Secondary Logo

Journal Logo

International Federation of Sports Medicine: Section Articles

The Fluidity of Gender and Implications for the Biology of Inclusion for Transgender and Intersex Athletes

Harper, Joanna MS1; Lima, Giscard BPhEd, MSc2; Kolliari-Turner, Alexander BA3; Malinsky, Fernanda Rossell PhD3; Wang, Guan PhD3; Martinez-Patino, Maria Jose4; Angadi, Siddhartha S. PhD5; Papadopoulou, Theodora MD, PhD6,7; Pigozzi, Fabio MD, PhD2,7,8; Seal, Leighton9; Barrett, James9; Pitsiladis, Yannis P. MMedSci, PhD2,3,7

Author Information
Current Sports Medicine Reports: December 2018 - Volume 17 - Issue 12 - p 467-472
doi: 10.1249/JSR.0000000000000543
  • Free



On April 23, 2018, the International Association of Athletics Federations (IAAF) released a new policy entitled “Eligibility Regulations for the Female Classification” and subtitled “Athletes with Differences of Sex Development (DSD)” (1). Implementation of the policy has been delayed until after a verdict, expected no later than March 26, 2019, in the Semenya versus IAAF trial in the Court of Arbitration for Sport. If the policy is enacted, it will restrict athletes from competing in the female athletics category with specific DSD in races from 400 m up to the mile in international level competitions unless the athletes lower their natural testosterone (T) levels below 5 nmol·L−1 (2). To thoroughly assess this new IAAF policy, one needs to appreciate its legal, sociological, and scientific underpinnings but also the history of previous policies attempting to define precisely how athletes should be divided into male and female categories. The challenge faced by the governing body of athletics, IAAF, is to formulate a policy that upholds both international law and the Olympic charter that stipulates athletes compete without discrimination of any kind (3).

Some view the new policy as a direct attack on the South African athlete, Caster Semenya (2,4), largely because the new regulations are restricted to a range of events that she competes in. However, narrowing the scope of the new regulations to only four events largely resulted from the Dutee Chand case; the Indian sprinter who won a temporary suspension of the 2011 Hyperandrogenism (HA) (5) rules in 2015 (6). The Court of Arbitration for Sport (CAS) in the Chand ruling in 2015 against the IAAF required the IAAF to produce more evidence to back its regulations; and indeed, the CAS panel stated “there is presently insufficient evidence about the degree of the advantage that androgen-sensitive hyperandrogenic females enjoy over nonhyperandrogenic females” (section 522). Notably, CAS also determined that the HA regulations were discriminatory in nature. In section 500 of the CAS ruling, the panel stated that it accepted Chand’s contention that the regulations discriminated against women based on a naturally occurring trait (i.e., higher T), and in section 501, the panel directed the IAAF to establish that the HA regulations were a proportionate means to provide fairness to all female athletes beyond a reasonable doubt. The discriminatory nature of the 2011 rules seems based on the fact that any intersex athlete who previously declined to lower their T level was prohibited from competing in any form of athletics. The CAS panel stated that “levels of endogenous testosterone are a key biological indicator of the differences between males and females. However, that is not the use to which endogenous T is being put under the HA Regulations. It is not being used to determine whether an athlete should compete as a male or as a female. Instead, it is being used to determine a new category of ineligible female athletes within the female category” (section 511) (6).

Unlike the 2011 HA rules, the new DSD regulations allow multiple avenues of competition for those DSD athletes who choose to maintain their natural T values. Hyperandrogenic athletes can compete in races longer or shorter than the restricted events; they can compete in the restricted events in nonelite competition, or in the men’s division, or in a future DSD division (1). The athletic careers of these DSD women are no longer over should they decide to maintain their high T levels, and this is a key legal difference between the new DSD rules and the old HA rules. Another important aspect of the DSD regulations is the recognition of the importance of legal sex (or legal gender) in the determination of relevant sporting category. In the 2015 CAS decision (i.e., the Chand case), the panel wrote that “whether a person is female is a matter of the law,” and followed this up with the statement that the IAAF had presented little evidence to indicate that “legal recognition as a female varies in most countries” (section 510) (6). It was only a matter of months after the CAS decision before one of the lawyers representing the IAAF in the case published an article that, among other things, explored the variety of laws affecting legal gender/sex recognition worldwide (7). We have previously stressed the need for evidence-based methods to determine appropriate criteria to separate male and female athletes (8). We also have suggested that 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” (9). The extent to which the rapidly expanding list of nations that are adopting a third legal sex would affect a future challenge to the new DSD regulations remains to be determined.

