Kempainen, Robert R. MD; McKone, Edward F. MD; Rubenfeld, Gordon D. MD, MSc; Scott, Craig S. PhD; Tonelli, Mark R. MD
Clinician–educator faculty play a pivotal role in patient care and medical education at U.S. medical schools. Although teaching and patient care are generally believed to be the most important aspects of this position,1–4 the extent to which clinician–educators should be required to publish in peer-reviewed journals and establish a regional or national reputation to gain promotion is unclear.1,5,6 Although department of medicine and promotion committee chairs consider clinical and teaching skills the most important aspects of a clinician–educator ’s performance,3,4 clinician–educators perceive extramural reputation, clinical research, and other forms of written scholarship as more important determinants of their promotions.5,7
Clinician–educators in primary care may be particularly disadvantaged in meeting expectations for reputation and publication compared with their specialist counterparts.5,8,9 Some authors believe the work of generalist clinician–educators is less amenable to publication and that generalists have fewer opportunities to lecture outside of their home institutions.5 We found medicine subspecialist clinician–educators in our institution had significantly greater numbers of peer-reviewed publications than did general internal medicine (GIM) faculty, but we did not identify differences in the time they had available for scholarship.10 Subspecialist clinician–educators also were more likely to give lectures outside their home institutions and to hold extramural committee memberships. These results suggest that requirements for peer-reviewed publication and reputation could result in fewer GIM clinician–educators gaining promotion. Further study, however, was needed to determine the generality of the findings.
We performed a multicenter, cross-sectional survey focusing on the research, teaching, clinical, and administrative activities of internal medicine clinician–educator faculty. The roles and expectations of clinician–educators vary substantially between institutions.1 Our primary objective was to determine whether medicine subspecialist clinician–educators had more peer-reviewed publications and higher levels of participation in extramural academic activities than did GIM clinician–educators and to seek explanations for any observed differences. For this reason, we did not address scholarly activity other than publications, and we focused on clinician–educators who had designated time available for scholarly activity and were expected to provide evidence of scholarship to gain promotion. We also sought the viewpoints of clinician–educators on the use of peer-reviewed publication and extramural reputation in determining their promotion.
To sample faculty from institutions with differing research intensities, we randomly selected five U.S. departments of internal medicine from each quartile of the 1999 ranking of research funding received from the National Institutes of Health (NIH). We contacted 25 schools to identify 20 with clinician–educator tracks. In all phases of recruitment, we used three criteria to define clinician–educator: (1) the faculty member spends much of her or his time on teaching and patient care, (2) scholarly activity is required for the faculty member to gain promotion, and (3) the faculty member is not dependent on grants to maintain his or her salary. We sent the medicine chairpersons up to three requests to submit the names of their clinician–educator faculty.
Of the 20 selected medicine departments, ten declined to participate. Of the ten participating institutions, one was in the top quartile of NIH research funding, two were from the second quartile, three were from the third quartile, and four were from the fourth quartile. A total of 577 clinician–educator faculty were identified by their departments. Fifty-one (8.8%) of these faculty were excluded from the study: 24 were members of divisions of dermatology, neurology, or pathology; 12 held a rank below assistant professor; five were on leave, retired, or emeritus; five were nonphysicians; and five had no current e-mail or mailing address. The remaining 526 faculty were included in the study. The number of participants per institution ranged from five to 126. The participating faculty were asked which of the following most closely characterized their current position: Choice A: “Nearly all of your time is spent on direct patient care or administration and scholarly activity is not a requirement for promotion”; Choice B: “The majority of your time is spent doing clinical or bench research and grant support is required to maintain your salary”; or Choice C: the study definition of clinician–educator as outlined above.
