Search

Where are all the Female Participants in Strength, Hypertrophy, and Supplement Research?

An oft-cited 2014 study found that females account for approximately 40% of the total subject pool in exercise science research. However, female lifters seem considerably more under-represented in many of the areas of research that lifters care the most about. So, we decided to investigate the matter for ourselves.

Note: This article was the MASS Research Review cover story for June 2022. If you want more content like this, subscribe to MASS.

For as long as I can remember, inter-individual differences in training responses have been one of my biggest research interests. It doesn’t take enormous powers of observation to see that two people can undergo the same type of training, but attain wildly different results. I’ve always been interested in learning more about this for a few reasons. First, I simply want to understand the phenomenon better: how much variability exists (1)? In my experience, people tend to underestimate how much training responses differ between individuals. Second, I want to learn more about the factors that are predictive of responsiveness to training. If we know what factors promote above-average training responses, we may eventually be able to use that knowledge to improve training results for everyone.

This interest in inter-individual differences naturally led to an interest in sex differences. Sex is a unique variable, in that it’s bimodal (most observable traits are more normally distributed), and it’s either associated with, or causally linked to, a host of other traits that differ between individuals, which may be predictive of training responsiveness (hormone levels, body size, muscle fiber types, etc.). So, learning more about sex differences seemed to be worthwhile, in order to better understand inter-individual differences more broadly. This interest led me to study sex differences in fatigability for my thesis research (2).

However, once you start trying to learn more about sex differences in domains related to resistance training, one thing becomes immediately apparent: most of the research in our field is solely conducted with male research subjects. When the topic of sex differences in research participation comes up, my go-to citation has always been a 2014 study by Costello and colleagues, showing that about 61% of the research subjects in our field are male, and 39% are female (3; the title of this article is an intentional homage to that study). However, those figures never quite sat right with me. In the research I was reading, it seemed like female research subjects made up considerably less than ~40% of the total research subjects.

To understand why there might be a disconnect, it’s worth understanding how Costello and colleagues came up with their estimate. They monitored three of the most prestigious journals in our field – Medicine and Science in Sports & Exercise (MSSE), the British Journal of Sports Medicine (BJSM), and the American Journal of Sports Medicine (AJSM) – for three years (2011-2013), noting the total number of male and female subjects in each article contained within each issue. Across that three-year span, 1,328 articles were published, containing nearly 6.1 million subjects, including about 2.4 million female subjects and about 3.7 million male subjects. This was a truly impressive undertaking, but it has one notable drawback for our purposes: a minority of the research published in those journals is directly related to the subjects we discuss most often in MASS and Stronger by Science, and a small minority of the research we discuss is published in those three journals. The most prestigious journals in our field will publish really cool research related to maximizing strength, hypertrophy, and resistance training performance from time to time, but a larger chunk of their publications focus on general health, injury prevention or rehabilitation, and aerobic fitness.

A couple years ago, I happened across an article in ScienceNews by Bethany Brookshire – a journalist with a PhD in physiology and pharmacology – who had similar concerns. She wasn’t specifically interested in strength and hypertrophy research, but she wanted to know if the proportion of male versus female research subjects differed by study type. She kept tabs on MSSE and the AJSM for the first five months of 2015, sorting studies into six categories: disease, basic physiology, metabolism and diet/obesity, injury, social, and performance. She found that female participants made up between 40-60% of the research participants in all six categories, but the “performance” category had a major caveat. One study on marathon pacing contained more than 90,000 subjects, accounting for the majority of the total research subjects in the “performance” category. When that single study was excluded, only 3% of the subjects in “performance” studies were female (Figure 1).

Graphic by Kat Whitfield

However, Brookshire’s analysis also has a couple of drawbacks for our purposes. First, and most notably, it was still centered around two journals that don’t prioritize strength and hypertrophy research. If I had to wager a guess, I doubt that most of the “performance” studies were focused on resistance training performance. Second, her analysis of performance studies had a pretty small sample. In the first five months of 2015, just 30 studies related to performance were published in the two journals she was monitoring. After excluding a single enormous study, she found that only 3% of the subjects in the other 29 studies were female. While that’s certainly concerning, it may not be representative – 29 studies published in two journals over five months in 2015 may not be a reliable reflection of all performance-related research in the field. It’s entirely possible that MSSE and AJSM just had a random run of very male-dominated performance studies during that five-month span. However, happening across Brookshire’s article reassured me that my suspicions probably weren’t misplaced: female subjects probably don’t account for 39% of the total subjects in the exercise research that I (and most of you, I assume) care the most about. After reading Brookshire’s article, I determined that I’d eventually do my own analysis. That’s what you’re reading now.

Finally, a 2021 paper by Cowley and colleagues (4) gave me another issue to ponder: what if the sex disparity in exercise research participation is actually getting worse over time? Cowley and colleagues updated and expanded on Costello et al’s analysis. They analyzed the research published in six journals (they looked at MSSE, BJSM, and AJSM like Costello, and added the European Journal of Sport Science, the Journal of Sports Science & Medicine, and the Journal of Physiology) over a seven-year span – 2014-2020. Approximately 5,300 studies with about 12.5 million participants were published in those six journals over the analyzed time span, including about 8.25 million (66%) male subjects and 4.25 million (34%) female subjects.

This gave me pause, because there’s a general belief that sex disparities in research participation are shrinking over time. In other words, there’s a general assumption that very early exercise science research used predominately male subjects, but that female subjects accounted for nearly 40% of all research subjects by 2011-2013 (the time period of Costello’s analysis), meaning that the sex disparity in research participation was decreasing over time. The natural assumption is that this disparity would be expected to decrease further until it became a non-issue. However, Cowley and colleagues found that female research subjects accounted for a smaller proportion of the total research subjects in the 2014-2020 time period (34%) than Costello and colleagues observed in the 2011-2013 time period (39%). These aren’t purely apples-to-apples comparisons, since Cowley and colleagues investigated more journals than Costello and colleagues did, but it at least suggests that sex disparities in exercise research participation aren’t continuing to shrink over time; in fact, they may be getting larger.

Purpose and Strategy

So, with that preamble out of the way, I decided to do my own analysis of sex disparities in research participation in the areas of research that MASS and Stronger by Science readers care the most about. Here’s what I wanted to accomplish:

  1. I wanted to be able to cast a wide net. Our monthly journal sweep combs through >140 journals (thanks, Kedric and Colby), so restricting my analysis to 3-6 journals wasn’t going to cut it.
  2. I wanted to be able to analyze trends over time, so I needed to have a way to see how sex disparities in research participation had shifted (or not shifted) over a period of decades, rather than a 3-7 year period.
  3. I wanted to be able to restrict my analysis to the sorts of research MASS and Stronger by Science readers care the most about. I’ll admit that this isn’t a completely objective criterion, but after putting out content, answering questions, and monitoring chatter in the fitness industry for over a decade, I think I have a pretty decent grasp on the research topics y’all care the most about.
  4. I wanted to be able to see whether the results of studies with female subjects systematically differed from the broader literature. I’ll discuss my reasoning for this below.
  5. I needed to be mindful of time- and labor-intensiveness. I’m writing this article on a deadline, so I needed to select a strategy that would allow me to do a thorough, representative analysis in about two weeks (not two months or two years). Furthermore, as a non-academic, I no longer have institutional journal access (5). I’m fortunate enough to have people who will send me papers when I request them, but I don’t want to push my luck. Requesting a few dozen papers is asking for a favor. Requesting a few thousand papers is asking someone to start a part-time job.

