career Page (1,2). Although the importance of establishing mentorship and the benefits of peer collaboration have been emphasized, the role of the fellowship program as the nidus for both research training and production of meaningful research outputs has yet to be established (1–3).
Typically, in the form of unstructured protected research rotations, fellowship programs and the Accreditation Council for Graduate Medical Education alike have emphasized exposure to research as a core requirement (4,5). However, research training is largely devoid of structure and dependent on the mentor identified, leading to highly variable training, even among trainees within a single program. Accordingly, fellows are graduating with inconsistent training in research methodologies and potentially beginning their careers without achieving the goals set out by accreditation guidelines (6).
The design, execution, and dissemination of research findings through publication represents the application of numerous learned processes akin to other skills developed during cardiology fellowship. For example, the development of longitudinal echocardiography curriculums with standardized performance measures ensures minimal standards in skills acquired for graduation (7). Similarly, one could argue that standardized training requirements in research could ensure that FITs achieve minimum competence to meet accreditation goals. Graduating with a standard skill set may then increase the probability of generating a significant contribution to the development of clinical/basic sciences, education, or performance improvement.
To address this perceived deficiency, our program has developed a structured longitudinal research curriculum during the 3 years of cardiology fellowship training. The curriculum was designed in 3 stages: first-year fellows undertake a 4-week didactic research rotation with lectures in study design, basic statistical analyses, literature searches, research regulations, local research board processes, and obtain research ethics and good clinical practice certifications. In addition, fellows meet with several potential mentors to identify projects of interest and develop a hypothesis-driven research question to form the foundation of their longitudinal project. To avoid affecting curricular requirements, this research block counted toward the fellows’ total elective time, and 1 mandatory clinical rotation was shifted into the third-year schedule. The second phase of the curriculum is comprised of research rotations (up to 6 1-month blocks), which require iterative objectives/goals and submission of a written progress report evaluated by a rotation supervisor independent of the project to ensure appropriate progress. The third stage requires fellows to demonstrate competency by presenting their project at a national or international conference and submitting the work for publication at a peer-reviewed journal. Formalization of the research curriculum ensures standardized research training, structured goal-oriented rotations, and formal evaluation.
As part of the evaluation of our curriculum, we sought to determine whether our structured program affected competency (as appraised by research productivity). Accordingly, we evaluated the in-training and early career publication records of graduates assessing both fellowship and early career research productivity. This was accomplished through a retrospective cohort analysis of the PubMed and Scopus databases for publications authored by former fellows who graduated within a 10-year period between 2006 and 2016. The formal research curriculum was instituted in 2012; therefore, the research productivity of fellows who completed the formal curriculum was compared with the cohort of historical control subjects who did not. Following data collection, the mean annual publication outputs were determined for each fellow. These data were also substratified into the mean annual publications during fellowship, and during early career (defined as up to 2 years after fellowship graduation). Mean publication outputs were then compared using an independent Student's t-test in version 24 of SPSS (IBM, Armonk, New York).
In total, 41 trainees graduated from our program over the ascribed period, with 14 trainees completing the structured research curriculum since its inception in 2012. Among fellows who received the curriculum, the mean publications per year was 2.57 compared with 0.64 for the historical control—a difference that existed both during fellowship and early career (Figure 1A) (p < 0.001). Moreover, in addition to the increase in total publications, research training increased the number of first or senior author publications 4-fold from an average of 0.29 to 1.27/year (Figure 1B) (p = 0.02).
Perhaps as expected, providing FITs with structured research training and formal evaluation was associated with a dramatic increase in research output during fellowship and early career. Not only did trainees increase their total publication volume through increased involvement in ongoing projects, but a corollary increase in first/senior author publications suggests that FIT-initiated projects were increasingly being produced. Finally, by establishing ongoing research productivity beyond fellowship and into the trainee’s early career, our results suggest that the increased interest in research is not simply a phenomenon of training requirements but may also extend into early practice.
Although the results we present are encouraging, several important limitations must be noted. First, due to the observational study design, we are unable to establish a causal relationship between the research curriculum and the increase in research productivity, only that the relationship exists. Similarly, as there was no prospective randomization of fellows, the possibility that unmeasured confounding variables affected the research productivity of our trainees, irrespective of the research curriculum, must be acknowledged. Finally, our results come from a small cohort of fellows in a relatively short time period. The true effect of our curriculum, and the ultimate research productivity of our graduates, cannot be fully ascertained with only a few years of follow-up. Rather, the career-long achievements of our fellows will represent the ultimate marker of the success of our program, and these will not be evaluable for years.
The results from our program demonstrate that a formal research curriculum can be implemented into a cardiology fellowship program without the need for increased training duration or a substantial detraction from other required rotations. Through the implementation of a formal curriculum into research rotations that already exist as a requirement in fellowship programs, we have been able to achieve meaningful gains in research productivity—a marker of competency—without placing undue barriers on the program or fellows.
