Demographic Characteristics of US Medical School Admission Committees FREE

Medical school admission committees are charged with selecting the next generation of physicians. The methods and results of the selection process are well studied, but the demographic characteristics of the committees have not been frequently examined. The most recent nationwide surveys were published in 1972¹ and 1984,² and a more recent study³ was limited in scope.

Since the publication of the earlier comprehensive studies, a number of new issues have confronted medical educators. Among these are the growing incentives to train primary care physicians, the rise of managed care and evidenced-based medicine, and the desire to produce greater racial and ethnic diversity in the physician workforce, as evidenced by the Association of American Medical Colleges' (AAMC's) Project 3000 by 2000 initiative.⁴ However, there has never been a nationwide survey of both the sex and underrepresented minority (URM) composition of US medical school admission committees.

We mailed a survey questionnaire with 19 questions regarding the demographic characteristics of the admission committee to the director or dean of admissions at the 122 AAMC member medical schools within the continental United States. A second mailing was sent to those institutions that did not reply within 60 days. The survey asked respondents to report on their committee with respect to total membership, sex ratio, URM membership (African American, Mexican American, Mainland Puerto Rican, Native American Indian, Native Hawaiian, and Native Alaskan), the academic career track of members, committee compensation status, medical student membership, allied health professions membership (those with PA, NP, RN, LCSW, PT, or OT degrees), and at-large or community membership.

The survey and accompanying cover letter specified that the investigators were seeking information specifically regarding the "voting members" of admission committees, ie, those individuals who are involved in the final selection process for applicants who are offered admission. In completing the survey, we asked respondents to refer to the committee that was convened to select the incoming first-year class for the 1999-2000 school year.

Admissions directors were assured that their responses would be kept confidential and that data analysis would be anonymous.

Of the 122 medical schools surveyed, 85 replied to either the first or second mailing for an overall response rate of 70%. A summary of the data is shown in Table 1. A composite portrait of the average US medical school admission committee appears in the first column of the table.

On average, there were 4.1 URM members (16%) per committee. Eleven schools (13%) had no URM members, 13 had 1, and 17 had 2, ie, 29% have 1 or fewer URM members, and 49% have 2 or fewer. One survey respondent did not provide data for this category.

The URM physicians comprised 54% of total URM committee membership, with a mean of 2.2 (8%) per institution. The range was 0 to 18, with a mode of 0. Fifty-seven committees (69%) had 2 or fewer, 42 (51%) had 1 or fewer, and 21 (25%) had no URM physician members. Non-URM physician members comprise the majority of admission committee members, averaging 14.5 members (55%) per committee. They comprised 65% of non-URM membership. This is statistically significant when compared with the 54% of URM committee members who are physicians (P<.001).

The mean number of physicians (MD or DO degrees) per committee was 16.7 (63%), with a range of 3 to 51. Of these, on average 8.3 (32%) were tenure-track faculty, 5.3 (20%) were clinical faculty, and the remainder were either volunteer or retired or emeritus faculty. Primary care physicians (defined for survey purposes as general internal medicine, family practice, or pediatrics) comprised 6.6 (40%) of physician membership and 25% of overall committee membership.

The mean number of female physicians per committee was 5.5 (21%), with a mode of 4. More than one third of medical schools (36%) have 3 or fewer female physicians, and more than two thirds (70%) have 5 or fewer female physicians. The ratio of men to women was 1.77 to 1, with an average of 9.5 women (36%) and 16.8 men (64%) per committee. Each medical school had at least 1 woman on the admission committee. Women constituted a majority at 6 schools (7%) and were represented in equal numbers at 2 (2%).

Medical student membership averaged 3.9 (15%) per committee, with a range of 0 to 21. Nearly three quarters (74%) of institutions have at least 1 medical student on their admission committee. Seventy-seven (91%) respondents reported that their committee provided no compensation to its members.

The mean number of members with PhD degrees (without an MD or DO degree) was 5 (19%) per committee. Of these, 3.7 (14%) on average were basic scientists, and 1.3 (5%) were clinical and/or social scientists.