To expand on my fourth criterion a bit, seeing whether the results of studies on female subjects differ from the broader literature helps us answer a couple of important questions.

First, it helps us generate informed assumptions about the generalizability of research conducted on single-sex samples. In areas of research where most findings come from studies on male subjects, it would be nice to know whether or not we can assume that those findings will generalize to female lifters. And, conversely, it would be nice to know whether research findings on female-only samples are likely to generalize to male lifters. Research interpretation always involves making assumptions about generalizability – better-informed assumptions help make more research more useful to more people.

Second, it can help inform research recruitment strategies. If we see that studies on female subjects commonly reach different results than male-dominated studies, that would imply that single-sex cohorts are probably preferable most of the time. In that situation, you should assume that a new intervention will produce different results in male and female lifters, meaning that early studies in the area should use male-only and female-only samples to generate effect estimates for each sex independently. However, if we see that studies on female subjects typically have similar results to those observed in male-dominanted studies, that would imply that mixed-sex cohorts are probably preferable for both practical reasons and logistical reasons.  Why invest double the time and double the energy to generate male- and female-specific effect estimates, if those effect estimates are likely to be similar? You can just used mixed-sex cohorts to generate a generalized effect estimate that should apply across the board. And why struggle trying to recruit 30 male subjects or 30 female subjects for a study? You’d have an easier time just recruiting 30 humans of any sex.

So, with all of that in mind, I decided to use recent systematic reviews and meta-analyses to do a lot of the heavy lifting for me. This approach fulfills the five criteria listed above:

  1. Systematic reviews and meta-analyses start with a comprehensive literature search, pulling in research from all of the indexed journals in our field, rather than restricting the search to a handful of journals.
  2. Systematic reviews and meta-analyses generally aren’t time-limited. They pull in research going back decades, allowing me to analyze trends in sex disparities over time.
  3. There are now systematic reviews and meta-analyses covering damn near every topic that MASS and Stronger by Science readers care about, which we catalog here. So, mining systematic reviews and meta-analyses was a convenient method of pulling in all of the research on highly relevant topics, while excluding research on less relevant topics.
  4. By comparing effect estimates from female-only research to pooled effect estimates in the included meta-analyses, I’d be able to see whether research on female subjects typically has meaningfully different results than research on male or mixed-sex cohorts.
  5. This strategy saves an enormous amount of time and labor, without sacrificing the scope of the project. Systematic reviews and meta-analyses generally contain a table listing the characteristics of the studies included, including the number of subjects and sex of the subjects in the study. Thus, a single meta-analysis can provide all of the relevant information about 20 studies, rather than needing to pull data from all 20 studies one-by-one. Systematic reviews and meta-analyses are also more likely to be open-access than original research. As a result, I only needed a kind soul (named Eric Helms) to hook me up with 12 papers I didn’t have access to, rather than (likely) 400+ papers.

After identifying an initial pool of 45 systematic reviews and meta-analyses, I whittled the list down slightly based on two factors. First, if two reviews covered very similar topics, I’d select the review that included the most studies. I wanted to maximize the scope of topics included in this analysis, while minimizing the overlap between reviews. Second, I’d exclude a systematic review or meta-analysis if it didn’t include a table listing the characteristics of the studies included. Fortunately, these two exclusion criteria only whittled my initial pool of systematic reviews and meta-analyses down by 6 papers, leaving me with 39 systematic reviews and meta-analyses for my final analysis.

After finalizing my list of systematic reviews and meta-analyses, I went through each one to extract the following information from the studies included:

  1. The title and author of the study.
  2. The number of male and female subjects in the study, along with the total number of participants. If the sex of the participants in a study wasn’t reported, or if a study was reported as mixed-sex without specific counts of male and female subjects, the subjects were assumed to be 50% male and 50% female.
  3. Whether the study used a male-only, female-only, or mixed-sex cohort.
  4. The publication year of the study.

Furthermore, I isolated all of the forest plots from the meta-analyses, and highlighted the effect estimates from the female-only studies in each forest plot. From the forest plots, I extracted the following information:

  1. The pooled effect estimate.
  2. The standard error of the pooled effect estimate.
  3. The effect estimate of female-only studies.
  4. Whether the 95% confidence interval of each female-only effect estimate overlapped with the 95% confidence interval for the corresponding pooled effect estimate.

Finally, I’d just like to make a note about the assumption that subjects were 50% male and 50% female when meta-analyses didn’t report sex of the subjects in a particular study, or when studies were reported to be mixed-sex without a precise delineation of the number of male and female subjects. Technically speaking, this is a cut corner, but I don’t think it materially impacts the value of this analysis for a few reasons: 1) over three-quarters of the studies included in these systematic reviews and meta-analyses were single-sex studies, 2) sex wasn’t reported in a very small minority of studies, and 3) the precise numbers of male and female subjects were reported for most of the mixed-sex studies. So, if the subjects in those studies were 70/30 or 30/70 male/female instead of 50/50, that would only shift the estimated proportions of male and female subjects by 1-2%, which is pretty immaterial for the purpose of interpreting this analysis. As I’ll cover in the next section, about 25% of the subjects included in these studies were female – if the “true” figure is actually 23% or 27%, I don’t think that’s an error that actually matters. Any figure within that range would lend itself to the same set of conclusions.

Results

As previously mentioned, 39 systematic reviews and meta-analyses were used for this analysis, covering topics ranging from training volume to rest intervals to ketone supplementation. They’re listed in Table 1.

Graphic by Kat Whitfield

These systematic reviews and meta-analyses covered 628 unique studies, with an average of 16.8 studies per systematic review or meta-analysis (range: 6-49 studies). Just 28 studies (4.5%) were included in multiple meta-analyses, suggesting that I did a pretty good job of selecting topics that would cover a broad range of topics while minimizing the overlap between topics.

Of these 628 unique studies, 408 (65.0%) had all-male samples, 133 (21.2%) had mixed-sex samples, 73 (11.6%) had all-female samples, and 14 (2.2%) did not specify the sex of the subjects (Figure 2).

