It Takes a Village: The Value of National Courses Focused on Training Physician-Scientists

In this issue of ATS Scholar, Corder and colleagues describe the “Impact of a Respiratory Disease Young Investigators’ Forum on the Career Development of Physician-Scientists” (1). Although a number of educational papers have been published throughout the years, these almost exclusively focus on training for clinical/procedural skills and are often conducted as part of either residency consortiums or as part of national conferences or bootcamps (2–4). In contrast, the Respiratory Disease Young Investigators’ Forum (RDYIF), which has been run since 2004 and continuously since 2011, is one of the few national forums focused on the education of physician-scientists at the fellow/junior faculty level. Adding to its unique nature is the collaboration between industry and academia to fund and host this forum. The focus on physician-scientists is critically important because numerous studies document the declining number of young investigators, which in turn can lead to a critical shortage of scientific investigators as established investigators retire (5). Although a number of factors contribute to this decline, what is clear is that with fewer and fewer individuals pursuing a career as a physician-scientist, the likelihood of there being a critical mass of peers at any single institution is becoming increasingly less likely. Therefore, the goal of the RDYIF was to create this critical mass by inviting fellows and physician-scientists from academic pulmonary critical care programs around the United States and to use this forum to provide career guidance and education on skills necessary for a career as a successful physician-scientist.

Although a minimum threshold for determining the effectiveness of an educational program is the assessment of learner satisfaction and the perceived value of the program at the conclusion of an activity, a more powerful evaluation is to reassess these perceptions at a future time point to determine the persistence of any benefit. However, the ultimate goal of any educational program is to impact learner behavior/outcomes, which is what the authors try to assess in this study. In 2019, they administered a structured, validated survey to all prior participants with the goal of assessing their current role in research/academia and the perceived value of the RDYIF. Overall, they had 107 complete responses (a 40% response rate). They demonstrated that 76% of subjects are working as physician-scientists, with 97% working in academic health centers and 93% involved in research. More than half the subjects were authors of at least 10 peer-reviewed publications, and 65% had gone on to receive K awards. When looking specifically at the RDYIF, all of the respondents would recommend participating in the RDYIF, with every domain evaluated being of perceived value. Notably, the respondents placed a high value on networking and establishing relationships with mentors. This highlights not only the importance of assessing the direct skills obtained by learners in a course but also the importance of assessing and establishing an environment for peer-to-peer interaction. Finally, the authors are to be commended for the thoroughness of their survey and the robustness of the outcomes data collected, not just in terms of academic productivity but in respondent demographics. These data form an important baseline to determine the success rate for the training of women and minorities for careers in science.

Despite these impressive results, one has to acknowledge that the low response rate introduces potential forms of bias. One is selection bias, as those who failed to pursue a research-oriented career may be less likely to participate in the survey. The second is nonresponse bias because of the long period of elapsed time between participation and survey distribution for some subjects. However, this highlights the difficulty in assessing educational programs with goals of improving outcomes measured over a period of years, such as obtaining grants or faculty positions. This is in stark contrast to many procedural-/knowledge-based educational forums, in which one cannot only use knowledge- or skill-based assessments at the end of a course but often the knowledge or skills taught are applied immediately after the course, greatly facilitating the ability to contact learners and assess their actual outcomes. Another limitation of this study is the lack of assessment of learner perception at the conclusion of the course. If performed, this would have potentially allowed for a better understanding of any bias in nonrespondents (e.g., did they have lower perceived value initially?) and the sustainability of the perceived value across all of the domains assessed.

Although these limitations impact the generalizability of the results and the absence of a control group makes it difficult to determine the true impact of this course, this study provides a blueprint not only for the design of courses focused on training of physician-scientists but also valuable lessons on how to better assess their value in the future.