This is a highly significant study showing that the population of teenagers referred for gender dysphoria is changing. It is noteworthy that this is happening in two different countries.
The sex ratio is changing:
The sex ratio of teenagers seeking help for gender dysphoria has changed at two clinics, one in Canada and one in the Netherlands.
Before 2006, more male than female teenagers sought transition at these clinics. Since 2006, they have seen more female teenagers than male teenagers.
Sex ratio for teenage patients at the Canadian Gender Identity Service:
- 1999-2005 – 68% male, 32% female
- 2006-2013 – 36% male, 64% female
Sex ratio for teenage patients at the Dutch Center for Expertise on Gender Dysphoria:
- 1989-2005 – 59% male, 41% female
- 2006-2013 – 37% male, 63% female
At the Canadian clinic, there was no change in the sex ratio of teenagers referred for psychiatric issues.* In both time periods, roughly two-thirds of their other patients were male.
According to the authors, “In adult samples [of transitioners], in almost all cases, the number of natal males either exceeds the number of natal females or the sex ratio is near parity.” Poland and Japan are exceptions; in those countries more females transition than males.
In addition, clinics for children with gender dysphoria have found that the number of males exceeds the number of females.
More teenagers are transitioning:
The number of teens of both sexes has increased over time, although the increase is larger for the female teenagers.
Increases at the Canadian clinic:
- 46 in 30 years (1976-2005)
- 129 in 8 years (2006-2013)
- 80 in 30 years (1976-2005)
- 73 in 8 years (2006-2013)
Increases at the Dutch clinic:
- 77 in 17 years (1989-2005)
- 148 in 8 years (2006-2013)
- 109 in 17 years (1989-2005)
- 86 in 8 years (2006-2013)
In other words, the Canadian clinic saw nearly nearly three times as many female teens in the past 8 years as they had seen in the previous thirty. The Dutch clinic saw nearly twice as many female teens in the past 8 years as they had seen in the previous seventeen.
Furthermore, “For many years in the Toronto clinic, the number of adolescent referrals was quite low. Between 1976 and 2003, for example, no more than five adolescents of one biological sex were assessed in a calendar year and, during this period, the number of males exceeded the number of females. Beginning in 2004, however, the number of adolescent referrals began to rise quite dramatically, which appears to be consistent with the observations of clinicians and researchers from other gender identity clinics.”
For earlier data on the increase in Canada, see this article.
Sexual orientation percentages have changed:
The Canadian clinic also looked at sexual orientation.
Sexual orientation of females:
- 1976-2005 – 89% primarily attracted to females; 11% other
- 2006-2013 – 64% primarily attracted to females; 36% other
Other could mean primarily attracted to males, bisexual, or asexual.
Sexual orientation of males:
- 1976-2005 – 67% primarily attracted to males, 33% other
- 2006-2013 – 44% primarily attracted to males, 56% other
Other could mean primarily attracted to females, bisexual, or asexual.
To put it another way, in the past most of the teenagers would have been gay if they weren’t transgender. If they transitioned, they would live their lives as straight people.
In 2006-2013 most of the male teenagers would have been straight, bisexual, or asexual if they weren’t transgender. If they transition, some of them will live their lives as lesbians.
One-third of the female teenagers in 2006-2013 would have been straight, bisexual, or asexual if they weren’t transgender. If they transition, some of them will live their lives as gay men.
What’s going on?
Why are we seeing more teenagers seeking help for gender dysphoria?
Why is the increase greater among female teens than males?
And why are we seeing a shift in the sexual orientation of these teens? Was it harder in the past to come out as transgender if you were seen as straight? Or is this a group of people who were less likely to have gender dysphoria in the past?
Has something changed in our environment that increases the number of people with gender dysphoria? What would affect more females than males? Why would it affect teenagers more than children (see this earlier article)? How would it fit with the changing percentages related to sexual orientation?
