LETTERS TO THE EDITOR The regulations about eligibility for women with hyperandrogenism to compete in women 0 s category are well founded. A rebuttal to the conclusions by Healy et al. Healy et al. 1 recently published an article 1 in Clinical Endocri- nology, reporting on body composition and endocrine profiles of elite athletes, 454 men and 239 women. Based on their find- ings, they severely criticize the recent Regulations by interna- tional sports organizations such as IAAF and IOC regarding eligibility of hyperandrogenic women to take part in competi- tions against other women athletes. 2 Their basic argument is that they find a ‘complete’ overlap between testosterone blood levels in men and women; therefore, testosterone is not a valid param- eter to be used in decisions on eligibility. Instead, it is argued that the difference in performance could be attributed to differ- ences in lean body mass. The article contains many misunderstandings of the IAAF/ IOC Regulations, as well as erroneous conclusions: • The authors conclude that ‘the IOC definition of a woman as one that has a normal testosterone level is untenable’. This is a false interpretation of the Regulations. On the contrary, the IOC and the IAAF have strongly stressed that the sex of an individual must never be questioned. Sex identity cannot be determined by any biological test. The Regulations focus on women with a combination of signs of hyperandrogenism (a male physique) and very high testosterone levels in blood. A high level (in the male range) alone is never a reason for non- eligibility. • Healy et al. find an overlap between testosterone levels in the male and female athletes, which in their opinion proves that there is no clear-cut separation between the testosterone levels in men and women. The authors do not mention the publica- tions that, under resting conditions, demonstrate no overlap in testosterone levels between healthy men and women. 3 While the IAAF/IOC Regulations state that blood samples must be drawn under resting conditions, Healy et al. obtained blood within two hours after completing national or international competitions. Others have shown that following strenuous exercise, testoster- one levels in men decrease, while in women, T is unchanged or moderately increased. 4 Much of the overlap seen in the cohort described by Healy et al. is due to some men with exceptionally low testosterone blood levels following exercise and some women with very high levels. Therefore, the testosterone levels reported by Healy et al. cannot be used as baseline data for male and female athletes at rest • A key statement in the Regulations is that for a woman with physical signs of hyperandrogenism, testosterone levels must not be in the (resting) male range. This is defined as above approximately 10 nmol/l (IAAF Regulations, dependent on the assay used). Healy et al. find 12 women with testosterone levels above 84 nmol/l, the lower (sedentary) male limit in their assay and argue that this proves that the Regulations are obsolete, because as much as 5% of their female athletes reach above this value. However, they do not present any follow-up of these women. They do not report if anyone showed signs of hyperan- drogenism, which in the Regulations is the only reason to mea- sure testosterone (apart from antidoping controls). Healy et al. believe that none of these 12 athletes were doped with testoster- one, because they declared themselves to be ‘clean’. Unfortu- nately, very few doped athletes would answer otherwise to this question. • The authors do not use state-of-the-art methods for tes- tosterone assay (tandem MSP), but rather immunoassay that in general have inherent specificity and sensitivity problems. Some of the high testosterone levels found in these 12 women might be explained by assay problems, but the majority is probably due to either hyperandrogenic disorders of sex development (DSD) or doping with testosterone. Some of them might have androgen insensitivity; such individuals would not show signs of hyperandrogenism and would not have any known undue benefits in sports. They would likely be deemed eligible to compete. As the authors do not present any follow-up, the cause of hyperandrogenism in these 12 women remains specu- lative. • The authors find that male athletes have a larger lean body mass (bone and muscles) and suggest that this, rather than testosterone levels, is the reason for males being superior to women in most sports. However, they do not discuss the reasons for this difference in muscle mass; the difference in testosterone levels is one obvious explanation. In androgen- sensitive individuals, testosterone directly stimulates increase in muscle mass, bone tissue and red blood cells, and androgens also influence behavioural patterns, such as drive to compete. All these effects of androgens are beneficial for physical performance. • The authors have not referred to the recent publication by Bermon et al. 5 This is interesting reading for anyone interested in this area and answers many of the questions raised by Healy et al. We hope that the erroneous conclusions made by Healy et al. will not in future be cited as evidence against the use of the Reg- ulations on eligibility of women with hyperandrogenism to com- pete in the women 0 s category. The Regulations were thoroughly discussed in several international fora before being accepted, with participation of both medical experts, representatives for women elite athletes as well as one DSD organization. Martin Ritz en*, Arne Ljungqvist†, Richard Budgett†, Pierre-Yves Garnier‡, Stephane Bermon§, Angelica Lind en- Hirschberg¶, Eric Vilain** and Maria Jos e Mart  ınez-Pati~ no†† Clinical Endocrinology (2015) 82, 307–311 Ó 2014 The Authors. Clinical Endocrinology published by John Wiley & Sons Ltd. 307 This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.