Two recent studies published in JAMA bring more data to the debate over how prevalent and meaningful gender disparities are in academic medicine. The first examined how academic faculty rank might differ by physician sex. It noted that the proportion of women at the rank of full professor hadn't increased since 1980, remains below men, and hasn't been affected by increasing numbers of women graduating from medical school. The authors used a comprehensive cross-sectional database created by Doximity that contains information on more than 1 million physicians.
The researchers specifically wanted to look at those physicians who had a appointment as faculty in a US medical school. This comprised just over 91,000 doctors, or about 9% of them. They used hierarchical models to analyze sex differences in full professorship as well as combined associate/full professorship, after adjusting for age, years since residency, specialty, authored publications, specialty, and school ranking in research.
There were 30,464 women and 60,609 in medical faculty positions. Overall, 11.9% of women were full professors versus 28.6% of men. Other differences existed as well. Women faculty were younger, and much more likely to be in certain specialties like internal medicine or pediatrics. Women had fewer publications on average, 11.6 versus 24.8 for men. About 6.8% of women had an NIH grant versus 10.3% of men. Fewer women had trials registered on ClinicalTrials.gov, too, 6.4% versus 8.8% for men.
However, even after adjusting for these factors, women were significantly less likely to achieve full professorship than men. The absolute adjusted difference in proportion was -3.8%.
In one sub-analysis, the researchers looked at cohorts of faculty who had completed residency in 1980, 1990, and 2000. This was to account for differences in years working towards promotion. Differences were seen in every group. Of those who finished residency in 1980, 61% of men had become full professors, versus only 47% of women. More than half of women in faculty rank for more 30 years or more were still Associate (23%) or Assistant (30%) Professors.
It's far too easy to dismiss claims like these as being due to characteristics of the women faculty themselves. Some will claim that women just aren't as ambitious, or that they take more time off, or that they work more part-time, or something else. But even after controlling for those things, disparities still remain. Even when you look at the beginning of their careers, when things should be more equal, differences exist.
The second study published in the same issue looked at application data from two New England biomedical research programs. They were able to collect data on how employers supported their faculty at startup. Once again, they also looked at scientific focus, terminal degree, and years since receiving that degree. About two thirds of the applicants held PhD degrees. There were no differences between men and women with respect to their terminal degrees or their years since getting that degree.
However, there were significant differences in the start-up packages that these applicants received. Men had a median startup support of $889,000 versus $350,000 for women. About 40% of the men had packages over $1 million versus only 12% of women. Men had higher support regardless of degree, and for PhDs, the difference was statistically significant. In basic science research, the difference was $980,000 for men and $585,000 for women. Men also had larger startup packages regardless of the stratum of NIH funding of their institution.
It's not clear what should be done. Some have suggested better negotiating skills for women as a solution, but other research shows that when they do they can be judged more poorly than men who do. Others think this is a fault of mentorship. Many believe that those in charge of the system still tend to be white and male, and harbor unconscious biases that prevent them from making the necessary changes.
Regardless, acknowledging that the system is not functioning properly is a necessary first step. How many more studies are necessary before we stop arguing about whether the problem exists and start to do something about it?