Graphic by Kat Whitfield

The box and whisker plot in Figure 3 shows the proportion of male-only, mixed-sex, and female-only studies included in each systematic review and meta-analysis.

Graphic by Kat Whitfield

The studies contained within these systematic reviews and meta-analyses had 16,683 total subjects, including 12,501 males (75.01%) and 4,182 females (24.99%). Only one meta-analysis included studies with more total female subjects than male subjects (34), while three meta-analyses didn’t include any studies with female subjects (8, 13, 27).

From the 1990s onward, the proportion of studies with male-only cohorts has actually increased, while the proportion of studies with female-only cohorts has slightly decreased. The proportion of studies with mixed-sex cohorts has basically remained flat since the 1990s (Figure 4). 

Graphic by Kat Whitfield

Finally, when analyzing the forest plots contained within these meta-analyses, I didn’t find evidence that the female-only studies systematically differed from the broader literature. There were 67 total forest plots that contained at least one effect estimate from a female-only study, and there were 185 effect estimates from female-only studies contained within these forest plots. The 95% confidence interval of female-only effect estimates overlapped with the 95% confidence interval for the corresponding pooled effect estimate … 94.6% of the time. The confidence intervals overlapped 175 times and didn’t overlap 10 times. Furthermore, on all 67 discrete forest plots, the confidence intervals from female-only studies overlapped with the confidence interval of the pooled effect estimate a majority of the time. Finally, across all of these meta-analyses, the average female effect estimates differed from pooled effect estimates by an average of -0.042 standard errors. In other words, if the pooled effect estimate from a meta-analysis was d = 0.5 (95% CI = 0.1-0.9), the mean effect estimates from female-only studies would be d = 0.49, on average – a completely inconsequential difference.

If the last paragraph sounded like it was written in a foreign language, Figure 5 illustrates what I’m talking about.

Graphic by Kat Whitfield

In this forest plot from García-Valverde’s meta-analysis (18), the pooled effect estimate is 0.86, with a 95% confidence interval from 0.51-1.21. The studies by Ayers (45) and Slovak (46) are female-only studies. Both of the confidence intervals from the Ayers study (95% CIs from 0.13-1.32 and 0.02-1.05) overlap with the confidence interval of the pooled effect estimate. However, the confidence interval from the Slovak study does not overlap with the confidence interval of the pooled effect estimate. Of note, the Slovak effect estimate in Figure 5 was one of the biggest outliers of any female-only effect estimate in any of these meta-analyses, and it’s not even the biggest outlier in that particular forest plot. The effect estimate from the Moore study (47) is quite literally off the chart.

Just to solidify this point, Figure 6 shows the “worst” forest plot of the bunch, from Heidel et al’s meta-analysis (24). There are five female-only effect estimates; three of them have confidence intervals that overlap with the pooled effect estimate (Nautilus leg press, Nautilus chest press, and Soloflex chest press), while two don’t overlap (Nautilus shoulder press and Soloflex shoulder press).

Graphic by Kat Whitfield

As you can see, all five effect estimates come from a single study (48). While the confidence intervals of the lowest and highest effect estimates from this study don’t overlap with the confidence interval of the pooled effect estimate, the mean effect of all measures in the Boyer study was –0.88, which is very close to the pooled effect estimate (-0.78).

In short, it appears that studies on female lifters produce results that are similar to the broader literature across every topic examined in these meta-analyses.

Interpretation

To summarize, this analysis found that female subjects are heavily under-represented in areas of research that are relevant to lifters. Furthermore, it found that female under-representation may actually be getting worse in recent decades. Finally, it found that studies on female lifters typically have results that are similar to those observed in mixed-sex and male-only cohorts.

On its face, mere under-representation doesn’t necessarily imply that there’s a bias against studying female lifters. After all, you could argue that males are more likely to participate in resistance training than females – if you studied, say, 10% of all male lifters and 10% of all female lifters, you’d still be studying more males than females. Therefore, you should expect there to be more male subjects than female subjects in areas of research that are relevant to lifters. Furthermore, you could argue that studies exclude female subjects for valid logistical reasons. For example, you may be concerned that performance fluctuations throughout the menstrual cycle would either increase the logistical complexity of a research project (i.e., it might require you to ensure female participants are always assessed during the same phase of their cycles – that’s not a concern with male subjects) or add noise to your results (if you didn’t account for menstrual cycle phase during assessments). Or, you might be concerned that results of a particular intervention would differ between normally menstruating women and women using hormonal contraceptives (again, not a concern with male subjects). Finally, you might be concerned that a particular intervention would affect male and female subjects differently, such that using a mixed-sex sample would simply result in noisier data. However, I think I can reasonably counter all of these concerns.

For starters, males are more likely to participate in resistance training than females. A recent review by Nuzzo found that women are between 9.4-44% less likely than men to meet public health recommendations related to participation in muscle-strengthening activities (49). However, even if we were to assume that research participation in resistance training-related research should match trends of general participation in resistance training, female lifters would still be under-represented in the scientific literature as it currently stands. If research participation scaled with general resistance training participation, you should expect female subjects to comprise 36-47.5% of the total pool of research subjects. The current proportion (25%) falls well below the bottom end of that range. Even if I cherry-picked the 20 systematic reviews and meta-analyses with the highest proportion of female subjects (out of my initial pool of 39), the proportion of female subjects in the studies included in those reviews is still just 33%, which would still fall below the bottom end of that range. In short, differing levels of participation in resistance training don’t explain the degree to which female subjects are under-represented in resistance training research.

Next, let’s address logistical concerns. Some researchers may opt to study male-only cohorts due to fears that including female subjects will add complexity to a research project or add noise to your results (largely relating to concerns about the menstrual cycle and hormonal contraceptives). Fortunately, while those concerns are certainly reasonable in a vacuum, recent research should significantly alleviate those concerns in most contexts. Meta-analyses by McNulty et al (50) and Elliot-Sale et al (51) have found that performance fluctuations throughout the menstrual cycle are typically trivial, and that hormonal contraceptives have little impact on most measures of performance. Furthermore, as I discussed in a recent article, hormonal contraceptives seem to have little impact on longitudinal muscle growth and strength gains following resistance training. These findings should mitigate most of the logistical concerns people raise when discussing the prospect of studying female lifters. Assessing female lifters at different points in the menstrual cycle or including female lifters who both use and don’t use hormonal contraceptives is unlikely to meaningfully alter your results or add unmanageable amounts of noise to your data in most contexts.

