The highest court in Germany ruled recently in favor of the introduction of a third gender category for individuals who do not identify as either male or female or are born with an ambiguous sexual anatomy (1). This ruling is in line with at least 10 nations or states that allow a third legal gender (2). This perspective outlines the implications of a third gender for elite sport and is a summary of proceedings of a session entitled “Beyond Fairness: The Biology of Inclusion for Transgender and Intersex Athletes” at the 10th European Sport Medicine — Congress of the European Federation of Sports Medicine Associations (EFSMA) that took place in Estoril, Lisbon, Portugal from November 16 to 18, 2017 (3).
Although it may seem simple enough to determine which athletes compete in the men’s division and which athletes compete in the women’s division — after all humans get sorted into male and female bins from the moment of birth — the question of precisely who should compete in the male and female categories is one of the most emotional and complex issues facing elite sport in recent times. Despite the separation of neonates into bundles of “pink” or “blue,” it should be noted that biology does not neatly divide humans into two sexes. There are two main groups of people (i.e., intersex and transgender) who fall outside of the binary division that most people take for granted. Intersex individuals have chromosomal or physical conditions that blur the line separating men from women. Intersex conditions are often referred to as disorders of sexual development or differences of sexual development (DSD). Transgender individuals can be defined as those whose gender identity — an innate sense of whether one is male, female, or somewhere in between — differs from their assigned sex at birth. Some athletes from each of these two groups would like to compete in women’s sport. For this perspective, we focus on issues raised by intersex and transgender athletes in women’s sport because it is generally assumed that transitioning from a woman to a man would not confer a similar competitive advantage.
The question of whether transgender and/or intersex women should be allowed to compete against other women, and if so then under what conditions, is extremely contentious. Examining the fundamental concepts of sex and gender can help provide some insights. For the purpose of this perspective, sex is defined as those biological qualities that make humans male, female, or somewhere in between, while gender can be thought of as a nonbinary social construct. Humans possess at least six distinct manifestations of biological sex; internal and external genitalia, chromosomes, hormone levels, secondary sex characteristics, and gender identity. Only when all of the elements of biological sex are examined can one say that someone is male, female, or somewhere in between. The inclusion, however, of gender identity as one of the biological components of sex is not without controversy, albeit an ever larger number of scientists seem prepared to endorse this position (4,5). There is a great deal of debate around the complex and multifaceted concept of gender as a social construct (6). Additionally, terms, such as gender assigned at birth (7) and legal gender (8), are commonly used with the understanding that one’s legal gender might be different from the gender assigned at birth. Given the multifaceted nature of gender, we propose the concept of an “athletic gender” for the purpose of elite competitive sport.
Individuals are assigned a gender at birth based on the appearance of one’s external genitalia, one component of biological sex. The external genitalia of many intersex individuals are atypical in appearance, and the process of assigning gender becomes far more complex. An enlightened and informed society should allow people to live as male, female, or somewhere in between, depending on their individual gender identity. In other words, an individual’s gender identity can be used to define their social gender. While self-defined social gender is very important in shaping the way an individual chooses to live, gender identity alone is not sufficient to determine the appropriate sports category for any individual. Our legal gender usually matches the gender assigned at birth but laws dealing with intersex and transgender individuals vary widely depending on the given country. In addition to the 11 or more nations or states that allow a third legal gender, a further seven countries allow people to change their gender markers by self-declaration (9). On the other hand, many other countries either refuse to consider any legal gender beyond that defined at birth or make it extremely difficult to make changes to the legal gender at birth (9). Legal gender, which does not define one’s athletic capabilities, cannot therefore be the sole criterion to determine appropriate sporting categories especially when dealing with elite sport. The concept of an athletic gender is in line with the Fundamental Principles of the Olympic Charter (10). In particular “The enjoyment of the rights and freedoms set forth in this Olympic Charter shall be secured without discrimination of any kind, such as race, colour, sex, sexual orientation, language, religion, political or other opinion, national or social origin, property, birth, or other status.” For the Olympic Charter to be upheld in a way that ensures meaningful sporting competition, it is imperative that the athletic gender concept be defined using scientifically determined performance-based metrics.
When the International Olympic Committee (IOC) and the International Association of Athletics Federations (IAAF) abandoned chromosome testing in the 1990s, they took differing paths in their treatment of intersex and later transgender athletes but both organisations have converged around the notion that testosterone levels should be the key component of their regulations on gender variant athletes (11–13). Additionally, the National Collegiate Athletic Association (NCAA) also adopted testosterone-based rules to govern the participation of transgender athletes in 2011 (14). Despite a paucity of research in this area, the general consensus is that differences in testosterone levels is currently the single most important factor contributing to the performance differences between male and female athletes (15). This view also is supported by the results of a recent study that demonstrated a significant advantage for women with higher testosterone levels in selected events at two recent world track and field championships (16). Specifically, when compared with the lowest female testosterone tertile, women with the highest tertile for testosterone performed significantly better in 400 m, 400 m hurdles, 800 m, hammer throw, and pole vault with margins of approximately 2.7%, 2.8%, 1.8%, 4.5%, and 2.9%, respectively (16). If there is equitable competition between the two groups of athletes, the two groups should be present in championship events in numbers that are consistent with their representation in the general population. At the 2011 IAAF world championships, hyperandrogenic athletes (i.e., a subset of intersex athletes) were overrepresented by a factor of 140, an indirect measure of a major advantage (15). Moreover, historical accounts of doping in the previous German Democratic Republic (GDR) with testosterone as part of a regime of different doping substances and methods indirectly confirm the performance enhancing effects of testosterone (17).
