Ephgrave, Kimberly MD; Margo, Katherine L. MD; White, Christopher MD; Hammoud, Maya MD; Brodkey, Amy MD; Painter, Thomas MD; Juel, Vern C. MD; Shaw, Darlene PhD; Ferguson, Kristi PhD
Clerkship directors (CDs) are key medical educators who shape students' clinical education and serve administrative functions for their institutions. In the post-Flexner era, CDs first met to address medical students' clinical education within large discipline-based educational groups such as the American Academy of Neurology and the Association of Professors of Gynecology and Obstetrics. Groups of educators with specific interest in the clinical education of medical students formed subsequently, beginning with Family Medicine in 1967, followed by Psychiatry (1975), Surgery (1981), Internal Medicine (1991), and Pediatrics (1992). The Alliance for Clinical Education (ACE) is a multidisciplinary group formed in 1992 to advance students' clinical education through collaboration of each of the above core CDs' groups. Subsequent activities of ACE include creation of the Guidebook for Clerkship Directors,1 multidisciplinary clinical education projects,2,3 and the development of a consensus statement detailing the expectations of and for CDs.4
Most discipline-based groups query their membership regularly5–8 or periodically.9–20 The data from these discipline-specific surveys suggest that CDs' characteristics, resources, and academic productivity have evolved since the formation of the ACE 18 years ago.5–21 Furthermore, the ACE leadership council noted that despite the 2003 publication of guidelines for CDs' resources,4 many CDs might not be receiving adequate support. Thus, ACE performed a single survey of all seven core CD groups simultaneously to identify which aspects of core clinical education, CD resources, and CD experiences are universal. In this report, we describe the survey and its findings.
The 2005 ACE administrative council assembled a working group with representatives from each constituent organization to address issues of resources and support for the direction of all core disciplines' clerkships. The working group created a Web-based survey of CDs at the 125 MD-granting U.S. medical schools at that time, using items from previous, discipline-based surveys and new items related to the ACE guidelines.4 The University of Iowa Carver College of Medicine's Office for Consultation and Research in Medical Education set up the online 50-item survey22 and performed the data analysis. The protocol was approved by the University of Iowa institutional review board on September 14, 2006.
Each constituent organization distributed the survey to their members during academic year 2006–2007. Two electronic reminders were sent over the course of the year. Because the data were collected anonymously, these reminders could not be targeted to nonresponders. All responses were received by the end of 2007.
Data analysis consisted of ANOVA to assess overall differences across groups, followed by pairwise comparisons when appropriate. Pearson product–moment correlations assessed bivariate relationships, while linear regression tested for the multivariate relationships of demographic and resource factors with academic productivity.
A total of 544 CDs responded, representing approximately 60% of the seven types of U.S. core clerkships. The distribution varied from a low of 56 responses (44%) from CDs in Obstetrics–Gynecology to a high of 96 (76%) from those in Internal Medicine, where the instrument was incorporated into their well-established annual survey.23 The numbers of all respondents, by discipline, are Internal Medicine (96), Family Medicine (91), Psychiatry (91), Pediatrics (79), Surgery (71), Neurology (60), and Obstetrics–Gynecology (56).
Just under half the respondents reported that they act as CDs alone, whereas the remainder share CD responsibilities. The percentage of solo directors varied by discipline (see Table 1); solo responsibility was least common for respondents in Pediatrics and Internal Medicine. Over half of the responding CDs (64%) report directly to their department head for this role, whereas 12% report to a vice chair, 10% to a director of medical education, 4% to their department's residency program director, and 10% to another administrator.
The ages of the responding CDs ranged from 30 to 75 years, with means clustered tightly between 45 (Neurology) and 50 (Surgery). The respondents' years as faculty members and years in the CD role were also similar across disciplines (see Table 1).
A striking finding of this study is that only 142/544 (26%) of current CDs report being on the tenure track; the lowest was 15/79 (19%) in Pediatrics, and the highest was 33/71 (45%) in Surgery. Of those 26% on the tenure track, 95 (68%) were already tenured. Thus, only 47/544, or 8.5%, of CDs at the time of this study were working toward promotion on the tenure track. Unlike the tenure-track percentage, the distribution of academic ranks did not vary significantly among disciplines (see Table 1). Overall, 103/544 (19%) reported a rank of full professor, with 207/544 and 204/544 (38%) at associate and assistant professor levels.