Finally, let’s address the concern that male and female subjects are likely to attain different results following some resistance training or supplementation intervention. This is a perfectly reasonable concern since there are fairly large visible differences (there are obvious anthropometric differences between the sexes) and invisible differences (52) between males and females. However, in most research contexts, we typically don’t care too much about baseline differences between subjects. Rather, we care if subjects experience different responses to a particular intervention. If subjects do experience meaningfully different responses to a particular intervention, that increases the variance in your data, and makes it harder to reliably detect the effect of the intervention. The present analysis found that studies on female lifters typically attain results that are in line with the broader literature (across numerous different bodies of research). Thus, while there are certainly baseline differences between males and females, it seems that male and female subjects have very similar responses to most interventions that would be relevant to MASS and Stronger by Science readers. Furthermore, looking beyond this analysis, we know that male and female lifters typically experience comparable muscle growth and strength gains in response to identical training interventions (53), and that protein needs are very similar between the sexes when scaled to lean body mass (54, 55). Thus, using mixed-sex samples should not be problematic in most contexts. Using mixed-sex samples should make subject recruitment easier, without making it more difficult to reliably detect treatment effects in most contexts.

With all of that said, there are still circumstances when it would make sense to use single-sex cohorts. For starters, there are some research topics that are specifically relevant to a single sex (research related to the menstrual cycle, pregnancy, hormonal contraceptives, menopause, prostate cancer, etc.). Furthermore, there are areas of research with known, notable sex differences where single-sex cohorts may offer advantages – research related to concussion and non-contact ACL injury risk immediately come to mind, as well as research related to osteoporosis and iron deficiency/supplementation. There can also be situations where you only have access to a single-sex cohort. For example, if you’re given the unique chance to study an elite male rugby team, turning down the opportunity because you don’t also have access to study an elite female rugby team doesn’t make a ton of sense. There are also research topics where cultural norms may dictate that a single-sex cohort would be preferable. For example, if you want to study pec hypertrophy, and all of the trained ultrasound technicians in a particular lab are male, studying a male-only cohort may be preferable. There are also research questions for which a mixed-sex cohort would increase the complexity of your statistical approach at best, or massively increase the noise in your results at worst. For example, if you were interested in the correlation between testosterone levels and some particular training outcome, employing a mixed-sex subject pool with a bimodal distribution of testosterone levels would present you with significant analytical challenges. Finally, there may be good a priori reasons to assume that sex would have a notable impact in some brand new area of research – in that context, it might make sense to conduct a couple of male-only and female-only studies first to validate or disprove your assumption.

Moving on, I want to briefly address the areas of research where female subjects are the most and least under-represented.

There were three meta-analyses in which female subjects accounted for at least 45% of the total subject pool: a meta-analysis by Schoenfeld and colleagues investigating the impact of eccentric vs. concentric muscle actions on muscle growth (40; 47% female subjects), a meta-analysis by Grønfeldt and colleagues investigating the impact of blood-flow restriction training on strength gains (22; 49.5% female subjects), and a meta-analysis by Murphy and colleagues investigating the impact of energy deficits of strength and lean mass changes (34; 82% female subjects). Furthermore, there were three meta-analyses that exclusively included studies with male-only cohorts (56): a meta-analysis by Cuthbert and colleagues investigating the impact of training frequency on strength gains (13), a meta-analysis by Baz-Valle and colleagues investigating the impact of training volume on muscle growth and strength gains (8), and a meta-analysis by Kassiano and colleagues investigating the impact of exercise variation on muscle growth and strength gains (27).

The three male-only meta-analyses cover research questions that are highly relevant to most lifters: how frequently should I train each muscle, how much training volume do I need to maximize my results, and can I improve my results by training each muscle group with multiple exercises (57)? Conversely, the two meta-analyses with near-parity cover more niche topics: blood-flow restriction training isn’t a staple in most people’s training arsenals, and few lifters do much eccentric-only or concentric-only training. Thus, while this is obviously subjective, the present analysis may still overstate the degree to which female subjects are represented in the research most people would use to make training decisions. In other words, females may be 25% of the total subject pool, but they may comprise 20% of the subject pool in the most practically relevant areas of research, and 30% of the total subject pool in more niche areas of research (58). Furthermore, I’ll note that the only meta-analysis I reviewed specifically investigating weight loss was also the only meta-analysis with more female subjects than male subjects. Take from that what you will.

Takeaways

I’ll start by stating the obvious: female lifters are significantly under-represented in the research that’s most relevant to lifters. Furthermore, female subjects are more under-represented in resistance training research than in general exercise science research, and the problem seems to be growing over time.

Thankfully, that doesn’t necessarily mean that the research in this area is uninformative for female lifters. The finding that studies on female-only cohorts reach results that are similar to those observed in the broader literature cuts both directions. It doesn’t just mean that researchers can confidently include female lifters in their studies without fearing that their male and female subjects will have meaningfully different responses to study interventions. It also means that, in general, research on male-only or mixed-sex cohorts should be expected to generalize to female lifters. As someone who has discussed research in public for a decade, I frequently encounter male lifters who disregard research findings from studies on female cohorts, and female lifters who disregard research findings from studies on male cohorts. I can certainly understand that impulse, but at least in the context of resistance training research, it’s probably unfounded most of the time. Most research findings generalize between the sexes quite well.

Finally, my biggest takeaway from this analysis is that more research should be conducted on mixed-sex cohorts. Most of the time, using a mixed-sex cohort comes with clear benefits (making subject recruitment easier and potentially increasing the generalizability of your findings), and it rarely has obvious downsides. As previously acknowledged, there are certainly situations where it makes sense to study single-sex cohorts, but there’s absolutely no reason why 65% of the studies in this area should use male-only cohorts. When female lifters account for approximately 40% of the general lifting population, there’s no reason why they should only account for 25% of the research subjects in the area. Thankfully, we know this is a solvable problem – as Brookshire found in 2016, most areas of exercise-related research have already achieved something resembling equal research representation for both sexes. It’s time for strength training research to follow suit.

Drawbacks

While I think the approach I took to addressing this problem is sound, it still has its drawbacks. If you’re primarily interested in practical takeaways, feel free to stop reading here. If you’re really hankering for several pages of self-criticism, then read on.

First, this analysis likely underestimates the overall proportion of studies on female lifters, for one simple reason: I only looked at “neutral” bodies of literature where participants can be either male or female. There are several bodies of research where all of the subjects are female: research looking at the impact of the menstrual cycle, hormonal contraceptives, and pregnancy necessarily have female-only cohorts. Conversely, there are a handful of topics that necessarily employ male-only cohorts (most notably, studies on exercise in subjects with prostate cancer). There are more sex-specific topics that require female-only cohorts than male-only cohorts, and the female-specific topics tend to garner more research attention because they affect more people in total (almost all females will menstruate, but most males won’t get prostate cancer). Cowley’s paper found that 20% of studies on female subjects investigate female-specific research questions, whereas only 0.6% of studies on male subjects investigate male-specific research questions (4). However, I don’t necessarily view this as a weakness of the analysis. There’s no reason why there shouldn’t be plenty of studies on contraceptives, and also plenty of studies on training volume with female cohorts. The fact that female-specific bodies of research exist doesn’t imply that female lifters shouldn’t be better represented in “neutral” bodies of literature.