Without further direction from CAS, it is unclear whether the new DSD rules will be subject to the same level of scientific rigor as the CAS panel required of the HA regulations. Despite this uncertainty, there were two relevant studies published in 2017 (10,11). The first was a review in which it was reported that there were 3 intersex athletes who had competed for 2 years after reducing their T from the typical male (21–25 nmol·L−1) to a level below 10 nmol·L−1 (the cutoff limit in the HA regulations) with a resulting performance reduction of 5.7% (10). Although the number of athletes is small, they were elite athletes of the very highest level. The second was an article that compared race times of female competitors divided into tertiles depending on their T levels measured at the 2011 and 2013 World Championships (11). Statistically significant differences were found between the high-T and low-T tertiles in five events, the 400- and 800-m races, the 400-m hurdles, the pole vault and the hammer throw. This study was not intended to suggest that these are the only events in which higher T is beneficial in athletics, but that the T-based advantage is particularly important in these events. It is widely recognized that the primary difference between male and female athletes is the higher levels of circulating T between the sexes after puberty (12). Given the paucity of data on the ergogenicity of T, indirect evidence for the significant benefits of T was found in a study on doping practices that highlighted the performance inequalities of lower-T compared to higher-T athletes as a result of anabolic steroids (13).

Support for the idea to limit athletes with DSD to certain events would require an assessment of previous successes of such athletes in the relevant events. There have been several champions in athletic events on the restricted list over the previous 25 years that, in all probability, had a DSD. Although none of the athletes, including Caster Semenya, have publicly confirmed being DSD, we estimate that these athletes won a total of at least 19 global athletics championships — that is, Olympic or IAAF World Championship gold medals — and a total of at least 30 medals of all colors in distances from the 400 m up to the 1500 m; this estimate is supported by unpublished documents from the IAAF. A study from the 2011 Daegu world championships (14) reported an overrepresentation of DSD athletes of 140-fold representing an indirect measure of a significant advantage. Morel et al. (15) calculated the prevalence of XY DSD at 1 in 20,000, meaning that of the 354 medals available in global athletic championships in the restricted events over the past 25 years, athletes with DSD would have been expected to win only 0.0177 medals. Hence, the presumed 30 medals are an over-representation of approximately 1700-fold at the podium level. The estimated 30 medals won by athletes with DSD exceeds considerably the six medals won by Caster Semenya in her career to date, demonstrating the urgent need for DSD rules well beyond a means to deal with the South-African athlete; rules intended to produce the stated goal of fair and meaningful competition for all female athletes.

The restricted events of the new DSD rules, that is, 400 m through the mile, deviate from the five events showing differences between the high-T and low-T tertiles in the study by Bermon and Garnier (11). Of the five events, pole vault and hammer throw have not been included in the restricted list possibly due to the lack of success by athletes with DSD in those events. Most DSD are autosomal recessive conditions and are more prevalent in areas in which consanguinity is common (16). In the 21st century, this means that the majority of elite DSD athletes can be traced back to rural areas in developing nations (17,18). Therefore, these athletes are extremely unlikely to have access to the level of coaching necessary to excel in field events in general, and in the pole vault or hammer throw specifically. It is certainly possible that this lack of coaching expertise in developing countries could change in the future, and the IAAF has stated that the regulations will be subject to periodic review (1) allowing for changes in the events regulated. The second difference between the Bermon and Garnier events (11) and the IAAF restricted list is the addition of the 1500-m and mile distances to the three track events identified in the Bermon and Garnier study. A reasonable justification for this addition is that the peak event for athletes with DSD has, in all likelihood, been the 800 m. A solid majority of the 30 medals estimated previously were awarded in the 800 m, and historically, it has proven very easy for many 800-m athletes to move up to the 1500 m, while preserving their level of excellence (19). This reasoning is further evidence that the new DSD rules were not developed with only one athlete in mind, but rather are rooted in the performance history of the sport in general.

