We determined two general categories regarding the clarity and depth of career path descriptions. Nineteen schools outlined between one and three career tracks that were not well defined in the faculty handbooks. The most common theme in this group was to offer two tracks—a “tenure track” and a “nontenure track”—whose responsibilities varied between schools. For example, at some schools the tenure track was available for “triple threat” faculty, whereas the nontenure track was offered to everyone else. At other schools the tenure track was provided to the majority of MD and PhD faculty, whereas the nontenure track was primarily for PhD researchers who worked in other faculty members' labs. Because these tracks were generally considered “one size fits all,” it was not possible to analyze them according to faculty participants' major professional activities.
Seventy-nine schools (81%) had between one and eight clearly defined tracks, each with a “major professional activity,” in which 50% to 85% of an individual's time is to be spent on either research, teaching, or clinical care. Together, these schools clearly described 172 research career tracks, 113 education tracks, and 39 clinical tracks. At least one research and one clinical track was offered at 78 of the 79 schools. The remaining school offered only a single all-inclusive option. Thirty-four of the 79 institutions offered an education track.
Among the 98 institutions we studied, tenure-eligible research tracks were generally allocated to MD or PhD faculty members with between 50% and 85% protected research time, and the rest of their time split between teaching, clinical, and administrative responsibilities. Nontenured research tracks usually were created for PhD faculty members who did not run their own labs and whose nonresearch contributions took up between 0% and 10% of their time. Of the 172 research tracks, 91 (53%) were tenure-eligible.
Clinical tracks at the 98 schools were generally reserved for MD faculty members whose time was spent providing clinical care 50% to 100% of the time. Of the 113 distinct clinical tracks described, 28 (25%) were tenure-eligible.
Education tracks were offered at 34 of the 98 schools. Most were open to both MD and PhD faculty members. In these positions, approximately 50% to 85% of a faculty member's effort was devoted to educational activities, with some descriptions including educational scholarship and educational administration into this time. Of the 39 education tracks, 16 (41%) were tenure-eligible.
Faculty title differentiation.
Of the 98 schools studied, 60 (61%) practiced title differentiation, defined as delineating faculty members by different titles depending on the track being pursued. Schools that offered a greater number of tracks were more likely to differentiate titles for nontenured positions (F1, 96 = 18.658, P < .001). In addition, a stepwise multiple regression analysis revealed that non-community-based schools were more likely than community-based schools to differentiate titles (r = 0.226, P < .01). However, this variable accounted for only about 0.07% of the variance. None of the other institutional characteristics were predictive of whether a school differentiated titles.
Of the 60 schools (61%) that differentiated titles, 59 differentiated their nontenured research track faculty, who devoted 85% to 100% of their time to research, with little or no teaching or clinical responsibilities. The title “Research Professor” was used at 46 of these 59 schools, whereas 13 of the 59 schools used a parenthetical or dashed suffix to differentiate the title (e.g., “Professor (Research Track),” “Professor (RT),” or “Professor–Research Track”).
Of the same 60 schools, 41 used modified titles for their nontenured clinical faculty (both clinician–educators and the more strictly focused clinicians). Of these 41 schools, clinical faculty were denoted with the following titles: “Professor of Clinical X” (13 schools), “Clinical Professor of X” (14 schools), and “Professor of X (Clinical Track/CT)” (8 schools). Additionally, 1 school used a combination of all three of these options, whereas 5 schools used either a comma (,), forward slash (/), or hyphen (-) to create a clinical track suffix (e.g., “Professor - Clinical Track,” “Professor/Clinical Track,” or Professor, Clinical Track”). Although this project did not focus on voluntary faculty tracks or titles, at least 11 of these schools appointed their voluntary faculty with the title “Clinical Professor of X,” the same title given to full-time, nontenured faculty at 14 other schools.