Second, the precise result of this type of analysis will necessarily depend on the systematic reviews and meta-analyses you select as your starting point. If someone wanted to cherry pick reviews to make it appear that female lifters are more under-represented or less under-represented, it wouldn’t be hard to put your thumb on the scale. For example, if I only analyzed the 20 systematic reviews and meta-analyses with the lowest proportions of female subjects, I could have estimated that female subjects compose just 14% of the total subject pool in the area. However, if I only analyzed the 20 systematic reviews and meta-analyses with the highest proportions of female subjects, I could have estimated that female subjects compose nearly 33% of the total subject pool in the area. To be clear, I didn’t do this; I didn’t pre-screen the original batch of 45 reviews I looked into, and the reviews I excluded were either excluded by necessity (due to insufficient reporting about the studies going into the review), or they would have had minimal impact on the analysis due to extensive overlap with reviews that were included. However, this general problem applies to essentially any approach one could take to addressing this basic research question. For example, if you used Costello’s and Cowley’s approach (screening all studies published in a fixed number of journals), you could decide to bin the results from a journal or two with particularly high or low proportions of female research subjects (to be clear, I don’t think that happened in the Costello and Cowley studies). I attempted to defray this risk by casting a really wide net. When you scan the list of systematic reviews and meta-analyses in Table 1, I suspect you won’t be able to think of very many highly active areas of research that are unrepresented in this analysis and that are highly relevant to lifters. That should ensure that I obtained a representative sample of studies.

Third, this approach is blind to areas of research that haven’t yet been systematically reviewed and meta-analyzed. So, it’s theoretically possible that less active areas of research or cutting-edge areas of research – bodies of literature with too few studies to warrant a systematic review or meta-analysis – have a proportionally greater number of male or female subjects than more established areas of research. If anything, I suspect this source of bias would result in a net overestimate of the proportion of female research subjects using the approach I took for this article. Based on my observations, it seems that the first couple of studies in a new niche tend to use male-only samples, with mixed-sex and female-only studies trickling in later. However, less active areas of research also tend to be areas of research that aren’t quite as interesting to lifters and coaches – topics with greater general interest also tend to attract greater research interest (and vice versa). So, I don’t think this potential drawback fundamentally impacts my ability to do what I set out to do with this article: analyze male and female representation in the areas of research that are most relevant to lifters.

Fourth, the approach I took in this article – letting systematic reviews and meta-analyses do a lot of the heavy lifting for me – may be slow to pick up on very recent trends. Research output is increasing every year, but this analysis only included nine studies from 2021 and one study from 2022 (compared to 63 studies from 2019 and 57 studies from 2020). The reason for this under-representation of very recent research is simple – there aren’t new meta-analyses about every topic, every month. If a meta-analysis was published in 2021, it may be based on a systematic search conducted in mid-2020, which would include all of the research output through 2019, half of the research output in 2020, and none of the research output in 2021 and 2022. To mitigate the impact of this drawback, I attempted to select the most recent systematic reviews and meta-analyses possible: none pre-dated 2017, and 27 of the 39 reviews were published in 2021 or 2022. However, it’s possible that a seismic shift in sex representation has occured within the past 18 months; if that has happened, the strategy I employed for this analysis would be unable to identify such a shift. For what it’s worth, I think this possibility is quite unlikely, but I still feel compelled to mention it in the interest of thoroughness.

Fifth, I could have been a bit more thorough and rigorous in my analysis of whether results from female-only studies differ from the rest of the literature. In theory, you could re-create every meta-analysis, and generate pooled effect estimates for male-only studies, mixed-sex studies, and female-only studies, and more rigorously compare those pooled effect estimates. In practice, doing this would have increased the time burden of this analysis by approximately 100-fold. Since I analyzed 67 forest plots, I would have needed to re-code all 628 studies and perform approximately 67 × 3 = 201 unique meta-analyses if I went this route. If someone wants to walk down that road, then more power to them. For practical purposes, just counting the number of overlapping versus non-overlapping confidence intervals is a quick and dirty method of analysis that’s way more feasible.

Sixth, if a meta-analysis made any data reporting errors (for example, reporting that a study had 28 subjects when it actually had 18 subjects, or reporting that a study had a mixed-sex cohort when it actually had a male-only cohort), those data reporting errors would be preserved in my analysis. I have to assume that such errors are rare and unlikely to be large enough to meaningfully change the outcomes of this analysis. Furthermore, this isn’t a unique drawback to this method of analysis – it’s not like original research never has data reporting errors, and it’s not like I’m immune to transcription errors. However, it’s worth mentioning in the interest of thoroughness.

Seventh, if the authors of a systematic review or meta-analysis failed to identify a relevant study or two when conducting their literature search, those studies would also be excluded from my analysis. Again, this isn’t a unique drawback – if I conducted my own systematic literature search, there’s no guarantee that I’d find every single relevant paper. However, I also don’t view this as a major drawback. Even if we were to assume that there’s a vast ocean of resistance training studies with female subjects that meta-analysts failed to find, the functional takeaways of this analysis wouldn’t change. If conventional systematic search strategies can’t find a study, that study doesn’t exist for all practical purposes. Research has utility insofar as it can be discovered, read, and used to inform future research and real-world application. If a study is indexed in any of the major databases, or if it’s ever been cited by other studies in its niche, a systematic literature search should be able to find it. If it’s not indexed and has never been cited, it basically doesn’t exist.

Eighth, general biases present in the scientific publishing industry will be preserved by any approach that analyzes published research output. Publication bias is the most prominent issue here: studies deemed to be more interesting by editors and reviewers are generally more likely to be published than studies deemed to be less interesting. Publication bias is mostly discussed in relation to statistical significance – statistically significant findings have an easier time getting published than null results. However, publication bias can also apply to novelty – studies addressing new research questions, or studies addressing old research questions in new populations are more likely to get published than studies addressing old research questions in well-trod populations. If publication bias affected this analysis, it would likely result in a slight overestimation of the proportion of female research subjects in the area. Since most lines of research start with male-only samples, studies with female samples are typically novel or under-represented in a particular body of research, which would tend to make it a bit easier to publish new studies with female cohorts than male cohorts.

To be clear, I don’t think any of the potential drawbacks of my analysis strategy fundamentally alter the validity of my findings. Rather, I considered the potential drawbacks before I started, and determined that they were all either acceptable (i.e., they were unlikely to meaningfully change the results) or unavoidable (drawbacks inherent to sampling a portion of the published literature). Furthermore, since I took a somewhat novel approach to analyzing sex representation within the literature (rather than cribbing Costello’s and Cowley’s approach), I wanted to preempt some of the potential questions and criticisms my strategy might provoke. I also just have a tendency to excessively fixate on the potential weaknesses of my own work.