We devised a 34-item questionnaire that covered five domains. The first domain asked for demographic information. The second domain addressed time available for scholarly activity and use of publications and extramural reputation as promotion criteria. Participants indicated the extent to which they agreed with a series of statements using a five-answer scale ranging from “strongly disagree” to “strongly agree.” The third domain requested information on coursework and degrees following medical school. The fourth domain asked participants to quantify their teaching, clinical, and administrative activities. In the final domain, participants indicated the numbers and types of publications listed on their curriculum vitae and estimated the proportion of their scholarly time spent on activities that “were unlikely to result in peer-reviewed publications (e.g., curriculum design, development of new teaching methods)” on a four-part scale: 0% to 25%, 26% to 50%, 51% to 75%, or > 75%. This item is hereafter referred to as “focus of scholarship.”
The questionnaire was sent in 2002, and all potential subjects received $5.00 along with a letter describing the study. For nine of the ten institutions, participants subsequently were sent an e-mail invitation to complete the questionnaire online. After four weeks, nonrespondents were sent an e-mail reminder encouraging their participation. Faculty without a known e-mail address or who were unable to access the study's Web site were sent hard copies of the questionnaire. All nonrespondents were mailed a third request along with a hard copy of the questionnaire. For one institution, all requests to participate and questionnaires were mailed. Questionnaires were coded to ensure confidentiality.
Funding for this research was provided by the University of Washington's Department of Internal Medicine, which had no role in the design and conduct of the study. The University of Washington Institutional Review Board approved the study.
Characteristics of responders and nonresponders were compared using the Pearson chi-square statistic. The survey's data were compared using the survey commands in Stata 8.0, with institution weighted as the primary sampling unit. Means and standard deviations were compared using t tests. Categorical variables were compared using the Pearson chi-square statistic.
Multiple linear regression was used to identify predictors of number of publications. Publications were log-transformed to reduce heteroscedasticity (unequal variance) of residuals. Predictor variables of interest were age, subspecialty training, gender, full-time equivalent, years in academics, years in a clinician–educator track, academic rank, inpatient and outpatient clinical duties, focus of scholarship, having acquired an advanced degree following completion of medical school, quartile of NIH funding (as outlined above), and number of committee memberships. All variables were included in the initial regression model, and statistically significant predictors were identified using backward stepwise modeling. Statistical significance was defined as a two-tailed p value of < .05. All statistical analysis was carried out using Stata 8.0 software (Stata Corporation, Texas).
The overall response rate was 69% (363/526), with a comparable proportion of GIM and subspecialist faculty responding (p = .35). Response rate by institution ranged from 33% (3/9) to 100% (5/5). Table 1 compares characteristics of responding and nonresponding faculty. Of responding faculty, 93 (25%) did not meet the study definition of clinician–educator. Fifty-three (15%) selected “Nearly all of your time is spent on direct patient care or administration and scholarly activity is not a requirement for promotion,” and 36 (10%) selected “The majority of your time is spent doing clinical or bench research and grant support is required to maintain your salary” as best descriptions of their positions.
The responses of the remaining 270 faculty were included in the analysis. Their demographic characteristics are summarized in Table 2. There was a significant difference in academic rank between groups, with a surplus of assistant professors among GIM clinician–educators and a larger proportion of professors among subspecialists. Among assistant professors, GIM faculty had spent a mean of 7.5 years in a clinician–educator track compared with 4.9 years for subspecialists (p = .09). The number of years in a clinician–educator track was comparable between groups at the associate professor and professor level (p = .43 and .28, respectively).
Table 3 summarizes respondents’ clinical, teaching, and administrative duties. On average, subspecialist clinician–educators spent 50% more time attending on inpatient services than did GIM clinician–educators (p = .04). Clinic time, which included outpatient procedures, did not differ significantly between groups (p = .13). The groups did not differ in the number of committee memberships held at the local, regional, or national/international level. Clinician–educators provided a large number of didactic sessions exclusive of teaching during inpatient rounds. Nearly a third averaged one such teaching session per week throughout the year. Although local teaching activities of generalists and subspecialists were comparable (p = .36), subspecialists were more likely to speak at a regional meeting or give grand rounds outside their home institution in the preceding year (p = .05 and < .001 respectively). Seventy-six percent of subspecialists and 69% of generalists “strongly disagreed” or “disagreed” with the statement “The amount of time you have available for scholarly activity is adequate” (p = .15 for comparison between groups).
