Auditing the Representation of Female Versus Male Athletes in Sports Science and Sports Medicine Research: Evidence-Based Performance Supplements

Abstract

Although sports nutrition guidelines promote evidence-based practice, it is unclear whether women have been adequately included in the underpinning research. In view of the high usage rates of performance supplements by female athletes, we conducted a standardised audit of the literature supporting evidence-based products: β-alanine, caffeine, creatine, glycerol, nitrate/beetroot juice and sodium bicarbonate. Within 1826 studies totalling 34,889 participants, just 23% of participants were women, although 34% of studies included at least one woman. Across different supplements, 0-8% of studies investigated women exclusively, while fewer (0-2%) were specifically designed to compare sex-based responses. The annual publication of female-specific studies was ~8 times fewer than those investigating exclusively male cohorts. Interestingly, 15% of the female participants were classified as international/world-class athletes, compared with 7% of men. Most studies investigated performance outcomes but displayed poorer representation of women (16% of participants), whereas health-focussed studies had the greatest proportion of female participants (35%). Only 14% of studies including women attempted to define menstrual status, with only three studies (~0.5%) implementing best practice methodologies to assess menstrual status. New research should target the efficacy of performance supplements in female athletes, and future sports nutrition recommendations should specifically consider how well female athletes have contributed to the evidence-base.

Keywords: menstrual status; nutrition; oral contraceptive; physical activity; women.

Figure 1

Flowchart demonstrating the primary screening process for each performance supplement audited, and the total number of individual studies included for extraction of metrics A–F.

Figure 2

(A) The total number of male and female participants included across all studies, separated per supplement, (B) the total number of studies including at least one male or female participant, per supplement, and (C) histogram displaying the total number of studies published in exclusively male or female participants between 1975 and 2021 across all performance supplements.

Figure 3

(A) The proportion of studies published in each population, separated by performance supplement, (B) the median number of male and female participants per study, and (C) total number of male and female participants, according to the population studied. Male versus female (MvF).

Figure 4

The total (A) and proportion (B,C) of male and female participants in each athletic tier [20]: Tier 0 (sedentary), Tier 1 (recreationally active), Tier 2 (trained/developmental), Tier 3 (highly trained/national), Tier 4 (elite/international) and Tier 5 (world-class). (D) The proportion of participants in each athletic tier according to the supplement investigated. Only classifiable participants are reported in all figures.

Figure 5

Proportion of studies classified into each tier [1], reflecting the standard of methodological control regarding ovarian hormonal profiles, as a proportion of the total number of studies including women across each supplement. Gold standard (best practice methodologies, as outlined by Elliott-Sale et al. [27]), silver/bronze (achieve some, but not all, methodological considerations), ungraded (menstrual status is defined, but methodological control is insufficient to award bronze/silver/gold) or unclassified (insufficient information to provide a robust classification of participants, or a mixed female cohort in which individual menstrual status cannot be discerned).

Figure 6

Percentage of studies in each theme, separated by (A) population (colours correspond to those used in Figure 3A) and (B) performance supplement. Number of participants represented as the (C) absolute number and (D) percentage of the total participants in each theme: performance (direct performance outcomes), health (outcomes related to health status/condition) and indirect associations with performance/health (a physiological or psychological adaptation/response that may subsequently transfer to athletic performance/health). Male versus female (MvF).

Figure 7

Median journal impact factor (A) and Altmetric score (B) across all performance supplements, separated by population studied. * Significantly greater impact factor compared to the nitrate studies in a female-only cohort (p < 0.01). Error bars display the interquartile range. Male versus female (MvF).