Medical school and clerkship characteristics
The majority of survey respondents (345/544; 64%) reported that their medical schools are public, reflecting the distribution of accredited MD-granting medical schools 76/125 (61%) in the United States at the time of this study. The entering class sizes varied from 20 to 301, with a mean of 143 (SD = 51).
The clerkship length varied between disciplines (see Table 2). Internal Medicine's mean at 9.6 weeks was significantly longer than Neurology's 4.2. Approximately 83/544 (15%) of clerkships had changed length in the previous three years, with two-thirds shortening their lengths to make room for new rotations. In-house call requirements for medical students persisted in approximately half of the clerkships (251/544; 46%), with wide variation between disciplines.
The majority of the responding CDs were physicians with substantial clinical responsibilities. The number of half-days of outpatient clinic responsibilities varied widely between and within disciplines, covering the entire range from 0 to 10 half-days. The difference between the lowest disciplinary mean (2.7 half-days/week in Surgery) and highest (4.7 half-days/week in Obstetrics–Gynecology) was noteworthy, with Psychiatry (3.3), Pediatrics and Internal Medicine (3.7 each), Neurology (4.1), and Family Medicine (4.2) falling in between.
Similarly, weeks of inpatient responsibilities for individual respondents spanned the entire range from 0 to 52, with significant differences between the two disciplines at the extremes. Reflecting differences in clinical practice, the lowest mean number of inpatient service weeks was reported by Family Medicine (8.3), with Surgery highest (40.7), and Neurology (11.4), Internal Medicine (11.7), Pediatrics (12.9), and Obstetrics–Gynecology and Psychiatry (23 weeks each) bridging the middle range.
Academic productivity was measured by four questions regarding total and medical-education-related presentations in the last five years, and total and medical-education-related publications in the last three years (see Table 3). Again, the individual variation across all 544 respondents was large, varying from 0 to 75 for total presentations in the last five years. However, differences in means between disciplines were negligible (see Table 3). Total publications averaged 5.2 during the last three years, whereas education-related publications averaged 1.7, or approximately one-third of the CDs' academic productivity. Testing for relationships by linear regression revealed that outpatient clinical responsibilities were inversely related to publications, both overall and education-related. In addition to the link between outpatient responsibilities and publications, academic productivity as measured by presentations in medical education was significantly linked to the number of years as CD, the percentage of work time the CD reported devoting to the clerkship, and the amount of support staff time dedicated to the clerkship.
Most CDs have little control over the budget for medical students' clinical education, and many have little knowledge of departmental resources. A slim majority (317/544; 58%) reported knowing the source of funds used for the clerkship. Less than half (210/544; 39%) reported knowing that there is a line item for student education in the department's budget, and 109/544 (20%) stated that medical student education was not on their departmental budget. However, the largest group (223/544; 41%) did not know whether or not medical student education appeared in their departmental budget. A total of 218/544 (40%) of the CD respondents stated that they have discretionary funds for educational materials or attendance at educational meetings, 277/544 (51%) stated they do not have such funds, and 48/544 (9%) stated they did not know.
Two of the most critical resources for the clerkship are the time devoted by the CD and the time devoted by his or her support staff. The mean percentage of work time spent on the CD role by all faculty involved for the responding clerkships was 57%, distinctly higher than the individual respondents' effort, because the majority of CDs now share the role. This total faculty effort for the CD role varied significantly by discipline, from a low of 31.8% (Neurology) to a high of 73.8% (Internal Medicine).
Time spent by nonprofessional staff in support of the clerkship varied even more than did faculty effort. Respondents reported zero to three support-staff full-time equivalents (FTE). The overall mean amount of time given by support staff to the clerkships was 0.69 FTE. Again, cross-disciplinary differences occurred, with Family Medicine (0.85 mean support staff FTE) reporting twice the mean for Neurology (0.40). As noted above, support from clerical staff correlated significantly with the CD's academic productivity. Support to the clerkship from a professional educator was reported by only 20% of respondents, however, ranging from 7% in Obstetrics–Gynecology and Neurology to 36% for Family Medicine.