Table 4 compares the number of publications of generalist and subspecialist clinician–educators. Subspecialists reported more than twice as many total and peer-reviewed publications as did GIM faculty. These differences remained significant when limiting the comparison to publications produced while in a clinician–educator track. On linear analysis, significant predictors of having a greater total number of publications were being a subspecialist (p < .01), having fewer half-days per week performing outpatient care (p < .01), academic rank (p < .01), higher quartile of NIH funding received by respondent's department (p < .01), focus of respondent's scholarship (p < .01), years on faculty (p = .02), and full-time equivalent (p = .02). Significant predictors of greater numbers of peer-reviewed publications were being a subspecialist (p < .01), having fewer half-days per week doing outpatient care (p < .01), academic rank (p < .01), higher quartile of NIH funding received by respondent's department (p < .01), years on faculty (p < .03), and focus of scholarship (p < .01). Age, gender, years in a clinician–educator track, inpatient attending duties, and number of committee memberships were not predictive of publication. Overall, specialists and generalists did not differ significantly in the proportion of protected time spent on activities that “were unlikely to result in peer-reviewed publications (e.g., curriculum design, development of new teaching methods)” (p = .14; see Figure 1). However, a larger proportion of GIM faculty spent more than 50% of their protected time in this fashion (44% versus 27%, p = .05).
Figure 2 summarizes clinician–educators’ views on the use of publications and reputation in determining their promotion. A minority (38%) of subspecialists felt there should be a requirement for original research publications and an even smaller proportion of GIM faculty agreed (18%, p < .01). Most subspecialists (77%) agreed that some form of peer-reviewed publication should be required compared with 44% of generalists (p < .01). More generalist and subspecialist clinician–educator approved the use of regional reputation as a promotion criterion than opposed it (p = .22 for comparison between groups), but GIM respondents were more divided on this issue.
This multicenter study examined the publications and extramural activities of clinician–educators and their views regarding promotion criteria. Interestingly, one-quarter of the respondents did not meet the study's definition of clinician–educator despite being identified as such by their department heads. This may reflect the institutions’ varied definitions and expectations of their clinician–educator faculty, movement of faculty between tracks, or faculty members’ lack of understanding of scholarly expectations. Among the clinician–educators meeting the study's definition, internal medicine subspecialists reported more than twice as many peer-reviewed publications as did GIM faculty, and subspecialty membership was an independent predictor of publication. On average, subspecialist clinician–educators gave more lectures outside their home institution than did generalists. The two groups held comparable numbers of extramural committee memberships.
Our study did not identify a definitive explanation for the higher rate of publication among subspecialist clinician–educators. We did not detect any overt differences in time available for scholarship. Both groups felt they had inadequate time for scholarship, subspecialists attended more on inpatient services, and clinic, administrative, and teaching duties did not differ significantly. Overall, more half-days in clinic per week predicted fewer publications. The fact that clinic time, but not inpatient attending duties, was associated with publication rates suggests outpatient attending duties may be more time-consuming or otherwise disruptive to scholarly pursuits.
Academic rank and focus of scholarship, both independent predictors of publication, may also account for some of the difference. A greater proportion of subspecialist clinician–educators held the rank of professor, and a larger proportion of GIM faculty spent more than 50% of their protected time on activities they perceived as unlikely to result in peer-reviewed publications. Although innovations in curriculum design and teaching are commonly reported in the literature, most routine curricular changes lack sufficient novelty to merit publication. Subspecialists continued to publish significantly more than generalists even after entering a clinician–educator track, making it unlikely that research generated during fellowship or while in traditional researchers’ tracks accounts for the finding. Previous studies found gender (male) was predictive of publication,11,12 but the greater proportion of male subspecialists in this study was not significant, and gender was not predictive of publication in our sample.
