What Caught Our Eye
By: Brian M. Feldman, MD, MSc, FRCPC | June 17, 2016 | What Caught Our Eye
We often hear of the glass ceiling that prevents the rise of women in academic medicine. In 2016, there is still much gender inequality in the world, as the UN clearly pointed out late last year. I’m old enough to remember the iconic “bra-burning” episodes in the 1960s; while much has been achieved, it’s remarkable how far we still have to go.
However, in academic medicine one would have hoped that we had solved the issues of gender inequality; so many new doctors are women. In fact, in some states and provinces more the half of the medical class is female. In North America, our governments are committed to gender equality. Especially in pediatric rheumatology, one would hope that gender equality has been achieved.
For academics, success is judged by productivity – and for us, as academic docs, productivity is often judged by research funding.
So, it was with great interest that I read a recent paper in Academic Medicine looking at how research proposals were scored at the NIH, and whether sex had anything to do with the scoring. As a rationale for their study, the authors wrote,
Extensive research documents women’s disadvantage in review processes… in fields that have historically been dominated by men, such as science. Such evaluation bias arises from gender stereotypes that characterize women without the “agentic” traits (e.g., independence, logic) associated with ability in male-typed fields, and can lead to the implicit assumption that women are less competent than men in those fields. Experiments show that this assumption can cause reviewers to hold women to higher performance standards than men by requiring more proof of their ability to confirm their competence. Such bias in judgment is often unconscious, unintentional, and demonstrated by both male and female evaluators equally.
The study looked specifically at the review process around NIH R01 grants – either new (type 1) or renewal (type 2). The authors were interested in how the new review process (since 2009) had affected the reviews; (they had already found gender inequality in a previous study of the old review process).
In order to examine the NIH review process, the authors used text analysis, a way of mining the data in the review critiques. They did this for new or renewal R01 applications, from the University of Wisconsin-Madison, that were reviewed after the new review process was in place. About ½ of the PIs at UW-Madison agreed to send their Summary Statements for analysis. The submitted critiques were done by 103 study sections in 21 institutes.
The study found that, while there was no sex difference in comments / critiques for new applications, female applicants were judged significantly worse (i.e. had higher scores) on all sections of renewal applications. More interestingly, females received markedly more critiques that contained words with standout adjectives and positive abilities. This suggested to the authors that “reviewers may have held male and female PIs of Type 2 applications to different evaluative standards”, perhaps based on gender stereotypes.
Failure to get renewal funding may be a reason for leaving the academic stream; this study suggests that part of the “glass ceiling” is due to having a harder time during the (supposedly unbiased) NIH review process.
I had hoped that all of this kind of behavior was in the past; apparently not. As a long-time reviewer for several funding agencies, I am going to have to take a hard look at my own “unconscious and unintentional” biases. How about you?