Of the 60 schools that differentiated titles, 30 schools modified titles for all nontenured faculty regardless of major professional activity. Nine schools differentiated titles for some nontenured faculty (usually clinical and research track faculty) while leaving others (usually clinician–educators, education, or clinician–researchers) with regular titles. Eighteen schools used modified titles only for research track faculty, leaving all other nontenured faculty with regular titles. One school had no tenure system, yet still modified titles for one-third of their faculty tracks. One school's documents did not clarify exactly which faculty members' titles were modified, and one school's documents became unavailable after initial data collection, so clarification was not possible.
Institutional characteristics and number of tracks offered.
No significant differences in the number of tracks offered were found based on a medical school's relation to a university or between public and private schools. However, differences were found in the number of tracks based on faculty size, degree of community focus, and research intensity.
* Faculty size. Univariate tests indicated a significant difference in the number of clinical tracks offered among schools of different faculty sizes (F2, 76 = 3.174, P < .05). Post hoc analysis further indicated that schools with faculty sizes greater than 1,000 had significantly fewer clinical tracks than those with faculty sizes less than 500.
* Community-based schools versus non-community-based schools. Community-based institutions offered a significantly lower total number of career tracks than non-community-based schools (F1, 96 = 4.77, P < .05). Further, when career tracks were analyzed by major professional activity, the number of clinical tracks offered at community-based institutions was significantly lower (F1, 77 = 5.773, P < .05). No differences were found between community-based and non-community-based schools for either research or educational tracks.
* Research-intensive schools versus non-research-intensive schools. Research-intensive schools, defined as the 40 schools receiving the most research funding from the National Institutes of Health (NIH), offered significantly more clinical tracks than schools in the bottom third of NIH funding (F3, 76 = 5.563, P < .01). In addition, these clinical tracks at research-intensive schools were more often tenure-eligible than at non-research-intensive schools (F2, 76 = 2.633, P < .05).
Discussion and Conclusions
Overview of the study findings
The goal of the current study was to describe the faculty career tracks at U.S. MD-granting medical schools and explore which institutional characteristics are associated with the types of career tracks offered. Although schools vary widely in their approach to delineating faculty career tracks, most schools (53; 54%) offered between three and four career tracks. Previous research has demonstrated a long-standing trend of schools adding new faculty tracks, suggesting that these numbers are probably higher than the average number of career tracks at institutions 30 years ago.3,6,7 There was a stark difference in the specificity and clarity of documentation from one institution to the next. Of the 98 schools whose data we analyzed, descriptions were vague and basic for 19 (19%). Among the 79 schools (81%) with clearly outlined career tracks, many more research tracks were offered than education or clinical tracks. This likely reflects the importance of research to academic institutions. However, there has been a recent rise in the number of faculty members pursuing clinical tracks,1 and our data show that 78 of 79 schools with clear promotion criteria (99%) offered clinical tracks. Our data also show that 34 out of 79 schools (43%) offered an education track; however, it is unclear whether this is an increase or decrease from previous years because of a lack of historical data.
Our study of the 98 schools' data revealed five general tracks as the most universal—the investigator track, the research track, the clinician–educator track, the clinical track, and the educator track.
The investigator track is most commonly referred to by schools as the “tenure track” or the “traditional track.” This track is for faculty members who spend greater than 50% of their time conducting research (generally 80%), with the remaining time split between clinical care, teaching, and administration based on institutional needs and the faculty member's interests and responsibilities. Most commonly, this faculty member is expected to obtain extramural funding for research projects, conduct and design valid, verifiable studies, publish results in respected, peer-reviewed journals, and supervise and mentor trainees. These tracks do not carry modified titles and are almost universally tenure-eligible (except at institutions where a tenure system is not available).
The research track is for faculty members who spend between 85% and 100% of their time conducting research, with little or no teaching or clinical responsibilities. Whereas a faculty member on this track may function as the principal investigator or as a team member on a research project, there is generally less expectation of obtaining extramural research funding. These tracks are generally occupied by PhD faculty members who wish to focus almost exclusively on their research endeavors. The research track is nontenured and most often carries a modified faculty title (e.g., “Research Professor”).