The advent of clinician–educator tracks has created a new challenge for promotion committees. Although excellent teaching and clinical skills are widely viewed as the most important attributes of being a clinician–educator,1–4 currently available tools for measuring these competencies may be inadequate or not uniformly applied.1,4–6 Committee members must also be proficient in evaluating the diverse scholarship recognized as appropriate for clinician–educators.13,14 These potential barriers to evaluation have raised concerns among clinician–educators that more readily quantifiable criteria, such as peer-reviewed publications, grant funding, and national reputation, play a central role in determining their promotion.1,4–7 In fact, a majority of recently surveyed department of medicine and promotion committee chairs specified a minimum number of peer-reviewed publications expected of clinician–educators seeking promotion.3,4
A single-institution study from the 1980s found clinician–educator internal medicine faculty with more publications were more likely to be promoted.15 Among assistant professors, GIM faculty had spent an average of 2.6 years longer than subspecialists in a clinician–educator track (p = .09), suggesting they may be experiencing a relative delay in gaining promotion. However, subspecialists with small numbers of publications may be more likely to leave academics, and the overall rates of promotion may be similar. Regardless, the differences in publications, extramural academic activities, and focus of scholarship we observed lend credence to the belief that requirements for peer-reviewed publication and extramural reputation may be particularly problematic for faculty in generalist fields. In our study, clinician–educators did not agree on what role peer-reviewed publications and extramural reputation should have in determining their promotion. Subspecialists were divided over a requirement for publication of original research, but a strong majority agreed with expectations for at least some form of peer-reviewed publication. In contrast, a strong majority of generalists were opposed to requirements for original research, but the group was split in its opposition to a requirement for peer-reviewed publication of all types. Not surprisingly, these opinions regarding promotion coincided with the publication rates of the two groups. GIM and subspecialist clinician–educators did not differ significantly in their support for a requirement for regional reputation.
The study has a number of limitations. Because our sample contained many institutions with lower-than-average amounts of NIH funding, our findings are potentially less applicable to clinician–educators at research-intensive institutions. The research intensiveness of a respondent's home institution was an independent predictor of peer-reviewed publication. However, we previously compared generalists and subspecialists at an institution within the top quartile of NIH funding and found similar results.10 Furthermore, our response rate compares favorably with previously published surveys of physicians.16
Another consideration is that the study's definition of clinician–educator does not apply to one quarter of the faculty in department-defined clinician–educator tracks. Our intent was to limit the scope of the study to faculty who serve predominantly as clinicians and teachers, yet who could still be substantially affected by the use of publication and extramural reputation as promotion criteria. Hence, we excluded clinician–educators with full-time clinical and/or administrative duties and no expectation for scholarship as well as faculty whose activities closely resembled those of traditional physician–scientists. We did not verify the accuracy of respondents’ self-reported work and scholarly activities, nor did we attempt to measure the time demands of various duties. Certainly, the relative impact of outpatient and inpatient attending duties on scholarly productivity merits further exploration. Lastly, our study explored publications of clinician–educators in detail but was not designed to delineate the diverse spectrum of other scholarly activities that might be performed by clinician–educators. Products of scholarly activity other than peer-reviewed publication, such as curriculum design, are often of significant value to an institution, though it is not clear how promotion committees value such contributions.
This study constitutes a substantial addition to the published data on the work of clinician–educators.10,14,17–19 Further research along similar lines ultimately may serve as a rough guide for establishing realistic expectations for clinical service and scholarship. However, promotion criteria must have sufficient flexibility to accommodate individual faculty whose diverse work and scholarly activities fall outside the realm of traditional research. How well institutions are meeting this relatively new challenge is unclear, and further study is needed to determine the extent to which peer-reviewed publications and national reputation dictate whether clinician–educators gain promotion.