The majority of respondents reported some salary support for the CD role, averaging 22% overall. Many respondents did not have any reductions within clinical assignments due to their CD role, however. The overall 16% mean protected professional time protected for the CD role is less than half of the 33% time that respondents reported devoting to the role.
In the survey, we asked whether the CD role had significantly enhanced, somewhat enhanced, had no effect, somewhat impaired, or significantly impaired the CD's professional advancement (see Figure 1). A majority (400/544; 74%) reported that the CD role at least somewhat enhanced their professional advancement. A total of 94/544 (17%) were neutral, whereas 45/544 (8%) felt the CD role impaired their professional advancement. The survey respondents' judgment regarding impact of the CD role on professional advancement was positively related to their number of years in the CD role and to higher productivity in terms of education-related research.
We also asked how the CD role had affected the respondents' satisfaction with their work. Here, a clear majority (303/544; 56%) stated that it had significantly enhanced their work-related satisfaction, whereas only 6% felt it had impaired their satisfaction (see Figure 1).
The largest factor associated with the CD role's enhancing satisfaction with work was the degree to which the CD felt that the CD role had helped his or her professional advancement. Additionally, a higher number of years spent as CD, a higher percentage of professional effort devoted to the role, and the degree of release from clinical responsibilities as a result of being a CD were all associated with the role's enhancing of work satisfaction.
In 2003, ACE first disseminated its consensus guidelines regarding expectations of and for CDs.4 Those guidelines recommended the allocation of more than 50% of a faculty member's time for the CD role, stated that at least 25% of that time was necessary for administrative responsibilities combined with educational scholarship, and said the rest (at least 25%) was needed for hands-on teaching. The CD's administrative responsibilities include addressing the Liaison Committee on Medical Education's (LCME's) clinical education standards, participating in local education committees, faculty and resident development, clerkship data analysis, and clerkship improvement as well as the scheduling, grading, and other daily activities of clerkship management.
Our study's cross-sectional data from seven core disciplines regarding CD demographics, resources, and academic productivity demonstrate the wide range of CD circumstances, often falling short of the 2003 ACE recommendations.4 Although our data were collected in 2007, nothing has led us to think that it differs from the CD situation today. With data from 544 CD respondents representing >60% of the core clerkships in the United States, we believe that this unprecedented database will allow the core disciplines as well as individual clerkships to compare themselves with national norms.
In keeping with previous disciplinary studies, one shift that our data demonstrate is that faculty serving as CDs are more frequently being appointed as clinician educators and are less frequently on a tenure track.8,10,12,15,17,20 Only 26% of core CDs at the time of this study were on the tenure track, and most of those were already tenured. Although achieving tenure was not reported as a source of stress by most of the responding CDs, they expressed other concerns, such as changes in LCME accreditation requirements and the CD's own clinical responsibilities.
Most CDs have heavy clinical responsibilities, with the distribution of outpatient and inpatient service varying according to discipline. Work by Hemmer et al5 demonstrated that academic productivity for Internal Medicine CDs was associated with having fewer than three half-days of clinic per week. Most respondents to our survey have more than this amount, as they reported an average of 3.7 half-days of clinic per week. The current dataset supports the findings of Hemmer et al in that half-days of clinic per week were inversely related to both the CDs' overall academic productivity and specific productivity in medical education. It is possible that this linkage between clinic time and academic productivity occurs because some CDs are not interested in academic pursuits and thus accept heavy clinical loads, but it is also possible that the outpatient clinical responsibilities impair their ability to pursue academic scholarship.
It is difficult to determine the exact amount of clinical responsibility that is ideal for CDs. Some clinical activity seems desirable, so that CDs can contribute to their departments' clinical mission and maintain their clinical skills. However, our data show that release from clinical responsibilities for the CD role and strong staff support are positively associated with CDs' academic productivity and the degree to which the role enhances satisfaction with work. The demands of the CD role are considerable and continue throughout the entire academic year, but our data indicate that many current CDs have little decrease of their clinical time in recognition of the CD role, and the majority of clerkships also fall short of the ACE guidelines for at least one FTE for support staff time.