The clinician–educator track is a track for faculty members who spend greater than 50% of their time in direct patient care (generally 75%–80%), with the remainder devoted to education, research, or administration. As the name implies, the secondary professional activity is usually education (10%–20% of faculty time), which may vary from teaching residents and medical students on the wards, teaching a basic medical science course, or running a clerkship. Usually, a small amount of time (5%–10%) is devoted to research, which is usually clinical or educational. These tracks are most often nontenured and sometimes carry a modified title (e.g., “Clinical Professor”).
The clinical track is for physicians who wish to spend the vast majority of their time providing health care services (from 90% to 100% of the time) and participate only in very limited teaching and research. Also known as “clinical faculty” and “clinical attendings,” they are extremely valuable to academic health centers because of the reimbursements their activities provide. Similar to clinician–educator tracks, clinical tracks are most often nontenured and frequently carry a modified title (e.g., “Clinical Professor”).
The educator track is for faculty members who spend greater than 50% of their time focused on educating medical students and residents. These physicians spend the rest of their time on some combination of patient care, research, and administration. A physician in this track might occupy the majority of his or her time being the course director for first- and second-year medical students, and may teach residents in the wards. These faculty members often conduct medical education research and produce scholarly educational resources for students, such as course handouts and book chapters. Approximately 40% of these tracks are tenure-eligible, and they only very rarely carry a modified title.
Tenure and its propriety in academic health centers have been hotly debated for some time.13,14 Our study shows that 143 (41%) of the 353 clearly described career tracks were tenure-eligible, with the majority being research tracks, although a number of schools offer tenure for education and clinical tracks. Whereas research tracks traditionally have been tenure-eligible, the increase in tenure eligibility for education and clinical tracks may indicate an increase in the explicit value institutions hold for the other major activities in academic medicine. Additional research is needed to document the number of faculty in these tenure-eligible tracks and how many actually receive tenure. This study did not address the various meanings of tenure from one institution to the next (e.g., job security, financial security).
Approximately two-thirds of schools used modified titles to differentiate faculty members in different career tracks, and this practice was more likely at schools with greater diversity of career tracks. There was considerable variance in the methods for modifying titles (clinical or research prefix versus clinical suffix versus parenthetical track abbreviation versus parenthetical/hyphenated track name). Surprisingly, at many schools a “clinical professor of medicine” is a full-time, clinically focused career track, whereas at an almost equal number of schools this is the same title given to a voluntary faculty member (nonsalaried). This difference adds complexity to interpretation of modified titles and further prompts the question, Why and for whom do schools use modified titles?
Analysis of institutional characteristics in relation to faculty track data showed that no single institutional characteristic was significantly predictive of the way a school structured its faculty tracks. This lack of association suggests that the creation of faculty tracks is influenced by more complex factors than simple institutional characteristics. More work in needed to assess the influence of variables such as institutional history, culture, politics, and faculty demographics.
Interestingly, faculty size, research orientation, and community orientation were significantly associated with a small number of dependent variables. First, schools with faculty sizes greater than 1,000 had significantly fewer clinical tracks than those with faculty sizes less than 500. This seems somewhat counterintuitive, given that larger faculty sizes likely include a large number of clinical faculty who might be interested in pursuing clinical tracks. Perhaps, however, smaller-sized schools have been able to cater their promotion criteria more fully to their clinical faculty because they had fewer cultural and/or administrative constraints to overcome. Second, research-intensive schools provided significantly more tenured clinical tracks. It is unclear why research-intensive schools are finding it favorable to make clinical tracks available with tenure eligibility. This may be an institutional initiative to elevate the status of clinical faculty and prevent the “second-class” stigma that can denigrate nonresearchers at a research-intensive university. Alternatively, there may be greater numbers of clinical faculty at research-intensive institutions who are pushing for these types of career tracks. Currently, clinical tracks have the lowest percentage of tenure eligibility. More research would uncover whether there is an upward or downward trend in this number.