References

  1. This curiosity about the sheer magnitude of variability has actually paid dividends. A couple years ago, I was involved in a project that helped expose implausible results coming from a highly productive exercise science researcher, which has resulted in several retractions. The first thing that made me skeptical of this research was the incredibly homogeneous training responses reported in this researcher’s papers.
  2. Nuckols G. The effects of biological sex on fatigue during and recovery from resistance exercise. Thesis, University of North Carolina at Chapel Hill (2019).
  3. Costello JT, Bieuzen F, Bleakley CM. Where are all the female participants in Sports and Exercise Medicine research? Eur J Sport Sci. 2014;14(8):847-51. doi: 10.1080/17461391.2014.911354. Epub 2014 Apr 25. PMID: 24766579.
  4. Cowley ES, Olenick AA, McNulty KL, Ross EZ. “Invisible Sportswomen”: The Sex Data Gap in Sport and Exercise Science Research. Women in Sport and Physical Activity Journal. 2021;29(2):146-151.
  5. I know what some of you are thinking. Yes, I know about it. After recent crackdowns, its coverage isn’t as good as it used to be, and some publishers now have systems to block access.
  6. Alvares TS, Oliveira GV, Volino-Souza M, Conte-Junior CA, Murias JM. Effect of dietary nitrate ingestion on muscular performance: a systematic review and meta-analysis of randomized controlled trials. Crit Rev Food Sci Nutr. 2021 Feb 8:1-23. doi: 10.1080/10408398.2021.1884040. Epub ahead of print. PMID: 33554654.
  7. Ashtary-Larky D, Bagheri R, Tinsley GM, Asbaghi O, Paoli A, Moro T. Effects of intermittent fasting combined with resistance training on body composition: a systematic review and meta-analysis. Physiol Behav. 2021 Aug 1;237:113453. doi: 10.1016/j.physbeh.2021.113453. Epub 2021 May 11. PMID: 33984329.
  8. Baz-Valle E, Balsalobre-Fernández C, Alix-Fages C, Santos-Concejero J. A Systematic Review of The Effects of Different Resistance Training Volumes on Muscle Hypertrophy. J Hum Kinet. 2022 Feb 10;81:199-210. doi: 10.2478/hukin-2022-0017. PMID: 35291645; PMCID: PMC8884877.
  9. Bello HJ, Caballero-García A, Pérez-Valdecantos D, Roche E, Noriega DC, Córdova-Martínez A. Effects of Vitamin D in Post-Exercise Muscle Recovery. A Systematic Review and Meta-Analysis. Nutrients. 2021 Nov 10;13(11):4013. doi: 10.3390/nu13114013. PMID: 34836268; PMCID: PMC8619231.
  10. Carey CC, Lucey A, Doyle L. Flavonoid Containing Polyphenol Consumption and Recovery from Exercise-Induced Muscle Damage: A Systematic Review and Meta-Analysis. Sports Med. 2021 Jun;51(6):1293-1316. doi: 10.1007/s40279-021-01440-x. Epub 2021 Mar 9. PMID: 33687663.
  11. Carvalho L, Junior RM, Barreira J, Schoenfeld BJ, Orazem J, Barroso R. Muscle hypertrophy and strength gains after resistance training with different volume-matched loads: a systematic review and meta-analysis. Appl Physiol Nutr Metab. 2022 Apr;47(4):357-368. doi: 10.1139/apnm-2021-0515. Epub 2022 Jan 11. PMID: 35015560.
  12. Coleman JL, Carrigan CT, Margolis LM. Body composition changes in physically active individuals consuming ketogenic diets: a systematic review. J Int Soc Sports Nutr. 2021 Jun 5;18(1):41. doi: 10.1186/s12970-021-00440-6. PMID: 34090453; PMCID: PMC8180141.
  13. Cuthbert M, Haff GG, Arent SM, Ripley N, McMahon JJ, Evans M, Comfort P. Effects of Variations in Resistance Training Frequency on Strength Development in Well-Trained Populations and Implications for In-Season Athlete Training: A Systematic Review and Meta-analysis. Sports Med. 2021 Sep;51(9):1967-1982. doi: 10.1007/s40279-021-01460-7. Epub 2021 Apr 22. PMID: 33886099; PMCID: PMC8363540.
  14. Dankel SJ, Kang M, Abe T, Loenneke JP. Resistance training induced changes in strength and specific force at the fiber and whole muscle level: a meta-analysis. Eur J Appl Physiol. 2019 Jan;119(1):265-278. doi: 10.1007/s00421-018-4022-9. Epub 2018 Oct 24. PMID: 30357517.
  15. Davies TB, Tran DL, Hogan CM, Haff GG, Latella C. Chronic Effects of Altering Resistance Training Set Configurations Using Cluster Sets: A Systematic Review and Meta-Analysis. Sports Med. 2021 Apr;51(4):707-736. doi: 10.1007/s40279-020-01408-3. Epub 2021 Jan 21. PMID: 33475986.
  16. Davies TB, Kuang K, Orr R, Halaki M, Hackett D. Effect of Movement Velocity During Resistance Training on Dynamic Muscular Strength: A Systematic Review and Meta-Analysis. Sports Med. 2017 Aug;47(8):1603-1617. doi: 10.1007/s40279-017-0676-4. PMID: 28105573.
  17. Doma K, Ramachandran AK, Boullosa D, Connor J. The Paradoxical Effect of Creatine Monohydrate on Muscle Damage Markers: A Systematic Review and Meta-Analysis. Sports Med. 2022 Feb 26. doi: 10.1007/s40279-022-01640-z. Epub ahead of print. PMID: 35218552.
  18. García-Valverde A, Manresa-Rocamora A, Hernández-Davó JL, Sabido R. Effect of weightlifting training on jumping ability, sprinting performance and squat strength: A systematic review and  meta-analysis. International Journal of Sports Science & Coaching. December 2021. doi:10.1177/17479541211061695
  19. Grgic J, Rodriguez RF, Garofolini A, Saunders B, Bishop DJ, Schoenfeld BJ, Pedisic Z. Effects of Sodium Bicarbonate Supplementation on Muscular Strength and Endurance: A Systematic Review and Meta-analysis. Sports Med. 2020 Jul;50(7):1361-1375. doi: 10.1007/s40279-020-01275-y. PMID: 32096113.
  20. Grgic J, Lazinica B, Mikulic P, Krieger JW, Schoenfeld BJ. The effects of short versus long inter-set rest intervals in resistance training on measures of muscle hypertrophy: A systematic review. Eur J Sport Sci. 2017 Sep;17(8):983-993. doi: 10.1080/17461391.2017.1340524. Epub 2017 Jun 22. PMID: 28641044.