There also have been several suggestions that the new DSD regulations are racist in nature (20,21). Although evidently it is the case that the majority of elite DSD athletes in the 21st century are women of color, it also is true that many of the women who stand to gain the most from the restriction of these athletes also are women of color. For instance, Track and Field News gave 31 of the 40 top 10 world-ranking spots in the four restricted events to women of color in 2017 (22). It also is important to note that the IAAF and the International Olympic Committee (IOC) have been concerned about the effect of DSD athletes on the female playing field dating back to the 1930s, a time when virtually all of the DSD athletes were Caucasian (23). In fact, it is likely that white DSD athletes swept the podium in one track event at the 1936 Olympic Games in Berlin (24,25).

The new DSD rules also address an important criticism of the previous HA regulations. It was revealed by Fénichel et al. (17) that four athletes with 5α-reductase type 2 deficiency (5-ARD) had undergone a gonadectomy and other surgeries including a “partial clitoridectomy.” The gonadectomy would have brought the athletes’ T levels into the female range, but there was no athletic purpose and a risk of serious side effects to the clitoral surgery. The release of the Fenichel study brought substantial condemnation to the IAAF, with a response by Ha et al. (26), suggesting that the athletes needed to have some sort of advocate on their behalf attending them. The new DSD regulations do indeed include the provision for an “appointment of athlete ombudsman” (1) and the ombudsman addition is a substantial improvement upon the HA regulations.

It is appropriate to review the very reasons for the division of most sports into male and female categories. At the elite level, the primary reason for dividing sport by sex is that women cannot hope to compete effectively with men in most sports (27). Thus, it is reasonable to restrict those women who benefit from a male-type puberty as a result of their Y chromosomes and testes. It would be possible to restrict athletes with DSD using various methodologies; however, given the current level of scientific knowledge available, setting a T limit on women with DSD is the most reasonable method to achieve this restriction.

In similar fashion, the IOC enacted testosterone-based regulations for transgender athletes in 2016 (28). There is a general belief that transgender(ed) women athletes retain inherent anatomical and biological advantages as compared to cisgender(ed) women (i.e., women birth assigned female who remain in that gender role) and the same could be said about athletes with DSD who chose to lower their T. In contrast to this belief, the Harper study (29) reported comparable performance of eight transgender distance runners in both the male and female categories. Although this study is valuable, it is very limited in scope, and further studies will be necessary to reach a more complete understanding of the capabilities of transgender women athletes. No known transgender athlete has ever participated in the Olympic Games, although their participation in the Olympic Games has been allowed since 2004 (30). However, this situation is likely to change at the next Olympics (i.e., 2020 Tokyo Olympic Games). In 2017, the New Zealand weightlifter, Laurel Hubbard, won a silver medal at the World Weightlifting Championships and only missed out on a gold medal in the Commonwealth Games in 2018 due to a serious elbow injury. Should she recover from this injury, she will compete in the 2020 Olympic Games. Another transgender athlete with a real chance of competing at the next Olympic Game is the Brazilian Volleyball player Tiffany Abreu, who was the highest scoring player in the Brazilian Volleyball League 2017/2018. It remains to be determined if Tiffany Abreu will be permitted to compete; the International Volleyball Federation (FIVB) has empowered each national federation to determine the rules for participation of transgender athletes in national club competitions. In January of 2018, the FIVB Medical Commission met in Lausanne to discuss the matter of inclusion of transgender athletes and reiterated their desire to establish an appropriate system for the participation of athletes in FIVB competitions (31).

There also is a certain amount of controversy over the fact that transgender men are permitted to take exogenous T and compete against cisgender men who are prohibited from doing so. However, trans men are only permitted to take enough T to bring their hormone levels up to typical male values (32). Tracking the performance of American swimmer Schuyler Bailar (the first openly transgender Division I swimmer in the National Collegiate Athletic Association ranks) shows that Bailar has progressed from swimming the 100 yard breast stroke in 1:03.1 in March of 2013 as a high school girl’s swimmer, to 57.7 s in February 2018 as a third year men’s university swimmer (33). This large improvement in swim times is partially due to the performance-enhancing properties of T injections, and yet Bailar finishes much lower in the standings in men’s swimming than he would have done in women’s swimming. Although scientific data are needed, it is clear that any advantage that transgender athletes possess over their competitors needs to be differentiated from the advantage athletes gain from doping (34).

