Abstract
Introduction
Several factors are known to affect the way clinical performance evaluations (CPEs) of medical students are completed by supervising physicians. We sought to explore the effect of faculty perceived “level of interaction” (LOI) on these evaluations.
Methods
Our third-year CPE requires evaluators to identify perceived LOI with each student as low, moderate, or high. We examined CPEs completed during the academic year 2018–2019 for differences in (1) clinical and professionalism ratings, (2) quality of narrative comments, (3) quantity of narrative comments, and (4) percentage of evaluation questions left unrated.
Results
A total of 3682 CPEs were included in the analysis. ANOVA revealed statistically significant differences between LOI and clinical ratings (p ≤ .001), with mean ratings from faculty with a high LOI significantly higher than from faculty with a moderate or low LOI (p ≤ .001). Chi-squared analysis demonstrated differences based on faculty LOI and whether questions were left unrated (p ≤ .001), quantity of narrative comments (p ≤ .001), and specificity of narrative comments (p ≤ .001).
Conclusions
Faculty who perceive higher LOI were more likely to assign that student higher ratings, complete more of the clinical evaluation and were more likely to provide narrative feedback with more specific, higher-quality comments.
Supplementary Information
The online version contains supplementary material available at 10.1007/s40670-021-01307-w.
Introduction
Gathering data from various methods of student assessment is important to accurately judge student competency. Assessment methods such as the National Board of Medical Examiners (NBME) subject matter examinations, objective structured clinical examinations (OSCEs), and reviews of clinical documentation are some examples of evaluations used in the clinical environment [1]. Most medical schools use direct observation and a clinical performance evaluation (CPE) as part of the assessment structure for clerkship performance. Obtaining high-quality evaluation data on student performance throughout clinical clerkships of the medical school curriculum is necessary to ensure validity of assessment.
CPEs, a type of workplace-based evaluation, are based on direct observation of a student in the clinical environment and generally consist of rating scales and descriptive evaluation. Not only do these evaluations often contribute to determining clerkship grades [2], but the quality of descriptive feedback provided can help shape student skill development and create a pathway for growth. Many factors are known to affect evaluation data, including the experience of the evaluator [3], clinical site, and even gender [4–6]. Lesser known is the effect of perceived “level of interaction” of the faculty evaluator on student CPEs, with a paucity of literature examining how perceived level of interaction between the student and the clinical evaluator may affect evaluations.
We sought to understand the impact of a faculty member’s perceived level of interaction (LOI) with respect to (1) clinical and professionalism ratings, (2) quality of narrative comments, (3) quantity of narrative comments, and (4) percentage of CPE questions left unrated.
Methods
Our institutional clerkship CPE [7] includes nine EPA-based performance questions (Appendix). Six questions relate to specific clinical skills and three relate to professionalism, rated on a 0–4 scale with 0 = critical deficiency, 2 = meets expectations, and 4 = exemplary. The evaluator is required to designate a perceived LOI (low, moderate, or high) with the student when completing the CPE. Faculty are asked to consider number of discrete student interactions, time spent with student, and robustness of the interaction to determine their LOI. Each CPE form was completed and submitted via a secure web-based platform. Each student receives an average clinical rating for clinical skills questions, and an average professionalism rating for professionalism questions.
Inclusion Criteria
We examined CPEs completed at Wake Forest School of Medicine during the academic year 2018–2019. Inclusion criteria included CPEs that were completed by full-time faculty who had completed evaluations during the 2017–2018 and 2018–2019 academic years. We selected faculty rather than resident evaluators because they have more experience with completing CPEs and evaluating student performance, in the hopes of reducing variability based on experience. Directed faculty training occurred regularly regarding the importance of feedback and strategies for providing specific, actionable comments. We selected faculty who had completed two consecutive years of student evaluation to assure familiarity with the new form and investment in the educational mission.
Clinical and Professionalism Ratings
For purposes of statistical analyses, each CPE’s total average clinical rating for clinical skill questions and professionalism questions were calculated and categorized based on the faculty member’s perceived LOI.