  21. Grgic J, Mikulic I, Mikulic P. Acute and Long-Term Effects of Attentional Focus Strategies on Muscular Strength: A Meta-Analysis. Sports (Basel). 2021 Nov 12;9(11):153. doi: 10.3390/sports9110153. PMID: 34822352; PMCID: PMC8622562.
  22. Grønfeldt BM, Lindberg Nielsen J, Mieritz RM, Lund H, Aagaard P. Effect of blood-flow restricted vs heavy-load strength training on muscle strength: Systematic review and meta-analysis. Scand J Med Sci Sports. 2020 May;30(5):837-848. doi: 10.1111/sms.13632. Epub 2020 Feb 21. PMID: 32031709.
  23. Hackett DA, Ghayomzadeh M, Farrell SN, Davies TB, Sabag A. Influence of total repetitions per set on local muscular endurance: A systematic review with meta-analysis and meta-regression. Science & Sports. 2022.
  24. Heidel KA, Novak ZJ, Dankel SJ. Machines and free weight exercises: a systematic review and meta-analysis comparing changes in muscle size, strength, and power. J Sports Med Phys Fitness. 2021 Oct 5. doi: 10.23736/S0022-4707.21.12929-9. Epub ahead of print. PMID: 34609100.
  25. Hickmott LM, Chilibeck PD, Shaw KA, Butcher SJ. The Effect of Load and Volume Autoregulation on Muscular Strength and Hypertrophy: A Systematic Review and Meta-Analysis. Sports Med Open. 2022 Jan 15;8(1):9. doi: 10.1186/s40798-021-00404-9. PMID: 35038063; PMCID: PMC8762534.
  26. Jones L, Bailey SJ, Rowland SN, Alsharif N, Shannon OM, Clifford T. The Effect of Nitrate-Rich Beetroot Juice on Markers of Exercise-Induced Muscle Damage: A Systematic Review and Meta-Analysis of Human Intervention Trials. J Diet Suppl. 2021 Jun 21:1-23. doi: 10.1080/19390211.2021.1939472. Epub ahead of print. PMID: 34151694.
  27. Kassiano W, Nunes JP, Costa B, Ribeiro AS, Schoenfeld BJ, Cyrino ES. Does Varying Resistance Exercises Promote Superior Muscle Hypertrophy and Strength Gains? A Systematic Review. J Strength Cond Res. 2022 Apr 1. doi: 10.1519/JSC.0000000000004258. Epub ahead of print. PMID: 35438660.
  28. Krzysztofik M, Wilk M, Wojdała G, Gołaś A. Maximizing Muscle Hypertrophy: A Systematic Review of Advanced Resistance Training Techniques and Methods. Int J Environ Res Public Health. 2019 Dec 4;16(24):4897. doi: 10.3390/ijerph16244897. PMID: 31817252; PMCID: PMC6950543.
  29. Liao K, Nassis GP, Bishop C, Yang W, Bian C, Li Y. Effects of unilateral vs. bilateral resistance training interventions on measures of strength, jump, linear and change of direction speed: a systematic review and meta-analysis. Biology of Sport. 2022;39(3):485-497. doi:10.5114/biolsport.2022.107024.
  30. Lim MT, Pan BJ, Toh DWK, Sutanto CN, Kim JE. Animal Protein versus Plant Protein in Supporting Lean Mass and Muscle Strength: A Systematic Review and Meta-Analysis of Randomized Controlled Trials. Nutrients. 2021 Feb 18;13(2):661. doi: 10.3390/nu13020661. PMID: 33670701; PMCID: PMC7926405.
  31. Moesgaard L, Beck MM, Christiansen L, Aagaard P, Lundbye-Jensen J. Effects of Periodization on Strength and Muscle Hypertrophy in Volume-Equated Resistance Training Programs: A Systematic Review and Meta-analysis. Sports Med. 2022 Jan 19. doi: 10.1007/s40279-021-01636-1. Epub ahead of print. PMID: 35044672.
  32. Morris SJ, Oliver JL, Pedley JS, Haff GG, Lloyd RS. Comparison of Weightlifting, Traditional Resistance Training and Plyometrics on Strength, Power and Speed: A Systematic Review with Meta-Analysis. Sports Med. 2022 Jan 13. doi: 10.1007/s40279-021-01627-2. Epub ahead of print. PMID: 35025093.
  33. Morton RW, Murphy KT, McKellar SR, Schoenfeld BJ, Henselmans M, Helms E, Aragon AA, Devries MC, Banfield L, Krieger JW, Phillips SM. A systematic review, meta-analysis and meta-regression of the effect of protein supplementation on resistance training-induced gains in muscle mass and strength in healthy adults. Br J Sports Med. 2018 Mar;52(6):376-384. doi: 10.1136/bjsports-2017-097608. Epub 2017 Jul 11. Erratum in: Br J Sports Med. 2020 Oct;54(19):e7. PMID: 28698222; PMCID: PMC5867436.
  34. Murphy C, Koehler K. Energy deficiency impairs resistance training gains in lean mass but not strength: A meta-analysis and meta-regression. Scand J Med Sci Sports. 2022 Jan;32(1):125-137. doi: 10.1111/sms.14075. Epub 2021 Oct 13. PMID: 34623696.
  35. Nunes JP, Grgic J, Cunha PM, Ribeiro AS, Schoenfeld BJ, de Salles BF, Cyrino ES. What influence does resistance exercise order have on muscular strength gains and muscle hypertrophy? A systematic review and meta-analysis. Eur J Sport Sci. 2021 Feb;21(2):149-157. doi: 10.1080/17461391.2020.1733672. Epub 2020 Feb 28. PMID: 32077380.
  36. Oranchuk DJ, Storey AG, Nelson AR, Cronin JB. Isometric training and long-term adaptations: Effects of muscle length, intensity, and intent: A systematic review. Scand J Med Sci Sports. 2019 Apr;29(4):484-503. doi: 10.1111/sms.13375. Epub 2019 Jan 13. PMID: 30580468.
  37. Pallarés JG, Hernández-Belmonte A, Martínez-Cava A, Vetrovsky T, Steffl M, Courel-Ibáñez J. Effects of range of motion on resistance training adaptations: A systematic review and meta-analysis. Scand J Med Sci Sports. 2021 Oct;31(10):1866-1881. doi: 10.1111/sms.14006. Epub 2021 Jul 5. PMID: 34170576.
  38. Rosa A, Vazquez G, Grgic J, Balachandran AT, Orazem J, Schoenfeld BJ. Hypertrophic Effects of Single- Versus Multi-Joint Exercise of the Limb Muscles: A Systematic Review and Meta-analysis. Strength and Conditioning Journal. April 6, 2022. doi: 10.1519/SSC.0000000000000720
  39. Sabag A, Najafi A, Michael S, Esgin T, Halaki M, Hackett D. The compatibility of concurrent high intensity interval training and resistance training for muscular strength and hypertrophy: a systematic review and meta-analysis. J Sports Sci. 2018 Nov;36(21):2472-2483. doi: 10.1080/02640414.2018.1464636. Epub 2018 Apr 16. PMID: 29658408.