Biological research studies on intersex or transgender athletes are rare, as it is extraordinarily difficult to gather data from small and at times secretive populations. In one of two notable studies, Bermon (18) measured the athletic performance of three intersex athletes who had undergone medical intervention. After 2 yr of testosterone suppression, the three athletes were collectively running 5.7% slower (18). In another study, race times were obtained from eight transgender distance runners before and after transition (19). The run times were then compared using the age-graded methodology, the standard method used by World Master’s Athletics to compare male and female athletes of all ages (20). Collectively, the eight runners had similar age-graded scores before and after transition. Given these results, it is unlikely that a transitioned male to female athlete with testosterone levels within the normal female range will retain the same relative ability after transition; a powerful argument for the performance enhancing effects of testosterone.
The intention for separating athletes into male and female categories is to provide women athletes with meaningful competition. It would be equally reasonable therefore to allow transgender or intersex women to compete with other female athletes if, and only if, the inclusion of these athletes did not unduly alter the playing field for other women. If one can minimize athletic advantages potentially held by gender-variant athletes, then one can satisfy this requirement. It may not be necessary, or even possible, to eliminate all individual advantages held by a transgender or intersex woman. Any remaining advantage, theoretical or real, held by a gender-variant athlete would be only one component of their unique makeup. It is paramount, however, that all athletes competing have a chance to succeed — albeit not necessarily an equal chance and in line with the true essence of sport.
The initial fear that transgender women would dominate women’s sport has simply not materialised, in large part due to the effects of testosterone suppression. Transgender women are majorly underrepresented in elite level sport. For instance, there were approximately 4700 female competitors (21) and 460 female medal winners (22) in the Rio Olympic Games. Assuming that transgender individuals make up approximately 0.6% of the population (23), then there should have been 20 to 25 transgender women competitors and two or three medal winners at the 2016 Olympics. In contrast to this, no openly transgender athlete has ever competed in any Olympic Games, let alone won a medal (24). Opponents of testosterone-based methods often suggest that testing is overly invasive (25), but this argument has little basis. A simple blood test, routinely performed in annual physical evaluations worldwide can be used to determine testosterone levels. Only when a woman’s testosterone levels are out of the normal range for women do the tests become more invasive and even then, the testing involved is not any more invasive than the routine drug testing that elite athletes are required to undergo. These urinalysis tests for doping control require an observer to watch the urine leaving the athlete’s body and enter the collection cup. It is understood by elite athletes that this loss of privacy is a price that must be paid for equitable sport.
While imperfect, serum testosterone levels are currently the best available method to fairly separate male and female athletes. Although there is an obvious need for more scientific data to support the idea of testosterone- and/or other biomarker-based athletic gender, the study of hormonal variations and muscle memory and their effect on athletic performance is attracting more research interest all the time (26). A concerted international research effort aimed at improving the identification of new biomarkers better able to discriminate athletic gender will undoubtedly emerge in the near future as research intensifies. For example, an improved approach may involve the use of bioavailable testosterone, that is, the testosterone that is taken up and used by the body’s cells, coupled with an allowance for androgen insensitivity.
In conclusion, to uphold the Olympic Charter and ensure meaningful sporting competition, it is necessary to use an evolving evidence-based scientific method to separate athletes into male and female categories. Although imperfect, the best currently available scientific approach is the use of serum testosterone levels. It is helpful to view the separation of athletes into male and female categories as the determination of an athletic gender. If the idea of an athletic gender is adapted, the increased use of this concept will result in clearer sporting policies and a reduction in the discord between various factions in the very complex world of sex, gender, and sport.
The authors declare no conflict of interest and do not have any financial disclosures.
5. Saraswat A, Weinand J, Safer J. Evidence supporting the biologic nature of gender identity. Endocr. Pract
. 2015; 21:199–204.
15. Bermon S, Garnier PY, Hirschberg AL, et al. Serum androgen levels in elite female athletes. J. Clin. Endocrinol. Metab
. 2014; 99:4328–35.
16. Bermon S, Garnier PY. Serum androgen levels and their relation to performance in track and field: mass spectrometry results from 2127 observations in male and female elite athletes. Br. J. Sports Med
. 2017; 51:1309–14.
17. Franke WW, Berendonk B. Hormonal doping and androgenization of athletes: a secret program of the German Democratic Republic government. Clin. Chem
. 1997; 43:1262–79.
18. Bermon S. Androgens and athletic performance of elite female athletes. Curr. Opin. Endocrinol. Diabetes Obes
. 2017; 24:246–51.
19. Harper J. Race times for transgender athletes. J. Sporting Cultures Identities
. 2015; 6:1–9.
20. Grubb HJ. Models for comparing athletic performances. The Statistician
. 1998; 47:509–52.
25. Katz RS, Luckinbill R. Changing sex/gender roles in sport. Stanford Law and Policy Review
. 2017; 28:215–43.
26. Pitsiladis Y, Harper J, Martinez-Patino MJ, et al. Beyond fairness: the biology of inclusion for transgender and intersex athletes. Curr. Sports Med. Rep
. 2016; 15:386–8.