Quality of Narrative Comments
Three investigators (KA, NH, DM) coded all narrative comments based on quality in both strengths and areas for improvement, defined a priori as none, general, and specific. In developing these categories, the investigators followed a process as outlined in other studies [8]. This process included defining and refining the definition of the different groups, both by consensus and application of the definitions to actual narrative comments. They applied these definitions to 200 narrative comments as a group. The investigators then independently coded an additional 200 narrative comments to assess for inter-rater reliability, with 91% consensus on 200 evaluations. Since greater than 80% agreement has been cited as the acceptable threshold, we interpreted this as acceptable agreement [9]. Each investigator then categorized a portion of the remaining narrative comments. When an investigator was unsure about how to designate a narrative comment, it was discussed between the three investigators until consensus was achieved. Examples of each category are shown in Table 1. Investigators were blinded to LOI.
Table 1.
Examples of each category of specificity for narrative comments
| Narrative comment category | None | General | Specific |
|---|---|---|---|
| Examples | Blank | Good job | Structured and logical presentations |
| Punctuation alone (.) | Read more | Ability to interpret tests beyond level of training | |
| N/A | See more patients | Came to rounds prepared and knowledgeable about patients | |
| None | Enthusiastic learner | Make your presentations as succinct as possible | |
| No specific behavior to improve | Pleasant to work with | Work on developing your neurologic exam |
Quantity of Narrative Comments
We identified the number of evaluations containing substantive narrative comments as those included in the general or specific categories, as noted above. We then assessed for differences in number of evaluations with substantive narrative comments completed by faculty with low, moderate, or high LOI.
CPE Questions Left Unrated
We identified the number of unanswered EPA-based clinical performance questions on each evaluation. We then assessed for differences in completion of the CPE questions between responses with low, moderate, or high LOI.
Statistical Analysis
CPE inclusion criteria, demographics along with frequency statistics regarding faculty self-identified LOI are reported. Clinical and professionalism ratings were analyzed using descriptive statistics. An analysis of variance (ANOVA) was undertaken to establish whether there are differences between clinical and professionalism ratings based on faculty-documented LOI. Post hoc analyses were undertaken using a Tukey test.
Descriptive statistics were conducted to analyze the percentage of “strengths” and “areas for improvement” narrative comments that were rated as “specific” by study investigators. Chi-square analyses were undertaken to explore whether there were statistically significant differences in the quality of comments made (none, general, and specific) based on the documented LOI. Chi-square analyses were conducted on quality of narrative comments for both strengths and areas for improvement.
Descriptive statistics were undertaken to investigate the quantity of substantive comments. General and specific comments were added together and compared to the number of CPEs submitted categorized by LOI. Chi-square analyses were conducted to establish whether there were statistically significant differences in the number of comments and LOI.
Descriptive statistics were conducted to analyze the percentage of CPE questions not rated based on LOI. Chi-square analyses were undertaken to investigate significant differences in the number of CPE questions completed versus not completed, based on documented LOI.
This study was approved by the Institutional Review Board of Wake Forest School of Medicine.
Results
A total of 325 faculty members meeting inclusion criteria submitted 3682 evaluations during the 2018–2019 academic year (see Fig. 1).
Fig. 1.
Individual evaluations that were completed by faculty during the 2018–2019 academic year and their designated LOI
Clinical and Professionalism Ratings
Descriptive statistics for clinical and professionalism ratings are presented in Table 2.
Table 2.
Descriptive statistics for clinical and professionalism ratings based on LOI
| Rating | Level of involvement | n | Mean | Standard deviation |
|---|---|---|---|---|
| Clinical | Low | 1131 | 2.56 | .54 |
| Moderate | 1698 | 2.75 | .60 | |
| High | 698 | 2.90 | .65 | |
| Professionalism | Low | 1230 | 2.82 | .65 |
| Moderate | 1717 | 3.03 | .65 | |
| High | 700 | 3.22 | .67 |
ANOVA results established significant differences between LOI and clinical ratings (F[2.3524] = 73.979, p ≤ 0.001). A post hoc Tukey test revealed that the mean rating from faculty with a high LOI (2.90, p ≤ 0.001) was higher than from faculty with a moderate LOI (2.75, p ≤ 0.001) and faculty with a low LOI (2.54, p ≤ 0.001). A post hoc Tukey test also demonstrated that the mean rating from faculty with moderate LOI was significantly higher than the mean rating from faculty with a low LOI.
