Over 30 Years, Minority Authors Make Little Headway in JAMA, NEJM

A fresh look at the diversity of authors whose work is published in the New England Journal of Medicine and JAMA shows little progress—and some setbacks—over the past few decades when it comes to sex and race/ethnicity.

The report, published recently in the Journal of Racial and Ethnic Health Disparities, tracked shifts in first and senior authorship from 1990 to 2020.

The rates of change are so slow that it will take decades if not centuries for female, Hispanic, and Black researchers to catch up. “Such underrepresentation and stagnation suggest that more work is needed before gender and racial equity is achieved in publishing in high-impact medical journals,” Moustafa Abdalla, PhD (Harvard Medical School, Boston, MA, and University of Oxford, England), and colleagues write.

Speaking with TCTMD, Abdalla said this work relied on a searchable database that was originally created for another purpose: to study the history of medicine. They had compiled nearly half a million articles published in JAMA and NEJM from the 1800s through 2020, with the goal of tracking changes in concepts and terminology.

As interest grew in health inequities during the COVID-19 pandemic, they realized the database could also be used to analyze the authors behind those publications, as well.

I think it’s important to overcome institutional inertia and effect change. One key way is making sure people hear [the data] and interact with it. Moustafa Abdalla

Abdalla said he wasn’t surprised by the low percentages of minority first and senior authors. “What shocked me more, I think, was the rate of change,” he said, specifically “how long it will take for the proportion of authors to match the proportion of individuals in the United States. That was the most concerning metric.”

It’s expected that there will be at least some underrepresentation of minority groups in academic medicine, Abdalla acknowledged. “But the question is, are we improving? And the numbers suggest we’re not.”

Even in the face of these discouraging numbers, Abdalla remains optimistic. Part of the reason why they did this work—and chose not to anonymize the journals—was to start a conversation that might spark change. In fact, Abdalla told TCTMD, he and his colleagues submitted their own study to NEJM and JAMA, though they were turned down.

The idea was to not to point fingers but instead to say “here are two journals, very prominent in the US medical establishment,” he said. The next step is to come up with a plan of action to improve representation among their authorships, Abdalla added. “I think it’s important to overcome institutional inertia and effect change. One key way is making sure people hear [the data] and interact with it.”

Women Authors Peaked in the Early 2000s

Abdalla et al ascertained gender by using a validated dictionary of 400,000 names, which then were checked in sensitivity analyses against the US Social Security Administration’s database of names and genders of newborns.

In JAMA, the proportion of research articles with a female first author rose linearly through the 1990s, hitting 29.6% in 2000. It peaked at 38.1% in 2011, though “the overall rate of change has been close to zero since the 2000s,” Abdalla et al note, adding, “At the current rate of change (0.16% per year since 2000), the fraction of JAMA research articles with a female first author will reach 50% in 74 years.”

Women as senior authors in JAMA followed a similar pattern but peaked at just 31.0% of articles in 2005, and the proportion has “since stagnated—with no increase over the past decade,” even decreasing by 0.31% per year since 2000.

With NEJM, female first authorship increased slowly and peaked at 28.2% in 2002. At its current pace, 0.03% per year, it will be another 725 years until women reach 50%. The fraction of female senior authors peaked at 22.9% in 2016.

Stark Trends for Black, Hispanic Researchers

Racial identity, meanwhile, was gauged by last name using clues from US census data and two statistical approaches. In the “best case scenario” math that might overestimate nonwhite authorship, the proportion of Black first authors in JAMA was highest in 1992, at 8.5%, then began to decrease. It reached a low point of 5.2% in 2008 then improved slightly over the last decade—but still hasn’t reached its prior peak. For senior authors, there were parallel trends, with the lowest proportion, 5.1%, seen in 2008.

“The trends for Hispanic authors in JAMA are similarly stark,” the researchers say, adding, “In fact, in 2018, the fraction of articles with a Hispanic first author was 2.8%, the lowest value since the turn of the century. At its current rate of growth (0.04% per year since 2000), it will take more than 337 years for the proportion of JAMA research articles with a Hispanic first author to match their current proportion in the US population (18.5%).” Here, too, senior authorship followed the same dismal trajectory.

For NEJM, meanwhile, the proportion of Black first authors “remained virtually unchanged since 1990, with a 0.03% annual increase,” the paper notes. Senior authorship also has been stagnant, reaching it’s lowest point, at 4.1%, in 2006 then rising to just 6.9% in 2019. “Hispanic authorship in NEJM has also remained unchanged over the past three decades,” Abdalla and colleagues observe.

These disparities “may both reflect and further contribute to disparities in academic advancement,” the authors conclude. They stress, though, that their study did not ascertain how many female, Black, and/or Hispanic researchers submitted their work for publication, but were rejected.