For the inclusion of DSD and transgender athletes in international sporting competition to be implemented in the fairest possible manner, decisions need to be evidence based. Given the paucity of relevant data and the need for an immediate solution (8,9), there is an urgent need for the scientific community to generate the essential know-how that can help resolve the complex issues surrounding the inclusion of both DSD and transgender athletes with the ultimate goal of achieving fair competition for all. With this challenge in mind, we recently proposed a system to deal with gender variant athletes that relies on a determination of an “athlete gender” (9). Briefly, the idea is to use serum T levels to separate athletes into male and female categories. To gauge opinion, the concept of an “athlete/athletic gender” was recently presented to the sports medicine community and the public and participants were invited to complete a short survey questionnaire reporting their opinion of the “athlete gender” solution (see Supplementary Digital Content 1, Questionnaire, Specifically, the concept of an “athlete gender” was presented at: 1) the 12th International Congress of the Sports Medicine Association of Greece (35); 2) a public lecture at the University of Brighton, England (36); 3) the 2018 Spring Conference of the British Association of Sport & Exercise Medicine (BASEM) (37). The BASEM presentation also was recorded ( and circulated to all members of BASEM and they were invited to complete the survey questionnaire reporting their opinion of the “athlete gender” solution. Overall, of the 154 participants who voted, 109 (71%) individuals agreed with the idea of the “athlete gender,” whereas 35 (23%) disagreed and 10 (6%) abstained from providing an opinion (Fig.A). There was the opportunity for those who disagreed with the idea of the athlete gender to provide their opinion on the best way to separate athletes into male and female categories according to five options. Of the 42 respondents (i.e., disagree and neither agree nor disagree), the most common option selected was “chromosomal pattern” (n=18, 43%), a method used and abandoned by the IAAF and IOC. The numerous cases of false-positive results, that is, originating from athletes with androgen insensitivity condition, and failing to identify, for example, women with virilizing forms of congenital adrenal hyperplasia (i.e., one of the restricted DSD in the new IAAF regulations), were the likely reasons why this approach was abandoned (30). The second highest option was “gonadal sex” (n=15, 36%), although this option has serious limitations, such as the practical difficulty of identifying one’s gender after medical or surgical reassignment (such as gonadectomy, genital surgery and hormone therapy). Few participants believed that a division is needed to be made on the basis of “legal status” (2, 5%); in contrast to the CAS ruling in the Chand case or “gender identity or self-determination” (n=3, 7%) (Fig.B).

Voting results on the idea of an "athlete gender" from three public events with a total of 154 respondents: 1) the 12th International Congress of the Sports Medicine Association of Greece; 2) public lecture at the University of Brighton; and 3) the 2018 Spring Conference of the British Association of Sport & Exercise Medicine (BASEM).

A total of 50 participants (of the 154) responded to the invitation by adding written comments. This written text was analyzed and responses were divided into six categories by two independent evaluators. Among those that agree with athlete gender (25 individuals), 8 (32%) (Fig.C) proposed the need for continuous monitoring of hormonal values (mainly T) to avoid doping manipulation. Furthermore, 6 (24%) justified their opinion as a means to include transgender athletes in the sport and 5 (20%) highlighted the need for more studies in this area (Fig.C) despite the difficulty in performing such studies, especially before and after hormone/surgery treatment. The term “inclusion” appeared frequently in the written responses, highlighting the need to ensure the rights and freedoms of all to be secured without any discrimination and in line with the Fundamental Principles of the Olympic Charter (3). 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, color, sex, sexual orientation, language, religion, political or other opinion, national or social origin, property, birth, or other status.” There also was the view that the final decision on inclusion should be made on a sport-by-sport basis (two participants) and/or after the emergence of scientific evidence (five individuals). Another view (seven comments) was that the detrimental effects on performance of hormone therapy, such as reduced muscle mass and aerobic capacity and lower levels of T, could offset any advantages maintained after the transition. In this way, transgender women with normal range T for females could compete against cis gendered women fairly. Although more studies are necessary to elucidate this question, we consider hormonal control the best way at present to ensure as fair as possible inclusion of transgender athletes in sports.