Significant differences were observed for professionalism ratings based on LOI (F[2,3644] = 88.720, p ≤ 0.001). A post hoc Tukey test demonstrated that the mean rating from faculty with high LOI (3.22, p ≤ 0.001) was higher than from faculty with moderate LOI (3.03, p ≤ 0.001) and faculty with a low LOI (2.82, p ≤ 0.001). A post hoc Tukey test revealed that the mean rating from faculty with moderate LOI was higher than the mean rating from faculty with low LOI.
Quality of Narrative Comments
Narrative comments related to strengths were categorized as “specific” in 63.3% of submitted CPEs that documented high LOI, 47.1% of CPE submissions with moderate LOI, and 26.3% of CPE submissions with low LOI. There was a difference between the investigators’ narrative comment categorization and faculty documented LOI (X2 (4) = 297.490, p ≤ 0.001).
Narrative comments related to areas for improvement were categorized as “specific” in 40.5% of submitted CPEs documenting high LOI, 29.9% of submitted CPEs with moderate LOI, and 20.7% of submitted CPEs with low LOI. There was a significant difference between the investigators’ narrative comment categorization and faculty documented LOI (X2 (4) = 170.073, p ≤ 0.001). These findings are summarized in Table 3.
Table 3.
Narrative comments categorization by LOI
| Level of involvement | Number of CPEs submitted | None (%) | General (%) | Specific (%) |
|---|---|---|---|---|
| Strengthsa | ||||
| Low | 1254 | 494 (39.4) | 430 (34.2) | 330 (26.3) |
| Moderate | 1727 | 412 (23.9) | 501 (29.0) | 814 (47.1) |
| High | 701 | 96 (13.7) | 161 (23.0) | 444 (63.3) |
| Areas for improvementb | ||||
| Low | 1254 | 840 (67.0) | 155 (12.3) | 259 (20.7) |
| Moderate | 1727 | 895 (51.8) | 316 (18.3) | 516 (29.9) |
| High | 701 | 260 (37.1) | 157 (22.4) | 284 (40.5) |
aStrengths: X2 (4) = 297.490, p ≤ .001
bAreas for improvement: X2 (4) = 170.073, p ≤ .001
Quantity of Narrative Comments
The number of general and specific comments were added together and compared to the number of CPEs submitted, categorized by LOI. Narrative comments regarding areas of strength were observed in 86.3% of submitted CPEs with high LOI, 76.1% of submitted CPEs with moderate LOI, and 60.6% of submitted CPEs with low LOI. Chi-squared analysis showed these differences to be statistically significant (X2 (2) = 168.431, p ≤ 0.001).
Similar results were found for narrative comments regarding areas for improvement. General or specific comments were observed in 62.9% of high LOI CPEs, 48.2% of moderate LOI CPEs, and 33.0% of low LOI CPE submissions. Chi-squared analysis showed these differences to be statistically significant (X2 (2) = 169.1701, p ≤ 0.001).
CPE Questions Left Unrated
The percentage of clinical performance CPE questions left unrated was 12.0% for high, 18.0% for moderate, and 35.3% for low LOI. Chi-square analysis revealed a significant difference between documented faculty LOI and whether CPE questions were rated or not (X2 (2) = 1086.9275, p ≤ 0.001).
The percentage of professionalism performance CPE questions left unrated was 9.9% for high, 16.7% for moderate, and 26.8% for low LOI. Chi-square analysis revealed a significant difference between documented faculty LOI and whether professionalism CPE questions were rated or not (X2 (2) = 281.9613, p ≤ 0.001). Results are summarized in Table 4.
Table 4.