The number of nonphysician, non-PhD dean's office personnel who hold voting positions on admission committees averaged 0.64 (2%) per committee. The range was 0 to 5, with a mode (58% of respondents) of 0 and a total of 54 members in this sample. Allied health professionals may represent some overlap with the dean's office category. They average 0.11 members (0.4%) per committee, with a mode (89% of respondents) of 0, a range of 0 to 2, and a sample total of 9. Community membership on committees averaged 0.45 members (2%), with a mode (75% of respondents) of 0, a range of 0 to 9, and a sample total of 38.

The final category in the survey was "other," and respondents were asked for a brief description of the members so designated. The mean for this category was 0.21 (0.8%), with a mode (86% of respondents) of 0 and a range of 0 to 4. Twelve institutions (14%) had at least 1 committee member in this category. A single university had a director of disabled students, an assistant professor of English, the director of a center for academic excellence, and the assistant director of alumni development on its admission committee. Other responses included 1 director of minority affairs, 3 pastoral care service members, 1 dentist (DMD), 1 member with an EdD degree, 2 with combined JD-MPH degrees, 1 with a combined JD-RN degree, 1 with an MA degree, and 1 with an MEd degree. Finally, 1 institution has created the position of visiting clinician, a role filled from a pool of community physicians who precept medical students and conduct applicant interviews and in return receive continuing medical education credit.

This is the first study in 28 years to report a comprehensive survey of the composition of admission committees in US medical schools. One focus of this study was URM membership in medical school admission committees. We found that on average 16% of committee members have URM status and that 87% of schools have at least 1 member with URM status on their committee. These statistics represent substantial gains since 1972, when just 55% of medical schools had 1 or more URM committee members, and African Americans, Mexican Americans, Puerto Ricans, and American Indians comprised just 6.1% of admission committee members.¹ A more recent study surveyed the 15 AAMC schools with the greatest percentage of URM students and found that URM representation on admission committees at schools that are not traditionally African American ranged from 10% to 25%.³

One trend that is reflected in the survey is the rise in the number of women in medicine. In 1982, women comprised 23% of committees and 31% of medical school freshmen²; today those percentages have grown to 37% and 44%,⁵ respectively. In 1972, women comprised just 8% of admission committees,¹ reflecting an increase of more than 300% during the past 28 years.

Medical students currently hold positions on 74% of admission committees, the same percentage as in 1982.² Prior surveys had shown that students were represented on just 56% of committees in 1972 and virtually none in 1957.⁶ Further study is needed to assess the effect of medical student membership on admission committees.

The arguments in favor of working toward racial and ethnic diversity in the US physician workforce have been made elsewhere.^7- 11 In the present climate, a reversal of historical gains in URM medical school matriculation is taking place,^12- 13 and the barriers to the promotion of diversity in medical education continue to grow.¹⁴ Investigators have concluded that the composition of a medical school class reflects the composition of the admission committee, particularly in terms of selecting students with an interest in primary care.^15- 16 This leads us to speculate that recruitment of people with URM status to sit on admission committees is a potential method of promoting diversity in the physician workforce. Our own experience at the University of Utah confirms that URM committee members are frequently able to offer salient interpretations and clarifications of the cultural factors relevant to an applicant's life history. Further study is needed to examine the relation between URM admission committee membership and the outcome of the admissions process.

Our study has some limitations. The 3 AAMC member institutions located in Puerto Rico were not surveyed because we anticipated that they would have a disproportionately large number of URM (Mainland Puerto Rican) committee members. In contrast, traditionally African American medical schools were included in the sample. A separate statistical analysis controlling for their inclusion is not feasible, because survey respondents were assured anonymity. We believe that medical school admission committees in our sample would appear even less diverse than they do if such an analysis were performed. The data ranges collected for some aspects of committees are large (Table 1). There are several possible explanations for this. First, respondents may have mistakenly provided data that represent all the individuals who are involved in the admission process. This seems unlikely, because both the cover letter and the survey instrument contained reminders that the survey pertained only to those involved in the "final selection process." It seems more likely that admission committee practices are not standardized with regard to committee size and structure.

In summary, we found that there is a wide variety among medical school admission committees in terms of both the size and character of their membership. The representation of women and people with URM status on medical school admission committees has improved but may not be optimal. Committee members with URM status are less likely than their counterparts to be physicians. Further study is required to elucidate what effect committee composition, especially in regard to URM membership, has on the outcomes of the admission process.