Our results suggest that if peer-reviewed publications and extramural reputation are, in fact, pivotal determinants of promotion, GIM clinician–educators may be less likely than their subspecialist colleagues to gain promotion. Based on our analysis of the focus of scholarship, the loss of GIM clinician–educators could, in turn, limit the degree to which institutions approach the integration, application, and teaching of medical knowledge in a scholarly fashion. These results underscore the need for promotion committees to rigorously pursue better ways of evaluating the contributions of their faculty apart from original research, and for institutions to seek out and address barriers that may be limiting the scholarly productivity of primary care clinician–educators. Such efforts will help ensure that medical schools fully benefit from the contributions of all faculty members.
1 Levinson W, Branch WT Jr, Kroenke K. Clinician–educator in academic medical centers: a two-part challenge. Ann Intern Med. 1998;129:59–64.
2 Jones RF, Gold JS. Faculty appointment and tenure policies in medical schools: a 1997 status report. Acad Med. 1998;73:212–9.
3 Beasley BW, Wright SM, Cofrancesco J Jr, Babbott SF, Thomas PA, Bass EB. Promotion criteria for clinician–educator in the United States and Canada. A survey of promotion committee chairpersons. JAMA. 1997;278:723–8.
4 Atasoylu AA, Wright SM, Beasley BW, et al. Promotion criteria for clinician–educator. J Gen Intern Med. 2003;18:711–6.
5 Levinson W, Rubenstein A. Mission critical- integrating clinician–educator into academic medical centers. N Engl J Med. 1999;341:840–3.
6 Lucey CR. Promotion for clinician–educator: time for a fresh approach? J Gen Intern Med. 2003;18:768–9.
7 Beasley BW, Wright SM. Looking forward to promotion: characteristics of participants in the prospective study of promotion in academia. J Gen Intern Med. 2003;18:705–10.
8 Culpepper L, Franks P. Family medicine research. Status at the end of the first decade. JAMA. 1983;249:63–8.
9 Weiss BD. Publications by family medicine faculty in the biomedical literature: 1989–1999. Fam Med. 2002;34:10–6.
10 Kempainen RR, McKone EF, Rubenfeld GD, Scott CS, Tonelli MR. Comparison of scholarly productivity of general and subspecialty clinician–educator in internal medicine. Teach Learn Med. 2004;16:323–8.
11 Carr PL, Ash AS, Friedman RH, et al. Relation of family responsibilities and gender to the productivity and career satisfaction of medical faculty. Ann Intern Med. 1998;129:532–8.
12 Tesch BJ, Wood HM, Helwig AL, Nattinger AB. Promotion of women physicians in academic medicine: glass ceiling or sticky floor? JAMA. 1995;273:1022–5.
13 Boyer EL. Scholarship Reconsidered: Priorities of the Professoriate. Princeton, NJ: The Carnegie Foundation for the Advancement of Teaching, 1990.
14 Lubitz RM. Guidelines for promotion of clinician–educator. The Society of General Internal Medicine Education Committee. J Gen Intern Med. 1997;12(2 suppl):S71–8.
15 Batshaw ML, Plotnick LP, Petty BG, Woolf PK, Mellits ED. Academic promotion at a medical school. Experience at Johns Hopkins University School of Medicine. N Engl J Med. 1998;318:741–7.
16 Asch DA, Jedrziewski MK, Christakis NA. Response rates to mail surveys published in medical journals. J Clin Epidemiol. 1997;50:1129–36.
17 Sheffield JV, Wipf JE, Buchwald D. Work activities of clinician–educator. J Gen Intern Med. 1998;13:406–9.
18 Sherertz EF. Criteria of the “educators’ pyramid” fulfilled by medical school faculty promoted on a teaching pathway. Acad Med. 2000;75:954–6.
19 Hafler JP, Lovejoy FH Jr. Scholarly activities recorded in the portfolios of teacher-clinician faculty. Acad Med. 2000;75:649–52.