Of the respondents that disagreed with the idea of an athlete gender (17 individuals), the majority (n=8, 47%) (Fig.D) were of the opinion that biological differences between males and females are retained after therapy (e.g., enhanced muscle mass, bone density, height, size, and the “muscle memory” mechanism (38)). In addition, three (17%) participants reported that there were insufficient data to meaningfully judge any performance advantage conferred to transgender women athletes (Fig.D). Furthermore, two respondents did not accept the existence of biological differences between individuals, whereas three sought to exclude transgender athletes from sporting competitions. Notably, one respondent made reference to the underrepresentation of transgender individuals in elite sport by commenting that most transgender individuals are consumed almost entirely with their struggle to cope with the complexities of their daily lives due to their transition than to consider a career in elite sport. Finally, among the eight participants that abstained from providing an opinion, three seem not to have understood the survey question and/or the term “transgender athlete,” and three considered that more studies are necessary.

Concluding Remarks

It would be unwise for any individual or organization to assume that the DSD rules are the ultimate solution to the vexing problems presented by the existence of gender variant athletes; the legal representatives of Caster Semenya have indicated that they will challenge these rules before the CAS (39). Nevertheless, the new DSD regulations are a perfectly reasonable attempt to create legislation ensuring “fair and meaningful competition” for all women. The IOC announced the inclusion of mixed gender competitions for the 2020 Tokyo Olympics, where athletes will compete outside the dichotomy of male/female division (40). This timely development will undoubtedly serve to provide better gender equality and should pave the way to a more flexible approach in dealing with gender variant athletes. These important changes will make the Olympic Games more youthful, more urban and will result in the participation of more women in sport. A more youthful and inclusive vision of the Olympic Games could help not only the greater inclusion of women but also DSD and transgender individuals. Finally, adoption of a fair solution for all would elevate sport as a modernizing power able to oversee positive social change well beyond sporting boundaries.

The authors acknowledge the British Association of Sport and Exercise Medicine (BASEM), the Sports Medicine Association of Greece (SMAG), the University of Brighton, as well as all participants who responded to the survey and contributed to the discussion.


