We found that residents and faculty agreed that lifelong learning is necessary to being an effective physician. However, a majority of residents, but a minority of faculty, believed that residents should be primarily responsible for directing their own learning. At the same time, a third of residents, but no faculty, thought that assigned clinical and curricular experiences were sufficient to make an individual a competent physician. Because these views are inconsistent, and the fact that residents believed this and faculty did not, suggests the possibility that the training received and experience gained during residency training and beyond are important in developing a full understanding of how self-directed learning fits into practice. This finding also underscores the importance of fostering the ongoing development of self-assessment and self-directed learning skills throughout the continuum of medical education.
Only half the residents and faculty had a good understanding of how to assess their own skills or were confident in their ability to do so. In addition, first-year residents were less likely than senior residents to have a good understanding of how to assess their own skills and were less confident in their ability to do so. The challenge of "determining one's own strengths and weaknesses" was also voiced in the qualitative comments, and is consistent with previous studies that have shown that physicians have limited ability to self-assess[10
]. Fortunately, previous studies have shown that physician self-assessment skills can be improved through performance feedback[14
]. When fourth-year medical students were shown their standardized patient ratings and how their performance compared with that of other students, their self-assessment of their clinical skills improved[14
]. In addition, studies of portfolios have indicated that reflection is enhanced with adequate faculty mentorship[15
]. Our findings, and those of these other studies suggest it would be useful to design ILPs in a way that requires residents to pair their own self-evaluation with external evaluations (such as multisource feedback) with the help of a faculty advisor in order to further develop self-assessment skills.
Our study also showed that residents were less confident in their ability to develop effective ILPs than they were in their ability to assess their own skills and develop specific goals to improve performance. This discrepancy is highlighted in our qualitative data, in which residents describe challenges to developing an ILP that include "creating a realistic plan" to accomplish their specific goals, "adequate time and motivation to really focus on it," and "follow [ing] through with the plan." These comments suggest that residents acknowledge that the challenge of self-directed learning is not just assessing weaknesses and setting specific goals to improve those weaknesses, but developing and implementing realistic plans to accomplish these goals. This commentary suggests that in order for ILPs to be effective, faculty should try to assist learners with the development and implementation of realistic plans.
Residents were less likely than faculty to continuously assess their own performance or continuously direct their own learning even though survey results demonstrated that they felt primarily responsible for their own learning. Our findings support previous reports that suggest that physicians-in-training may not be well-prepared to self-direct their learning[17
]. For example, when third-year medical students in a family medicine clerkship were given the opportunity to choose their own learning goals, most selected from prewritten learning goals rather than develop their own[17
]. In a pilot study of ILPs in a pediatric continuity clinic, most (90%) residents were able to develop learning goals, but few (25%) documented progress toward achieving their goals[18
]. Our results and the results of these previous studies suggest the need to develop tools to improve self-directed learning skills. Exposure to problem-based learning curricula[19
] has been shown to improve self-directed learning in pediatric residents, but other tools to promote self-directed learning, such as ILPs[21
], have not been well studied.
Our study has several limitations and identified several potential barriers to implementation of ILPs that must be considered. The survey was administered at a single institution; therefore, generalizability is limited because it remains possible that residents and faculty at other institutions and other specialties may have different attitudes toward physician self-assessment and self-directed learning. However, our institution is similar to other academic medical centers in that faculty have trained in a broad representation of training programs and residents come from a wide range of medical schools throughout the United States. The demographic differences we found between residents and faculty (practice setting, subspecialty, gender) in our study may be a confounding factor, but our study had insufficient power to test this possibility. By definition, all faculty practice in an academic institution, most academic pediatricians are subspecialists, and until recently, there were fewer women in pediatrics (and medicine, in general)[22
]. A larger study will need to be done in order to more closely examine these factors. Faculty who responded to the survey may be different from faculty who did not respond to the survey. However, demographics of faculty respondents and all faculty were similar. Finally, in our survey we discovered that some faculty and residents did not have a good understanding of what an ILP was. In addition, many residents were unsure how to write an effective ILP and faculty were unsure of their ability to help them. While these findings may be due to the fact that the survey was distributed prior to initiation of ILPs, they underscore the need to provide appropriate training for faculty and residents prior to implementation of an ILP.
Our paper describes initial attitudes, knowledge and skills surrounding self-assessment, self-directed learning and ILPs prior to the implementation of ILPs in our program. Our study suggests that self-assessment and self-directed learning are not innate skills and need to be developed over the course of training. While ILPs may be a tool to improve self-assessment and self-directed learning skills through deliberate practice, our study suggests that several challenges will need to be addressed before ILPs can be successfully implemented. These include: assisting residents and faculty in viewing ILPs as a valuable tool to improve lifelong learning, not merely a "time-consuming" exercise; building flexibility into ILPs in order to meet the evolving goals of learners with different learning styles; and better integrating ILPs into residency training. Ideally, dedicated time would be provided to develop and implement ILPs. Resident and faculty development is needed on how to construct an effective ILP. Such training would most likely involve how to (1) compare external assessment with self-assessment to identify strengths and weaknesses, (2) write specific goals focused on individual improvement, and (3) construct realistic plans to achieve those goals. ILPs could be actively integrated into clinical rotations by having residents share their learning goals with attendings at the beginning of each clinical rotation, so that learning opportunities surrounding specific learning goals could be optimized and learning would become a more "collaborative" partnership.
Further research is needed to better understand how best to integrate ILPs into graduate medical education and whether ILPs help residents develop self-assessment and self-directed learning skills. A repeat survey after implementation of ILPs may help determine whether actual implementation of ILPs affects resident attitudes, knowledge or skills surrounding self-assessment, self-directed learning and ILPs. A national survey of residents at multiple training programs could validate the findings from our study and improve its generalizability. Research is needed on the efficacy of ILPs to improve lifelong learning and best implementation strategies if ILPs are to be most effective.