Women Are Underrepresented In Exercise Science, And That Is A Problem

Last week, we made the questionable mental health decision of absent-mindedly scrolling Twitter. What started as a generally bad life decision became a specifically infuriating one when we saw an article highlighting the benefits of training in a depleted state of low glycogen availability.

The problem? The article was posted in a way that was marketed toward female athletes. Zero female athletes were included in the cited studies. And intentionally depleted training approaches that may have some benefit in male athletes in moderation could have negative long-term consequences for female athletes.

Understanding and reversing the lack of female athlete representation in exercise science studies isn’t just a question of inclusiveness. It could be the difference between good science and actively harmful misinformation.

The publication deleted the post after readers brought the problem to their attention. There was no ill intent. But it underscores a broader point that applies to all athletes, independent of gender. Understanding and reversing the lack of female athlete representation in exercise science studies isn’t just a question of inclusiveness. It could be the difference between good science and actively harmful misinformation.

The topic of depletion training is a microcosm of the representation problem.

Nearly every study on potential benefits of low-glycogen training includes only male participants, with some of the review articles on underlying physiological mechanisms including no discussion of gender at all. Meanwhile, as outlined by Dr. Stacy Sims in the book Roar and her important work, many of the principles that could make fasted training beneficial in moderation for some male athletes may work in the opposite direction for most female athletes. For women, low-energy states could cause a cascade of hormonal responses, from increased cortisol to sex-hormone perturbations that contribute to amennorhea, along with increased risk of bone-stress injuries and reduced performance.

A 2010 study in the Journal of Science and Medicine in Sport is one of the only articles looking at gender differences directly. In that study, eight female and six male participants underwent four weeks of biking training, with random assignment to groups that trained fasted (FAST) and a group that trained after a meal (FED). In terms of markers of adaptation, “men responded better to FAST and women responded better to FED.” And based on the underlying hormonal consequences of within-day energy deficits outlined in this 2017 study in the Scandinavian Journal of Medicine and Science in Sports, the problems for women conducting fasted training would likely be magnified long-term. It’s possible that a man and a woman with the same background and genetic profile in almost every way could undertake the same training approach, controlling for every variable except gender, and the man would become a world champion while the woman suffers from a major health crisis.

It’s possible that a man and a woman with the same background and genetic profile in almost every way could undertake the same training approach, controlling for every variable except gender, and the man would become a world champion while the woman suffers from a major health crisis.

Or not. There’s a possibility that the negative consequences theorized by some researchers would not be broadly applicable to all female athletes. We just don’t know because there is not enough data. And there is not enough data because women are underrepresented in research.

Research Discrepancy

A 2014 article in the European Journal of Sport Science reviewed 1382 studies in three major exercise science journals from 2011 to 2013 and found that 39% of study participants were women. In 2016, Dr. Bethany Brookshire looked at 188 studies published over the first five months of 2015 in two major journals. She next divided the studies into six categories: metabolism, non-metabolic diseases, basic physiology, social studies, sports injury and performance. The findings generally aligned with the previous study, except for one major offset: performance studies—the type that are used heavily to influence training theory and writing.

There were 30 performance studies. One of those studies was an outlier with 90,000 participants, looking at pacing variation in marathons based on gender. Removing that study left 29 of the types of studies often cited in articles like those we write—X participants doing Y intervention for Z time. What percent of the participants do you think were women?

Lower.

Lower still.

3 percent.

Representation in studies is getting greater attention. Journalist Christine Yu writes wonderful articles on the topic (like this one from 2018 in Outside) and is writing a book on the subject (sign up for updates here). A 2019 journal article in the Journal of Sports Physiology and Performance highlights some potential solutions. Discussion of the problem is a key step, but there is still a long way to go.

In the outlines for our podcast, when it comes to training topics, it seems that there is a disclaimer nearly every week that says, “Ugh! Studies with no female participants!”

For example, in November 2020, we wrote an article on the performance studies for antidepressant medications. Zero female participants were included across all of the major studies. We see the same issues over and over again, even as progress is made. In the outlines for our podcast, when it comes to training topics, it seems that there is a disclaimer nearly every week that says, “Ugh! Studies with no female participants!” (There have been some notably great studies in the last year on training and performance related to the menstrual cycle in particular, and many researchers are taking important steps in all of their work to ensure representation based on gender and race.)

There are some valid reasons for studies that focus on male athletes, as outlined in this 2019 article in Sports Medicine. Hormonal variation at different phases of the menstrual cycle could impact results depending on how studies are designed, especially for studies that lack substantial funding and/or time. And in informal discussions, some researchers say it can be more difficult to recruit female participants. There is probably no ill intent in most circumstances. But lack of intent for underrepresentation doesn’t change the possible outcome of underrepresentation: an understanding of exercise physiology and training that risks causing long-term damage for some female athletes.

There are three major conclusions for writers, coaches and researchers. 

First, when reporting on training theory or physiological principles, try to pay attention to gender disparities and talk about them even if there isn’t a satisfying conclusion. The same principle applies in how we talk about training and physiology issues in social groups, like running clubs or online. It’s key to use language that doesn’t assume or imply that the default human is male. Personally, we will try to improve by including relevant discussion of women and transgender athletes in all of our writing and speaking.

Second, coaches must understand potential gender differences and coach each athlete as an individual, particularly related to the menstrual cycle and hormonal fluctuations. As Dr. Sims says, “Women are not small men,” and incorporating that idea into training requires an open understanding that it’s essential to talk about topics that may be uncomfortable for some people. We will try to improve by starting more conversations related to how hormonal and physiological differences may change training approaches over time.

As Dr. Sims says, “Women are not small men,” and incorporating that idea into training requires an open understanding that it’s essential to talk about topics that may be uncomfortable for some people.

Third, studies should strive for gender inclusion. When that is not possible, the rationale should be discussed in the paper and ideally in the abstract. If a study only includes male athletes, even if past studies show a lack of gender disparity on the topic, it should not be assumed in subsequent research without showing it. When possible, studies should discuss possible implications for transgender athletes as well.

Inclusion goes beyond what is morally right, to what is scientifically right. Gender inclusion in exercise science studies is essential for good science, good science reporting and good science-based coaching.

Megan Roche received her medical degree from Stanford University and is a researcher pursuing a PhD in Epidemiology from Stanford. David and Megan partner with runners of all abilities through their coaching service, Some Work, All Play. They host the Some Work, All Play podcast on running (and other things), and they wrote a book called The Happy Runner.