Clinical performance and professionalism questions rated or left unrated, based on faculty documented LOI
| Level of involvement | Number of CPEs submitted | Total number of questions | Number of questions rated (%) | Number of questions unrated (%) |
|---|---|---|---|---|
| Clinical performancea | ||||
| Low | 1254 | 7,524 | 4,866 (65%) | 2,658 (35%) |
| Moderate | 1727 | 10,362 | 8,498 (82%) | 1,864 (18%) |
| High | 701 | 4,206 | 3,701 (88%) | 505 (12%) |
| Professionalismb | ||||
| Low | 1254 | 3762 | 2,753 (73%) | 1,009 (27%) |
| Moderate | 1727 | 5181 | 4,315 (83%) | 866 (17%) |
| High | 701 | 2103 | 1,895 (90%) | 208 (10%) |
aX2 (2) = 1086.9275, p ≤ .001; 77.2% of the clinical questions had ratings (17,065/22,092)
bX2 (2) = 281.9613, p ≤ .001; 81.1% of the professionalism questions had ratings (8963/11,046)
Discussion
Our study showed that higher faculty LOI was associated with higher ratings, increased quantity and quality of the narrative feedback and with fewer questions left unrated. The reasons for these associations are unclear and demand further investigation, but there are a few possibilities that can be postulated. The demonstrated association of higher scores with higher LOI may be the result of a more thorough student assessment leading to a better understanding of the students’ skills, whereas limited interaction caused the evaluator to regress to a default score closer to “meets expectations.” Faculty with higher LOI likely had more in-depth observations and/or higher frequency of direct observations of the student’s clinical skills, resulting in more questions rated and more specific feedback for the student. Improved feedback may have resulted from an enhanced “educational alliance” forged by working in closer proximity or for a greater length of time [10]. Indeed, the conceptual framework of the “educational alliance” could illuminate these associations by placing the feedback and evaluation process in the context of the relationship between student and faculty member [11]. In that light, more interactions and a closer relationship might be expected to lead to exactly the kinds of associations observed here. Another, less promising, possibility would be that students who fared worse were assigned lower ratings, got less and lower quality narrative feedback, and the evaluating faculty attempted to soften all this by rating their LOI lower. Further study, perhaps using qualitative methods, will be needed to more definitively explain these results.
There was an inverse relationship between LOI and number of questions left unrated on the CPEs. This held true in both the clinical skill-based and professionalism CPE questions. Questions related to global professionalism attributes and teamwork were almost universally completed. This aligns with previous research showing that teaching faculty can readily assess student professionalism in the context of a clinical rotation [12]. Our data suggests that higher LOI may allow faculty to observe and assess specific skills that may otherwise not be formally assessed, including entrustable professional activities related to medical record documentation and contributing to a culture of safety.
LOI was related to the frequency of documented narrative feedback, and importantly the frequency of specific feedback. Previous work supports use of narrative-type feedback in workplace-based assessments for more nuanced and meaningful student feedback. Students are better poised to use narrative comments to change their behavior compared to numerical ratings [13, 14]. Given this context and the results of our investigation, it appears that faculty having more in-depth interaction with a student may lead to more actionable feedback.
These findings may have implications for how quantitative and qualitative data are used together to create holistic evaluations. Relying solely on quantitative data has numerous potential disadvantages: inability of quantitative assessments to accurately depict student competency in complex skill acquisition, negative impact of test-taking anxiety on quantitative assessments, and lack of correlation between quantitative scores and future clinical performance.12 Previous studies have demonstrated that narrative comments in the context of medical training carry value by providing a more comprehensive overview of a learner’s abilities and are as reliable as quantitative data [15–22].
Regardless of faculty LOI, there remain weaknesses in faculty-student workplace-based assessments which must be mitigated. These include demonstrative differences between novice and expert assessors [23] and biases pervasive among even the most experienced assessors [24]. Evaluators can be influenced by these biases and the environment in which the assessment is taking place [25]. Even skilled evaluators tend to overlook weaknesses in otherwise strong students and strengths in “problem” students tend to be minimized [26]. Assessors may infer student ability from global observation of the student instead of forming an assessment based on direct observation of the skill [12]. Targeted faculty development can mitigate these biases in specific domains [27]. Faculty development should be continuous and purposeful to drive effective and accurate assessment of students [28, 29]. The findings in this study may simply add to the mix of variables that ultimately are reflected in clinical evaluation data for students.
Limitations
This was a single institution study and it is possible our results may not be generalizable to other institutions. This was retrospective, observational data, and so conclusions around causation should not be inferred. We purposefully selected a subset of year 3 CPE data to analyze for this study to include only faculty evaluators with at least a year of experience at our institution. The results may not accurately reflect the entirety of the student-faculty CPE experience, but we believe our selection methods functioned to focus attention on the actual effect of LOI rather than on confounding factors such as evaluator experience. Further, faculty self-determined their level of student involvement without the use of a rubric, although general guidance was given during faculty development sessions. Several of the authors were among the 325 faculty who submitted CPEs as part of the study.