The provision of support staff to the CDs is critical, but highly variable. The ACE consensus for one FTE support staff dedicated to the clerkship is approached only in Family Medicine, with an average of 0.85 FTE. Neurology, which is not a required clerkship in all medical centers, is significantly less well supported than Family Medicine, with a mean of 0.4 FTE support staff. Neurology respondents were also most likely to report that they had no budget for medical education (70%), and they expressed concern regarding the impact of the role on their professional advancement.
It is likely that many current CDs and their department heads are unaware of the ACE guidelines4 for clerkship resources. Our data show that most CDs are also unfamiliar with their own financial resources. Some CDs reported that there is a line item in their departmental budget for medical student education, but the largest proportion (223/544; 41%) stated that they do not know whether medical education appears in the departmental budget.
Despite the challenges and time required for the CD role, 400/544 (70%) of our respondents reported that the role had a positive effect on their academic advancement. Indeed, medical education presentations accounted for a mean of over half of all respondents' presentations and also accounted for a mean of about one-third of their publications.
For all core clinical disciplines, it is clear that the majority of respondents value their CD role. In addition to the 70% who report feeling that the role enhanced their academic advancement, >90% reported that being a CD enhanced their satisfaction with their work. This satisfaction with the CD role correlated with feeling that the role enhanced academic productivity, and with the number of years and percentage of effort in the role. The strong, positive impact of the CD role on career satisfaction is in keeping with the medical literature, which suggests that a variety of physicians, including rural nonacademic physicians24 as well as academic physicians,25–27 link their career satisfaction to their opportunities to teach.
Although it is encouraging that most CDs find the role intrinsically rewarding, it is concerning that lack of release from clinical loads and/or lack of staff support for many make it difficult to excel as CDs or to be academically productive. The major gaps we detected are between the amount of time the respondents devote to the clerkship (∼33%) and the extent to which CDs have decreased clinical responsibilities (∼16%), and between the recommended staff support of approximately 1 FTE and the reality of 0.69 FTE.
Our study has several limitations. Because the survey was anonymous, comparisons were not possible within or between schools. We do not have data regarding nonrespondents, but respondents may well have been more satisfied or invested in their roles than were the nonrespondents. We also did not ask the sex of the CDs we surveyed, which previous studies show has been steadily shifting to a larger proportion of women. In addition, we did not ask if the CDs are satisfied with the amount of time and support they have, as in other surveys21; we only asked if they feel the CD role enhanced their work satisfaction overall. Our data suggest that CDs are a resilient group, but there may be more negative effects of the role than our data reveal.
In summary, our cross-disciplinary CD data contain both encouraging and discouraging news. Being a CD is perceived as a positive experience for the majority, despite the substantial responsibilities and less-than-optimum support. Over half of current clerkships meet the ACE guidelines for total faculty time; the same is not true for support personnel.4 Thus, the most low-cost strategy for improving current clinical education may be to supply additional support staff. Investing additional support staff in core clerkships could free up CDs' time to innovate, monitor educational outcomes, and disseminate their findings for the benefit of all.
The authors gratefully acknowledge the support of the Alliance for Clinical Education, in particular from Lou Pangaro, MD, and Gary Beck, who supported the entire process. Additionally, the authors thank Carlen Ribble for her able assistance with preparation of the manuscript, tables, and figure.
1 Fincher FE, Cox S, DaRosa DA, eds. Guidebook for Clerkship Directors. 3rd ed. Omaha, Neb: Alliance for Clinical Education; 2005.
2 Brodkey A, Sierles F, Weiner C, Spertus I, McCurdy F. Clerkship directors' perceptions of the effects of managed care on medical students' education. Acad Med. 2002;77:40–48.
3 Sierles F, Brodkey A, Cleary L, et al. Medical students' exposure to and attitudes about drug company interactions: A national survey. JAMA. 2005;294:1034–1042.
