Two of the most influential and esteemed medical journals — if not the top two — are the New England Journal of Medicine (NEJM) and the Journal of the American Medical Association (JAMA). JAMA is more widely circulated than any other medical journal in the world. NEJM has the highest impact factor (number used to measure the importance of a journal) of any medical journal (IF 74.7). So, the combined authorship of articles in these two journals is a reasonable yardstick for assessing the diversity of researchers represented in the most influential medical studies.
That’s exactly what a new study published in the Journal of Racial and Ethnic Health Disparities has analyzed. The findings are discouraging in light of all the lip service in the past decade about needing to improve parity and diversity in medical research. Before I go into more details about the study, here are a few key takeaways:
• Women and racial/ethnic minorities aren’t just heavily underrepresented — their representation as lead or senior authors isn’t increasing in any meaningful way in either NEJM or JAMA.
• It will take centuries for the proportion of Black and Hispanic lead and senior authors to match the population of Blacks and Hispanics in the U.S.
• As journalists, we can’t change who study authors are but we can control who we contact for outside comments. We can and should also make a point to seek out women, gender minorities and Black and Hispanic researchers and clinicians.
• We can also pay attention to the authors of the studies we review. And when presented with two equally impactful studies to cover, we can opt for the one with more diversity among the authors.
A deeper dive into the research
The study authors, led by Moustafa Abdalla, DPhil, currently at Oxford’s Radcliffe Department of Medicine, analyzed the gender and racial/ethnic demographics of authors of every article published in NEJM and JAMA from 1990 through May 2020. (Note the JAMA articles are only in JAMA, not in the other JAMA publications such as JAMA Network Open or JAMA Pediatrics.) They included articles with a digital object identifier (DOI), meaning their data set would consist of actual studies as well as editorials, commentaries and essays.
The authors calculated the proportions of males/females and different races/ethnicities who appeared anywhere in the author list and junior and senior authorship positions. Typically, the first listed author of a medical study is the lead author, who usually does most of the research. The last listed author is usually the senior author, often (but not always) a senior researcher who has typically has less of a role but may have overseen the project or maybe a department chair or mentor of the lead author.
The study findings, which will surprise precisely no one who regularly keeps up with medical research, reveal that women are heavily underrepresented across the board. Though the proportion of female lead authors in JAMA increased during the 1990s to 30% in 2000, it’s stalled since then, peaking at just 38% in 2011. Senior female authors had a similar trend and peaked at only 31% in 2005 and hasn’t increased at all in the past decade. (The authors noted it’s declined slightly since 2000.)
NEJM’s track record is even worse. Though female lead authors also increased in the 1990s, it peaked at just 28% — way back in 2002. And since then? Nada. “The rate of increase is so slow that it will take more than a century for both journals to reach gender parity,” said the study authors. And they were being charitable. NEJM’s current rate of increase in female lead author representation is so slow it would take more than 725 years to reach a 50/50 party. (It’s worth noting that the authors acknowledged one limitation in their data set: the inability to include minority gender authors.)
Little to no progress in race/ethnicity
The outlook is just as bleak for Black and Hispanic researchers, whose “appearance as authors has remained stagnant for three decades despite attention to structural inequalities in medical academia,” said the authors.
Relying on names is far less precise for identifying racial and ethnic minorities since last names for white and Black Americans aren’t distinguishable. So, the authors developed an averaging formula to estimate likely breakdowns using the 2000 and 2010 U.S. Census reports.
An overview of some of the findings, assuming the “best-case scenario” for estimating racial/ethnic authorship:
• JAMA’s Black lead authors dropped between 1990 and 2010 before returning to 1990 levels by 2020.
• JAMA’s highest proportion of Black lead authors was 8.5% in 1992. (For reference, the U.S. population is approximately 13.4% Black.)
• JAMA’s proportion of Hispanic authors hasn’t changed since 2000, and its lowest proportion since then, 2.8%, was in 2018. It would take more than three centuries for JAMA’s Hispanic authorship to match that of U.S. Hispanic demographics.
• NEJM’s highest proportion of Black senior authors was 6.9% in 2019.
• At the current rate of increase, it will take more than two centuries for the proportion of NEJM Black authors to match the proportion of Black Americans.
• NEJM’s Hispanic authorship hasn’t changed in three decades — and at its current rate of 0.01% annual increase since 2000 — it’ll take over a millennium — 1,213 years to be exact — for NEJM’s Hispanic first authors to match the U.S. proportion of 18.5% Hispanics.
As Abdalla told STAT News, the low numbers weren’t the surprising part. It was the disturbing slow rate of change and overall stagnation. “It appears that even in my lifetime, we won’t see change,” he said. Editors from both JAMA and NEJM sent statements to STAT News regarding their commitment to diversity and initiatives aimed at including more underrepresented groups, but the inequities grow from far greater systemic issues, including representation in medical schools and graduate programs. Journalists can’t fix that on their own, and we can’t change the proportions of authors in high-impact journals, so what can we do?
We can start by looking at who our outside sources are. Just because the majority of study lead (and corresponding) authors we might cover may be white, we have complete control over the independent researchers we contact to comment on articles. And while the largest factors in what studies we decide to cover should rely on the significance of the findings and their impact on our audiences, it’s impossible to cover every important study. We can therefore strive to explicitly seek out impactful studies whose lead and/or senior authors aren’t white males.