Conclusions
Faculty reporting higher LOI with a student completed more CPE items, provided more specific feedback, and assigned higher ratings for students than when they reported a lower LOI. Further study may elucidate the reasons these associations are seen. Better understanding of these associations may be helpful in designing evaluation systems and when creating meaningful clinical experiences for students and faculty.
Supplementary Information
Below is the link to the electronic supplementary material.
Declarations
Ethical Approval
This study was approved by the Wake Forest Health Sciences IRB.
Conflict of Interest
The authors declare no competing interests.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.AAMC. Assessment Methods in Clinical Clerkship Experiences. https://www.aamc.org/data-reports/curriculum-reports/interactive-data/assessment-methods-clinical-clerkship-experiences-formative-and/or-summative. Accessed 2 Feb 2020.
- 2.Hauser LJ, Gebhard GM, Blumhagen R, et al. Applicant characteristics associated with successful matching into otolaryngology. Laryngoscope. 2017;127(5):1052–1057. doi: 10.1002/lary.26236. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Carline JD, Cook CE, Lennard ES, et al. Resident and faculty differences in student evaluations: implications for changes in a clerkship grading system. Surg. 1986;100(1):89–94. [PubMed] [Google Scholar]
- 4.Fay EE, Schiff MA, Mendiratta V, et al. Beyond the ivory tower: a comparison of grades across academic and community OB/GYN clerkship sites. Teach Learn Med. 2016;28(2):146–151. doi: 10.1080/10401334.2016.1146603. [DOI] [PubMed] [Google Scholar]
- 5.Maxfield H, Kennedy M, Delzell JE, Paolo AM. Performance of third year medical students on a rural family medicine clerkship. Fam Med. 2014;46(7):536–538. [PubMed] [Google Scholar]
- 6.Riese A, Rappaport L, Alverson B, et al. Clinical performance evaluation of third year medical students and association with student and evaluator gender. Acad Med. 2017;92(6):835–840. doi: 10.1097/ACM.0000000000001565. [DOI] [PubMed] [Google Scholar]
- 7.Askew K, Manthey D, Potisek N, et al. Practical application of assessment principles in the development of an innovative performance evaluation in the entrustable professional activity era. Med Sci Educ. 2019;e-pub ahead of print. [DOI] [PMC free article] [PubMed]
- 8.Pelgrim EA, Kramer AW, Mokkink HG, et al. Quality of written narrative feedback and reflection in a modified mini-clinical evaluation exercise: an observational study. BMC Med Educ. 2012;12:97. doi: 10.1186/1472-6920-12-97. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.McHugh M. Interrater reliability: the kappa statistic. Biochem Med. 2012;22(3):276–282. doi: 10.11613/BM.2012.031. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Bowen L, Marshall M, Murdoch-Eaton D. Medical student perceptions of feedback and feedback behaviors within the context of the “educational alliance”. Acad Med. 2017;92(9):1303–1312. doi: 10.1097/ACM.0000000000001632. [DOI] [PubMed] [Google Scholar]
- 11.Telio S, Ajjawi R, Regher G. The, “educational alliance” as a framework for reconceptualizing feedback in medical education. Acad Med. 2015;90(5):609–614. doi: 10.1097/ACM.0000000000000560. [DOI] [PubMed] [Google Scholar]
- 12.Pulito AR, Donnelly MB, Plymale M, et al. What do faculty observe of medical students' clinical performance? Teach Learn Med. 2006;18(2):99–104. doi: 10.1207/s15328015tlm1802_2. [DOI] [PubMed] [Google Scholar]
- 13.Hanson JL, Rosenberg AA, Lane JL. Narrative descriptions should replace grades and numerical ratings for clinical performance in medical education in the United States. Front Psychol. 2013;4:668. doi: 10.3389/fpsyg.2013.00668. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Ginsburg S, van der Vleuten CPM, Eva KW. The hidden value of narrative comments for assessment: a quantitative reliability analysis of qualitative data. Acad Med. 2017;92(11):1617–1621. doi: 10.1097/ACM.0000000000001669. [DOI] [PubMed] [Google Scholar]