4 Pangaro L, Fincher RM, Bacicha J, et al. Expectations of and for clerkship directors: A collaborative statement from the Alliance for Clinical Education. Teach Learn Med. 2003;15:217–222.
5 Hemmer PA, Elnicki M, Albritton TA, et al. The responsibilities and activities of internal medicine clerkship directors. Acad Med. 2001;76:715–721.
6 Elnicki DM, Hefner J, Hemmer PA, et al. Does being clerkship director benefit academic career advancement: Results of a national survey. Teach Learn Med. 2003;15:21–24.
7 Hemmer PA, Alper EJ, Wong RY. Participation of internal medicine department chairs in the internal medicine clerkship—Results of a national survey. Acad Med. 2005;80:479–483.
8 Durning SJ, Pap KK, Pangaro LN, Hemmer P. Expectations of and for internal medicine clerkship directors: How are we doing? Teach Learn Med. 2007;19:65–69.
9 Johnson VK, Michener JL. Attitudes, experience, and influence of family medicine predoctoral education directors. Fam Med. 1994;26:309–313.
10 Gazewood J, Margo K, Jerpbak C, Burge S, Ballinger T, Usatine R. Predoctoral directors: Who are they and what do they do in these trying times? Fam Med. 2007;39:171–177.
11 Greenberg L, Sahler OJZ, Siegel B, Sarkin R, Sharkey SA. The pediatric clerkship director, support systems, professional development, and academic credentials. Arch Pediatr Adolesc Med. 1995;149:916–920.
12 White C, Greenberg L, Waller J, et al. The state of undergraduate pediatric medical education in North America: The COMSEP survey results. Teach Learn Med. 2007;19:264–270.
13 Sierles FS, Magrane D. Psychiatry clerkship directors: Who they are, what they do, and what they think. Psychiatr Q. 1996;67:153–162.
14 Kuhn TW, Cohen MJ, Polan HJ, Campbell EC, Clegg KA, Brodkey AC. Standards for psychiatry clerkship directors. Acad Psychiatry. 2002;26:31–37.
15 Shaw D, Borckardt JJ, Feldman TB. Psychiatry clerkship directors: Characteristics of the leaders in medical student education and their perceptions about the adequacy of available resources to be effective educators of tomorrow's healthcare professional. Teach Learn Med. 2006;18:181–182.
16 Magrane DM, Fenner D. A profile of directors of clerkships in obstetrics and gynecology in the United States and Canada. Obstet Gynecol. 1997;39:785–789.
17 Goepfert AR, Phelan SP, Cliver SP, Herbert WN. An evaluation of trends in obstetrics and gynecology clerkships in the United States and Canada. Am J Obstet Gynecol. 2004;191:354–358.
18 Boehler ML, Rogers DA, Schwind CJ, Williams RG, Dunnington G. Who are the surgery clerkship directors and what are their educational needs? Am J Surg. 2003;185:216–218.
19 Zun LS, Downey L. Is a third year clerkship in emergency medicine correlated with a career choice in emergency medicine? Teach Learn Med. 2004;16:14–17.
20 Coates WC, Gill AM, Jordan R. Emergency medicine clerkship directors: Defining the characteristics of the workforce. Ann Emerg Med. 2005;45:262–268.
21 Margo K, Gazewood J, Jerpbak C, Burge S, Usatine R. Clerkship directors' characteristics, scholarship, and support: A summary of published surveys from seven medical specialties. Teach Learn Med. 2009;21:94–99.
24 Lepnum R, Dobson R, Backman A, Keegan D. Factors associated with career satisfaction among general practitioners in Canada. Can J Rural Med. 2007;12:217–230.
25 Clem KJ, Promes SB, Glickman SW, et al. Factors enhancing career satisfaction among female emergency physicians. Ann Emerg Med. 2008;51:729–731.
26 Bell DJ, Bringman J, Bush A, Phillips OP. Job satisfaction among obstetrician–gynecologists: A comparison between private practice physicians and academic physicians. Am J Obstet Gynecol. 2006;195:1474–1478.
27 Kelly AM, Cronin P, Dunnick NR. Junior faculty satisfaction in a large academic radiology department. Acad Radiol. 2007;4:445–454.