1. IAAF introduces new eligibility regulations for female classification| News | [cited 2018 June 21]. Available from:
2. De Villiers O. IAAF using flawed science to target Caster Semenya. IOL Sport. [Internet]. 2018 [cited 2018 June 21]. Available from:
3. International Olympic Committee. Olympic Charter. Olympic Charter. [cited 2018 June 18]. Available from: Library/OlympicOrg/General/EN-Olympic-Charter.pdf#_ga=2.17871673.1051009108.1512427396-2086914248.1494153900/.
4. New IAAF testosterone rules could slow Caster Semenya by up to seven seconds | Sport | The Guardian. [cited 2018 June 21]. Available from:
5. Amended IAAF Rules and new and updated IAAF Regulations - in force as of 1 May 2011| News | [cited 2018 June 21]. Available from:
6. Interim Arbitral Award delivered by the Court of Arbitration for Sport - CAS 2014/A/3 759 Dutee Chand v. Athletics Federation of India (AFI) & The International Association of Athletics Federation s (IAAF). Court of Arbitration for Sport. [cited 2018 June 10]. Available from: Published 2014.
7. Riley L. The participation of trans athletes in sport – a transformation in approach? Law in Sport. [cited 2018 June 21]. Available from:
8. Gleaves J, Lehrbach T. Beyond fairness: the ethics of inclusion for transgender and intersex athletes. J. Philos. Sport. 2016; 43:311–26.
9. Harper J, Martinez-Patino MJ, Pigozzi F, Pitsiladis Y. Implications of a third gender for elite sports. Curr. Sports Med. Rep. 2018; 17:42–4.
10. Bermon S. Androgens and athletic performance of elite female athletes. Curr. Opin. Endocrinol. Diabetes Obes. 2017; 24:246–51.
11. 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. doi:10.1136/bjsports-2017-097792.
12. Handelsman DJ, Gooren LJ. Hormones and sport: physiology, pharmacology and forensic science. Asian J. Androl. 2008; 10:348–50.
13. 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.
14. Bermon S, Garnier PY, Hirschberg AL, et al. Serum androgen levels in elite female athletes. J. Clin. Endocrinol. Metab. 2014; 99:4328–35.
15. Morel Y, Rey R, Teinturier C, et al. Aetiological diagnosis of male sex ambiguity: a collaborative study. Eur. J. Pediatr. 2002; 161:49–59.
16. Ellaithi M, Kamel A, Saber O. Consanguinity and disorders of sexual developments in the Sudan. Sudan J. Med. Sci. 2011; 6. doi:10.4314/sjms.v6i4.
17. Fénichel P, Paris F, Philibert P, et al. Molecular diagnosis of 5α-reductase deficiency in 4 elite young female athletes through hormonal screening for hyperandrogenism. J. Clin. Endocrinol. Metab. 2013; 98:1055–9.
18. The treatment of Caster Semenya shows athletics’ bias against women of colour | Katrina Karkazis | Opinion | The Guardian. [cited 2018 June 21]. Available from:
19. Part 2: Should our athletes move up? — Science of Running. [cited 2018 June 21]. Available from:
20. IAAF female classification rules slammed as “blatantly racist”. 2018. [cited 2018 June 21]. Available from:
21. IAAF accused of "blatant racism" over new testosterone level regulations | Sport | The Guardian. [cited 2018 June 21]. Available from:
22. T & FN Lists - Track & Field News. [cited 2018 June 21]. Available from:
23. Hanson SK. The Life of Helen Stephens: The Fulton Flash. Southern Illinois: University Press; 2004.
24. Sports gender controversies - Telegraph. [cited 2018 June 21]. Available from:
25. Walters G. Berlin Games. John Murray; 2007.
26. Ha NQ, Dworkin SL, Martínez-Patiño MJ, et al. Hurdling over sex? Sport, science, and equity. Arch. Sex. Behav. 2014; 43:1035–42.
27. Handelsman DJ. Sex differences in athletic performance emerge coinciding with the onset of male puberty. Clin. Endocrinol. (Oxf). 2017; 87:68–72.
28. International Olympic Committee. IOC Consensus Meeting on Sex Reassignment and Hyperandrogenism. November 2015. [cited 2018 June 21]. Available from:
29. Harper J. Race times for transgender athletes. J. Sport Cult. Identities. 2015; 6:1–9.
30. Cavanagh SL, Sykes H. Transsexual bodies at the Olympics: the International Olympic Committee’s policy on transsexual athletes at the 2004 Athens summer games. Body Soc. 2006; 12:75–102. doi:10.1177/1357034X06067157.
31. FIVB — Volleyball. [cited 2018 June 21]. Available from:
32. Medical Information to Support the Decisions of TUECs — Transgender Athletes | World Anti-Doping Agency. [cited 2018 June 24]. Available from:
33. Schuyler Bailar - CollegeSwimming. [cited 2018 June 24]. Available from:
34. Teetzel S. On transgendered athletes, fairness and doping: an international challenge. Sport Soc. 2006; 9:227–51.
35. 12th International Congress of Sports Medicine Society of Greece / 12th Hellenic-Cypriot Conference. [cited 2018 June 21]. Available from:
36. Beyond fairness: The biology of inclusion for transgender and intersex athletes. [cited 2018 June 21]. Available from:
37. Programme - Gender in Sport - Genetic and hormonal determinants of sex for athletes. [cited 2018 June 21]. Available from: Published 2018.
38. Egner IM, Bruusgaard JC, Eftestøl E, Gundersen K. A cellular memory mechanism aids overload hypertrophy in muscle long after an episodic exposure to anabolic steroids. J. Physiol. 2013; 591:6221–30.
39. Caster Semenya: Olympic champion will challenge “unfair” IAAF testosterone ruling - BBC Sport. [cited 2018 June 21]. Available from:
40. Tokyo 2020 event programme to see major boost for female participation, youth and urban appeal - Olympic News. [cited 2018 June 21]. Available from:

Supplemental Digital Content

Copyright © 2018 by the American College of Sports Medicine