  40. Schoenfeld BJ, Ogborn DI, Vigotsky AD, Franchi MV, Krieger JW. Hypertrophic Effects of Concentric vs. Eccentric Muscle Actions: A Systematic Review and Meta-analysis. J Strength Cond Res. 2017 Sep;31(9):2599-2608. doi: 10.1519/JSC.0000000000001983. PMID: 28486337.
  41. Valenzuela PL, Morales JS, Castillo-García A, Lucia A. Acute Ketone Supplementation and Exercise Performance: A Systematic Review and Meta-Analysis of Randomized Controlled Trials. Int J Sports Physiol Perform. 2020 Feb 10:1-11. doi: 10.1123/ijspp.2019-0918. Epub ahead of print. PMID: 32045881.
  42. Vårvik FT, Bjørnsen T, Gonzalez AM. Acute Effect of Citrulline Malate on Repetition Performance During Strength Training: A Systematic Review and Meta-Analysis. Int J Sport Nutr Exerc Metab. 2021 Jul 1;31(4):350-358. doi: 10.1123/ijsnem.2020-0295. Epub 2021 May 19. PMID: 34010809.
  43. Vieira AF, Umpierre D, Teodoro JL, Lisboa SC, Baroni BM, Izquierdo M, Cadore EL. Effects of Resistance Training Performed to Failure or Not to Failure on Muscle Strength, Hypertrophy, and Power Output: A Systematic Review With Meta-Analysis. J Strength Cond Res. 2021 Apr 1;35(4):1165-1175. doi: 10.1519/JSC.0000000000003936. PMID: 33555822.
  44. Zabaleta-Korta A, Fernández-Peña E, Santos-Concejero J. Regional Hypertrophy, the Inhomogeneous Muscle Growth: A Systematic Review. Strength Cond J. 2020 Oct;42(5):94-101. doi: 10.1519/SSC.0000000000000574.
  45. Ayers JL, DeBeliso M, Sevene TG, Adams KJ. Hang cleans and hang snatches produce similar improvements in female collegiate athletes. Biol Sport. 2016 Sep;33(3):251-6. doi: 10.5604/20831862.1201814. Epub 2016 May 10. PMID: 27601779; PMCID: PMC4993140.
  46. Slovak, Bárbara et al. EFFECTS OF TRADITIONAL STRENGTH TRAINING AND OLYMPIC WEIGHTLIFTING IN HANDBALL PLAYERS. Revista Brasileira de Medicina do Esporte [online]. 2019, v. 25, n. 3 [Accessed 13 May 2022] , pp. 230-234. Available from: <https://doi.org/10.1590/1517-869220192503210453>. Epub 01 July 2019. https://doi.org/10.1590/1517-869220192503210453.
  47. Moore EW, Hickey MS, Reiser RF. Comparison of two twelve week off-season combined training programs on entry level collegiate soccer players’ performance. J Strength Cond Res. 2005 Nov;19(4):791-8. doi: 10.1519/R-15384.1. PMID: 16287374.
  48. Boyer BT. A Comparison of the Effects of Three Strength Training Programs on Women. Journal of Strength and Conditioning Research: August 1990 – Volume 4 – Issue 3 – p 88-94
  49. Nuzzo JL. Sex Difference in Participation in Muscle-Strengthening Activities. J Lifestyle Med. 2020 Jul 31;10(2):110-115. doi: 10.15280/jlm.2020.10.2.110. PMID: 32995338; PMCID: PMC7502892.
  50. McNulty KL, Elliott-Sale KJ, Dolan E, Swinton PA, Ansdell P, Goodall S, Thomas K, Hicks KM. The Effects of Menstrual Cycle Phase on Exercise Performance in Eumenorrheic Women: A Systematic Review and Meta-Analysis. Sports Med. 2020 Oct;50(10):1813-1827. doi: 10.1007/s40279-020-01319-3. PMID: 32661839; PMCID: PMC7497427.
  51. Elliott-Sale KJ, McNulty KL, Ansdell P, Goodall S, Hicks KM, Thomas K, Swinton PA, Dolan E. The Effects of Oral Contraceptives on Exercise Performance in Women: A Systematic Review and Meta-analysis. Sports Med. 2020 Oct;50(10):1785-1812. doi: 10.1007/s40279-020-01317-5. PMID: 32666247; PMCID: PMC7497464.
  52. Haizlip KM, Harrison BC, Leinwand LA. Sex-based differences in skeletal muscle kinetics and fiber-type composition. Physiology (Bethesda). 2015 Jan;30(1):30-9. doi: 10.1152/physiol.00024.2014. PMID: 25559153; PMCID: PMC4285578.
  53. Roberts BM, Nuckols G, Krieger JW. Sex Differences in Resistance Training: A Systematic Review and Meta-Analysis. J Strength Cond Res. 2020 May;34(5):1448-1460. doi: 10.1519/JSC.0000000000003521. PMID: 32218059.
  54. Malowany JM, West DWD, Williamson E, Volterman KA, Abou Sawan S, Mazzulla M, Moore DR. Protein to Maximize Whole-Body Anabolism in Resistance-trained Females after Exercise. Med Sci Sports Exerc. 2019 Apr;51(4):798-804. doi: 10.1249/MSS.0000000000001832. PMID: 30395050.
  55. Bandegan A, Courtney-Martin G, Rafii M, Pencharz PB, Lemon PW. Indicator Amino Acid-Derived Estimate of Dietary Protein Requirement for Male Bodybuilders on a Nontraining Day Is Several-Fold Greater than the Current Recommended Dietary Allowance. J Nutr. 2017 May;147(5):850-857. doi: 10.3945/jn.116.236331. Epub 2017 Feb 8. PMID: 28179492.
  56. I double checked, and none of them included the presence of female subjects as an exclusion criterion
  57. A 2020 meta-analysis by Hagstrom and colleagues surveyed the literature examining longitudinal resistance training outcomes in females. They found that, at the time of publication, only 24 such studies existed. It feels like 24 longitudinal resistance training studies in male subjects are published every month. So, I do think it’s likely that female lifters are even more under-represented in the most relevant and impactful areas of resistance training research.
  58. The finding that female subjects account for just 25% of the total subject pool in resistance training-related research comports well with a recent analysis by Smith and colleagues. They investigated sex representation in supplement research, and found that female subjects accounted for just 23% of the total subject pool in studies investigating the effects of β-alanine, caffeine, creatine, glycerol, nitrate/beetroot juice and sodium bicarbonate. 1826 studies with 34,889 participants were included in their analysis.
Scroll to Top