According to a recent study, women are less likely than men to engage in misconduct when publishing biomedical research. The analysis, published in PLoS One, found that women were markedly underrepresented among the authors of retracted papers, particularly in cases involving multiple retractions.

Drawing on an examination of 878 retracted articles from more than 130 high-impact medical journals, the researchers reported that although women account for 41% of authorship in biomedical literature, only 16% of all retracted papers listed a woman as the first author.

In an interview with Univadis Spain, part of the Medscape Professional Network, Paul Sebo, MD, University Institute for Primary Care, University of Geneva, Geneva, Switzerland, and the study’s lead investigator, said the findings showed that “retractions are not neutral: women are clearly underrepresented among authors of retracted publications, especially in senior authorship positions and in cases of multiple retractions or misconduct.”

The study analyzed more than 2500 authors across biomedical disciplines using an artificial intelligence-based tool.

Gender Patterns

As Sebo noted, most previous research has shown that women are underrepresented in authorship and editorial roles, “but relatively few studies have examined retractions specifically or compared the gender representation of retracted authors with reference data from the same medical specialties and time period.”

One key strength of this study is that it links a large, carefully curated dataset of retractions with established publication benchmarks from a previous study.

This approach “shows that women are not only underrepresented in retractions but also relative to their overall presence among authors.”

Context Matters

These findings suggest that retractions, particularly those related to misconduct, “do not simply reflect the baseline gender distribution in medical publications but are embedded in a landscape in which gender inequalities are already present and sometimes more pronounced among retracted authors,” Sebo said.

He added that context is crucial when interpreting retraction data. “Retraction statistics should not be interpreted in isolation from the underlying demographics of the research workforce. If men dominate senior authorship and leadership positions, they may also be more exposed to the risk for retraction, whether due to unintentional errors or misconduct.”

Sebo stated that his interest in this issue emerged in earlier research. “In that study, we observed that women appeared to be underrepresented among the authors of retracted medical articles, particularly in cases related to misconduct. That analysis was limited by a relatively small sample and some methodological constraints, so I wanted to assess the hypothesis more rigorously in a larger, well-defined cohort of retractions from high-impact journals.”

Study Limits

Given the broader gender gap in academic publishing and senior authorship, the researchers suspected that women were underrepresented among the retracted authors. “But the magnitude of the disparity, especially for first and last authors and for serial retractors, remained surprising,” he added.

Among the study’s limitations, Sebo cited missing or ambiguous author names in some publications. As a result, “authors with initials only or ambiguous names were excluded from the gender analysis, which could introduce some selection bias.”

The comprehensive retrieval and classification of retractions are also challenging. Although the researchers used three major bibliographic databases to identify retracted articles and relied on the Retraction Watch database to categorize the reasons for retraction, the classification ultimately depended on the information contained in the retraction notices. These notices are not standardized across journals and vary widely in terms of clarity, detail, and transparency. Because of this heterogeneity, Sebo said, “some degree of uncertainty in the categorization of retractions is inevitable.”

Additionally, the analysis was limited to English-language high-impact journals across the nine medical specialties. “The patterns observed here may not be generalized to journals with lower impact factors, nonmedical fields, or journals published in other languages,” Sebo said. The study could not adjust for factors such as country, research team size, funding, or open access status.

For Sebo, a key conclusion is the close link between equity and integrity in research. He noted that “efforts to strengthen research integrity must be designed with attention to existing inequalities.”

He added that he hopes the study will “encourage the research community to view retractions not only as corrections to the literature but also as windows into how responsibility, power, and opportunity are distributed across genders in science.”

Sebo and the other authors reported no relevant conflicts of interest. 

This story was translated from Univadis Spain.