Materials and Methods

The conduct of this investigation was approved by our 2 institutional review boards (University of New Mexico Human Research Review Committee no. 08-442; American College of Obstetricians and Gynecologists institutional review board no. IRB00003428). Data were obtained using the Faculty Salary Survey of the AAMC. Since the early 1960s, the survey has been the most comprehensive source of the total compensation attributed to teaching, patient care, and research by faculty in MD-granting accredited allopathic schools in the United States. This survey includes full-time faculty at university-based programs and private programs with university affiliations. Data for total salaries of the board-certified subspecialties (gynecologic oncology, maternal-fetal medicine, and reproductive endocrinology) were first published in the survey in 2002 for the academic year 2000-01.

During the fall of every academic year (July 1 through June 30), the AAMC distributes the same web-based survey to all accredited US medical schools. The principal business officer at each medical school, or his or her designees reports each faculty member’s pretax total compensation, along with the faculty member’s degree, academic rank, and primary department/specialty affiliation. Real-time edit checks as well as post-upload edit checks by the AAMC flag those salaries that fall outside the expected range for a particular combination of degree, department/specialty, and rank. These medical schools are requested to verify or correct the questionable compensation data. This salary information, gathered each October, is then summarized in tables in the annual Report on Medical School Faculty Salaries, which is distributed in January.

Total compensation for every faculty member included the fixed or contractual salary, variable (supplemental and bonus or incentive pay) components, and outside earnings, for consistency in making comparisons. Fixed or contractual salary was compensation (exclusive of fringe benefits) that was set at the beginning of the fiscal year and contractually obligated assuming standard satisfactory performance. The medical practice supplement was income that is not fixed at the beginning of the fiscal year but is directly tied to the amount of productivity during the year. Data on bonus or incentive pay, earned as a result of achieving specific performance goals, were collected on all eligible faculty. Outside earnings were self-reported and unregulated outside income related to the health professions (eg, patient services income, royalties, and consulting fees). Also reported were faculty contributions to salary reduction programs for retirement (eg, 403B/Tax Sheltered Annuity plans). Employer contributions to retirement plans and additional benefits provided by the institution, such as tuition benefits and employer-paid health, life, and disability insurance, were excluded. Salaries of new faculty whose appointments began beyond the beginning of the academic year were converted to an annual equivalent.

This investigation examined published data on rank (assistant professor, associate professor, professor, chair) and discipline (general obstetrics-gynecology, gynecologic oncology, maternal-fetal medicine, reproductive endocrinology). The AAMC Faculty Salary Survey has not yet distinguished the evolving specialty urogynecology from general obstetrics-gynecology because of the lack of board certification. We included only full-time faculty at university-based or university-affiliated departments who were MDs, MD equivalents (eg, DOs), and MD/PhDs. PhD faculty in obstetrics-gynecology were excluded because they constituted a small percentage of the total faculty, and their salaries were more consistent with those in basic science departments. As a result of these adjustments, not all faculty were included (usually 70% from all schools), although we agree with Studer-Ellis et al who critically reviewed the AAMC Faculty Salary Survey and found the data to represent the faculty overall.

Salaries for each faculty rank and department/specialty were reported at the median or 50th percentile, which corresponded to the salary below which half of all salaries appeared. Median salaries were used rather than mean, since the median is less influenced by unusually high or low compensation outliers. We also assessed the percent change in median salaries from year to year for faculty in general obstetrics-gynecology with general practitioners in other core clinical departments through which medical students rotated during their initial clinical clerkships.

We examined compensation levels that were unadjusted and adjusted for inflation. The Consumer Price Index (CPI), an economic indicator determined monthly by government economists at the Bureau of Labor Statistics of the US Department of Labor, was used as a means of adjusting dollar values. While sometimes referred to as an index of inflation or cost of living, the CPI is actually a measure of the average change over time in the price paid by urban households for a set of consumer goods and services (real purchasing power). We used the CPI-U (CPI-urban), since it reflected the spending patterns of all-urban consumers, which represent about 87% of the US population and most US medical school locations.

Data from the AAMC Report on Medical School Faculty Salaries were entered into a spreadsheet and coded by academic year, designated AAMC geographic region (northeastern, southern, midwestern, western), whether the school was public or private, and faculty rank and department division. Data were aggregated in such a way that it was possible only to cross-classify by public/private and rank or by geographic region and rank. Data included the number of schools, mean, median, and quartiles in each aggregate. SDs were not available in the data but were estimated using the consistent estimator for normal distributions defined by (interquartile range)/1.35. Multiple comparisons employed Bonferroni adjustments. Student t test was used to establish differences between faculty salaries in 2000-01 and 2008-09. Statistical significance was set at the α = 0.05 level.