- 15.Cohen GS, Blumber P, Ryan NC, et al. Do final grades reflect written qualitative evaluations of student performance? Teach Learn Med. 1993;5:10–15. doi: 10.1080/10401339309539580. [DOI] [Google Scholar]
- 16.Battistone MJ, Pendleton B, Milne C, et al. Global descriptive evaluations are more responsive than global numeric ratings in detecting students’ progress during the inpatient portion of an internal medicine clerkship. Acad Med. 2001;76:S107. doi: 10.1097/00001888-200110001-00035. [DOI] [PubMed] [Google Scholar]
- 17.Schwind CJ, Williams RG, Boehler ML, et al. Do individual attendings’ post-rotation performance ratings detect residents’ clinical performance deficiencies? Acad Med. 2004;79:453–457. doi: 10.1097/00001888-200405000-00016. [DOI] [PubMed] [Google Scholar]
- 18.Eva KW, Hodges BD. Scylla or Charybdis? Can we navigate between objectification and judgement in assessment? Med Educ. 2012;46:914–919. doi: 10.1111/j.1365-2923.2012.04310.x. [DOI] [PubMed] [Google Scholar]
- 19.Guerrasio J, Cumbler E, Trosterman A, et al. Determining need for remediation through post rotation evaluations. J Grad Med Educ. 2012;4:47–51. doi: 10.4300/JGME-D-11-00145.1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Nasca TJ, Phillibert I, Brigham T, et al. The next GME accreditation system-rationale and benefits. N Engl J Med. 2012;366:1051–1056. doi: 10.1056/NEJMsr1200117. [DOI] [PubMed] [Google Scholar]
- 21.Bartels J, Mooney CJ, Stone RT. Numerical versus narrative: a comparison between methods to measure medical student performance during clinical clerkships. Med Teach. 2017;39(11):1154–1158. doi: 10.1080/0142159X.2017.1368467. [DOI] [PubMed] [Google Scholar]
- 22.Lefebvre C, Hiestand B, Glass C, et al. Examining the effect of narrative commentary on evaluators’ summative assessments of resident performance. Eval Health Prof. 2018;e-pub ahead of print. [DOI] [PubMed]
- 23.Govaerts MJ, Schuwirth LWT, Van der Vleuten CPM, et al. Workplace-based assessment: effects of rater expertise. Adv Health Sci Educ Theory Pract. 2011;16(2):151–165. doi: 10.1007/s10459-010-9250-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Williams RG, Klamen DA, McGaghie WC. Cognitive, social and environmental sources of bias in clinical performance ratings. Teach Learn Med. 2003;15(4):270–292. doi: 10.1207/S15328015TLM1504_11. [DOI] [PubMed] [Google Scholar]
- 25.Gingerich A, Kogan J, Yeates P, et al. Seeing the 'black box' differently: assessor cognition from three research perspectives. Med Educ. 2014;48(11):1055–1068. doi: 10.1111/medu.12546. [DOI] [PubMed] [Google Scholar]
- 26.Ginsburg S, McIlroy J, Oulanova O, et al. Toward authentic clinical evaluation: pitfalls in the pursuit of competency. Acad Med. 2010;85(5):780–786. doi: 10.1097/ACM.0b013e3181d73fb6. [DOI] [PubMed] [Google Scholar]
- 27.Morewedge CK, Yoon H, Scopelliti I, et al. Debiasing decisions: improved decision making with a single training intervention. Policy Insights Behav Brain Sci. 2015;2(1):129–140. doi: 10.1177/2372732215600886. [DOI] [Google Scholar]
- 28.Steinert Y, Mann K, Centeno A, et al. A systemic review of faculty development initiatives designed to improve teaching effectiveness in medical education: BEME Guide No 8. Med Teach. 2006;28:497–526. doi: 10.1080/01421590600902976. [DOI] [PubMed] [Google Scholar]
- 29.Oller KL, Mai CT, Ledford RJ, et al. Faculty development for the evaluation system: a dual agenda. Adv Med Educ Pract. 2017;8:205–210. doi: 10.2147/AMEP.S124004. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.