Community-based schools generally offered fewer clinical tracks (both tenured and nontenured) than do non-community-based institutions. This finding was surprising, given that community-based schools have a large focus on clinical care. Perhaps at these schools one flexible clinical option has been sufficient, given that research is not the primary concern and promotion criteria might not be dominated so fully by academic scholarship.
Contributions and limitations of the present study
Our study is unique in several aspects. First, this is the only study to conduct detailed content analysis of individual faculty policies across a large number of institutions. Previous studies of career tracks used surveys of school administrators or examined just one medical school.2–7 With detailed analysis of the majority of medical schools, our conclusions provide institutions with a more comprehensive understanding to inform meaningful conversations about future faculty policy decisions. Second, this research analyzed institutional characteristics and their relationships with faculty tracks.
There are several limitations in this research. We analyzed career tracks in academic medicine based on information collected from available faculty policies. Despite our explicit request for the status of these policies, some policies may be outdated. Forty of the 98 schools verified that their policies were, in fact, the most current ones available, which does give us some confidence that the policies were not grossly outdated. Other considerations are that (1) actual institutional practices may vary from published policies, (2) individual departmental policies may vary from published institutional policies, and (3) more specialized faculty members may not fit into published roles but may still be employed appropriately. Also, we were able to collect only faculty policies from 98 of the 129 U.S. MD-granting medical schools accredited as of July 2008, representing 76% of these institutions. Although a sample of over 75% is usually sufficient to generalize to the population, the variability of schools that we analyzed suggests that those schools that we did not analyze may be different in some additional ways. Since July 2008, more MD-granting schools have been accredited, and our results do not include those institutions. Despite these limitations, we believe that these results are meaningful and would be upheld if our research were replicated using data from more schools.
Ideas for future research
We see a number of important directions for this line of inquiry. Publicly available information about medical school career tracks would facilitate a replication and extension of this work. It will be interesting to see whether the trend in increasing numbers of education tracks continues. It would be worthwhile to examine the extent to which individual institutions follow their own faculty policies, to analyze the degree to which departmental policies align with institutional policies, and to investigate the results of promotion decisions in each of these tracks relative to institutional characteristics. The AAMC should consider expanding the kinds of information it collects from schools about career tracks to include variables from this study.
The authors would like to thank Valarie Clark and Sara Whiteley of the AAMC for their assistance in procuring faculty policies from participating institutions, the AAMC's Group on Faculty Affairs (in particular, the research subcommittee for their support of this project), and Dr. Emil Petrusa of Vanderbilt University for his review of this manuscript. We also thank Sarah A. Bunton of the AAMC and Dr. Scott Rodgers of Vanderbilt University for their guidance in the early stages of this research project's design.
(1) Coleman (Sullivan) M, Richard GV. “Faculty career tracks at U.S. allopathic medical schools.” Oral presentation, AAMC annual meeting; November 2008; Washington, DC. (2) Coleman (Sullivan) M, Richard GV. “Faculty career tracks and associated institutional characteristics at allopathic medical schools.” RIME abstract oral presentation, AAMC annual meeting; November 10, 2010; Washington, DC.
4 Beasley BW, Simon SD, Wright SM. A time to be promoted. J Gen Intern Med. 2006;21:123–129.
8 Kelley WN, Stross JK. Faculty tracks and academic success. Ann Intern Med. 1992;116:654–659.
10 Williamson JC, Schrop SL, Costa AJ. Awarding faculty rank to non-tenured physician faculty in a consortium medical school. Fam Med. 2008;40:32–39.
11 Association of American Medical Colleges. AAMC Statement on the Physician Workforce. Washington, DC: Association of American Medical Colleges; June 2006.
13 Clawson D. Infrastructure: Tenure and the future of the university. Science. 2009;324:1147–1148.
© 2011 Association of American Medical Colleges
14 Marks ES. Does tenure still fit the needs of medical school faculty? Am J Med. 2000;108:751–754.