Forming strategies that include establishing mentorship and developing collaboration among peers is crucial to excelling at research during fellowship. Our center’s unique experience—through the development of a structured research curriculum—has established that competence can be increased across an entire training program. Assessment of research output suggests that participation in a formalized research curriculum results in augmented total publication counts and publication of first/senior author papers, both during fellowship and in early clinical practice. Akin to the many skills that FITs acquire in training, research is a skill that must be systematically taught and evaluated to ensure competence. We contend that structured research training should become an integral component of cardiovascular fellowship programs worldwide.
- Michael S. Lauer, MD (, )
- P. Kay Lund, PhD and
- Sherry Mills, MD, MPH
RESPONSE: Working to Save the “Endangered Species” of Physician Scientists
It has long been said that the physician scientist is an “endangered species.” This is not simply an impression. At the National Institutes of Health (NIH), we have seen that over time fewer physicians—and especially early-career physicians—are applying for and receiving research project grants (1). About 3 years ago, the NIH Advisory Committee to the Director issued a report (2) that confirmed the problem and offered the agency 9 specific recommendations: strong support of the MD-PhD Medical Scientist Training Program; greater support for individual fellowship awards; steps to address the gap in research project grant funding rates; adoption of rigorous tools to assess the biomedical research workforce; physician-scientist–specific grant mechanisms; expansion of the loan repayment program; pilot programs to test novel approaches to “improve and/or shorten research training for physician-scientists”; increased efforts to increase the diversity of the physician-scientist workforce; and leveraging of the resources available through Clinical and Translational Science Awards. The agency has taken steps to address all 9 recommendations (3).
Over the past 2 years, the NIH has focused on considering and developing pilot programs to test novel approaches, including embedding research into residency (and fellowship) training. We held 3 workshops with leading physician scientists, academics, trainees, and representatives from medical examination boards and specialty societies. We heard 2 clear messages that echo the careful work and sentiments offered by Marbach and colleagues in their intriguing and encouraging essay. First, embedding research training into residency and fellowship may increase the likelihood that physicians will pursue serious research (4). Second, research training, just like clinical training, entails exposure to and mastery of certain core competencies. Just as we expect trained cardiologists to know how to perform and interpret a Doppler echocardiogram, we should also expect trained physician scientists to know the fundamentals of experimental design—to know, for example, that it is not acceptable to simply ignore missing data.
Marbach and colleagues describe what sounds like a rigorous clinical research curriculum embedded within a fellowship program. They find that fellows who completed the program published more peer-reviewed papers during and shortly after fellowship. They also describe the limitations of their work, including the observational nature of their data, relatively small study sample, and need for longer-term evaluation of outcomes in additional cohorts.
The NIH is now taking the concept of “research-residency (including fellowship)” to a higher level by offering a new program, “Stimulating Access to Research in Residency (StARR) (R38).” As we state in our announcement (5), “The program will support institutions to provide support for up to 2 years of research conducted by Resident-Investigators in structured programs for clinician-investigators with defined program milestones.” Trainees who matriculate through the program will be eligible to apply for a subsequent mentored research K38 award (6).
We are pleased that the National Heart, Lung, and Blood Institute is participating in the StARR program, and therefore we urge readers of the Journal to read the announcement and consider applying.
- ↵Lauer MS, Mills S, Lund PK, et al. Addressing the physician/clinical-scientist challenge: an update for the NIH Advisory Committee to the Director. Available at: https://acd.od.nih.gov/documents/presentations/12082016-PSW.pdf. Accessed October 4, 2017.
- ↵Lauer MS, Lund PK. Physician-scientist working group update. Available at: https://acd.od.nih.gov/documents/presentations/06092017Lauer.pdf. Accessed October 4, 2017.
- ↵Department of Health and Human Services. Stimulating Access to Research in Residency (StARR) (R38). Available at: https://grants.nih.gov/grants/guide/rfa-files/RFA-HL-18-023.html. Accessed October 4, 2017.
- ↵Department of Health and Human Services. Notice of intent to publish: Stimulating Access to Research in Residency Transition Scholar (StARRTS) (K38). Available at: https://grants.nih.gov/grants/guide/notice-files/NOT-HL-17-533.html. Accessed October 4, 2017.
All authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- 2017 American College of Cardiology Foundation
- Negi S.I.,
- Koifman E.
- Liu Y.,
- Dai X.
- Lala A.,
- Mentz R.J.
- ↵Accreditation Council for Graduate Medical Education. ACGME program requirements for graduate medical education in cardiovascular disease (internal medicine). Available at: https://www.acgme.org/Portals/0/PFAssets/ProgramRequirements/141_cardiovascular_disease_2017-07-01.pdf. Accessed October 16, 2017.
- Williams E.S.,
- Halperin J.L.,
- Arrighi J.A.,
- et al.
- Ryan T.,
- Berlacher K.,
- Lindner J.R.,
- Mankad S.V.,
- Rose G.A.,
- Wang A.