With women in medicine, medical research starts asking different questions. Guest post by Fatima Aqeel

This is the 25th in this year’s series of posts by PhD students on the job market.

A missed diagnosis and a bigger puzzle

In 2014, cardiac patient Joy O'Neal was sent home from the emergency room after her heart failure symptoms were mistaken for emotional distress. Her case was not unusual: decades of male-focused medical research have contributed to higher misdiagnosis rates among women (Mayo Clinic, 2022; Yale School of Medicine, 2023; British Heart Foundation, 2016). This imbalance shapes the knowledge guiding medicine. Who becomes a doctor and researcher may influence what questions get asked, what trials are run, and which patients benefit. My job market paper asks: When the gender composition of doctors changes, does medical research itself become more inclusive?

A natural experiment in Pakistani medical schools

To answer this, Thea How Choon, Yvette Zhang and I leverage a medical school reform that changed gender ratios unevenly across Pakistani specialties and districts. Until the early 1990s, women were capped at 20 percent of medical school seats. After female students challenged this cap in 1987, arguing they were denied admission despite higher scores, the court struck it down in 1992 and mandated gender-neutral admissions (LUMS University, 2024). Women's share in medical schools jumped 10 percentage points in one year and more than doubled within a decade.

This increase varied across specialties and districts. Specialty-district pairs with higher pre-reform female shares experienced smaller increases, while those starting lower saw larger increases. This variation let us compare how specialty-districts more affected by the reform differed from those less affected in the research their doctors later produced.

To establish causality, we employ an instrumental-variables difference-in-differences (DDIV) strategy, using pre-reform female share in each specialty-district as an instrument for the change. We account for trends using fixed effects and present evidence supporting parallel trends across specialty-districts.

Figure 1: Female share among enrolled medical students and registering doctors

(a) Enrolled students

(b) Registering doctors

Note: The left graph is reproduced from Aqeel (2024) and plots the mean share of female students enrolled in medical schools. The vertical line is at the year 1992, and data are from administrative data (provincial development reports). The right graph plots the female share of doctors who register to practice. The vertical line is at 1997, and data are from the Pakistan Medical and Dental Commission (PMDC) medical doctor registration website.

Linking medical careers to medical knowledge

We combined three data sources: PubMed records of every medical paper with a Pakistan-based author from 1982 (15 years pre-reform) to 2019 (~17,000 articles), author geographic location data (on ~7,000 unique authors), and doctor registration data showing female shares by specialty-district.

PubMed indexers categorize articles with "Female" or "Male" labels based on relevance, providing an objective measure of research inclusiveness.

We track how inclusiveness changed as more women entered medicine, examining it by disease, topic, and trial participant gender. To quantify the latter two, we created a lexicon distinguishing medical words associated with female health conditions or anatomy from those referring to participant gender in trials or surveys. We then use text-analysis of article abstracts to identify references to trial participant gender, health/anatomy topics, or both.

More women, more inclusive science

A 1-percentage-point rise in the female share of doctors increased female-relevant research (measured by the "Female" tag) by 3 percent of the mean (0.62), and exclusively female-focused research by 9 percent of the mean (0.17). Of papers tagged "Female," 24 percent of the increase reflected including women in clinical trials or surveys, 29 percent came from studies on female anatomy and health conditions, and the remainder combined both. Female-focused topics included osteoporosis in post-menopausal women, histological evaluations of breast masses from mastectomies, and vitamin A supplements for post-partum mothers.

Topics that matter most to women in Pakistan

These shifts occurred in areas central to Pakistani public health. Research on tuberculosis became 11 percentage points more female-inclusive, diarrhea research 2 percentage points more inclusive, and cancer research 1 percentage point more inclusive. TB and diarrhea rank among the top 10 causes of mortality in Pakistan (World Health Organization (2021)), and cancer is predicted to join this list by 2030.

These diseases often also affect women differently. Breast cancer cases in Pakistan have been misdiagnosed due to medical awareness gaps, and women with tuberculosis face severe pregnancy risks (World Health Organization, 2024). Our findings suggest gender-balanced medical workforces help illuminate these differences, improving science for everyone.

Did female researchers drive the change?

The rise in female-focused research didn't come solely from new female doctors. By following authors active before the reform, we show both men and women shifted toward more inclusive topics once more women entered medicine.

Pre-reform female researchers became 0.8 percentage points more likely to produce female-inclusive research, but male authors did too, suggesting diversity influenced collaboration and inquiry across the profession. Focusing on pre-reform authors whose selection is fixed demonstrates the change in female shares had a treatment effect.

New collaborations and mentoring networks

The reform changed both what people studied and who worked with whom. After the influx of women, female researchers were more likely than men to collaborate with other women and mentor junior female co-authors. These networks likely sustained female participation in academic medicine and amplified inclusive approaches to science.

These patterns appeared even in non-indexed, local journals with less formal peer review, suggesting inclusiveness permeated the broader scientific ecosystem beyond elite research.

Figure 3: Event Studies

A. "Female"-tagged research

B. Collaborations with female authors

Note: The graphs plot event studies for the likelihood of ”Female” tags in articles, and the likelihood of female-female co-authorship relative to female-male co-authorship.

Why this matters beyond Pakistan

Although our setting is Pakistan, the implications extend globally. Women's entry into medicine has accelerated worldwide. For example, it has increased from 6 percent of U.S. medical students in 1960 to over 55 percent in 2025 (AAMC 2023; AAMC, 2025). Yet women remain under-represented in clinical trials and female-specific health conditions are neglected.

Our findings suggest increasing women's participation in medicine not only improves educational access but changes the questions researchers ask and makes scientific knowledge more representative of half the population.

Fatima Aqeel is a Ph.D. economist and Senior Economist at the City of Boston’s Planning Department.