Is it just that there were always this many teenagers with gender dysphoria and now they are able to get care at an earlier age? How does that theory fit with the change in the sex ratio of teens applying to the clinic? with changes in their sexual orientation?
Clearly, we need more research to sort out these questions.
The authors speculate about possible explanations for the change in the sex ratio at their clinics.
They suggest that the general increase in patients might be due to a combination of destigmatization and more awareness of the biomedical treatments available to teens. However, they point out that this does not explain why more females would apply for treatment.
I don’t think we can know why the number of patients has increased without further research – research which is desperately needed.
The increase in the number of female patients at the Toronto clinic was not caused by a change in the severity of cases; they found that there was no significant relationship between severity of dysphoria and year assessed.
However, for male teens in Toronto, there was a weak correlation between severity of dysphoria and year assessed. “More recently assessed cases had moderately higher GD severity.” This only explained 6.7% of the variance. Therefore “it is unlikely that the recent inversion in the sex ratio can be accounted for by a substantive change in severity variation.”
On the other hand, they only have data on the severity of dysphoria starting in 2001 and the number of cases began increasing in 2004.
The change in the sex ratio was not due to females entering puberty at an earlier age; both clinics found no significant difference for the mean ages when females and males came to the clinic.
The sex ratio did not change due to the shift in sexual orientation. A logistical regression analysis did not find evidence for a sex x sexual orientation interaction.**
The authors suggest that perhaps the explanation for the change in the sex ratio is that it is harder for males to transition to a female role than for females to transition to a male one.
I find this unconvincing as this would have been true in the past when more male teenagers than females applied to their clinic. Nor would this hypothesis explain the shift in sexual orientation.
Here is their full explanation:
“Given that there is at least some overlap in the gender-variant developmental histories of early-onset individuals with GD and some gay men and lesbians, it might, therefore, be asked whether or not degree of stigmatization for gender-variant behavior might account for the recent inversion in the sex ratio of GD adolescents. It is well-known that cross-gender behavior in children is subject to more social stigma (e.g., peer rejection and peer teasing) in males than in females, in both clinic-referred adolescents with GD and in the general population[26–30]. Thus, it could be argued that it is easier for adolescent females to “come out” as transgendered than it is for adolescent males to come out as transgendered because masculine behavior is subject to less social sanction than feminine behavior. Some support for this was found in Shiffman’s  study of peer relations in adolescents with GD, in which adolescent males with GD reported more “social bullying” than adolescent females with GD. Given that a transgendered identity as an “identity option” has become much more visible over the past decade, it is conceivable, therefore, that such an identity option is easier for females to declare than it is for males because it does not elicit as much of a negative response. Thus, it could be argued that it is this sex difference in degree of stigmatization that accounts for the inversion in the sex ratio that we have identified in the two studies reported here. In other words, there are greater costs for a male to adopt a female gender identity in adolescence than it is for a female to adopt a male gender identity.”
A few more details about this study:
The first study looked at 328 teens (13-19) who were referred to the Toronto clinic between 1976 and 2013. The mean age at the time of referral was 16.66 years with no difference between the ages of males and females.
All of the teens met criteria for Gender Identity Disorder or Gender Identity Disorder Not Otherwise Specified. They were diagnosed using criteria in the relevant version of the DSM – this changed over time. The assessment of severity of dysphoria began in 2001.
The control group was 6,592 teens referred to their general clinic for psychiatric issues between 1999-2013. Eleven teens originally referred for psychiatric issues who were later referred to the Gender Identity Service were not included in this group.
The teens’ sexual orientation was determined by either clinical chart data or measurements on the Erotic Response and Orientation Scale and the Sexual History Questionnaire. This data was not available for five probands (aka people in this study).
The numbers for the sexual orientation of the teens at the Canadian clinic were:
1976-2005 (30 years)
- 52 males primarily attracted to males
- 26 males in the “other” category
- 39 females primarily attracted to females
- 5 in the “other” category
2006-2013 (8 years)
- 32 males primarily attracted to males
- 41 males in the “other” category
- 82 females primarily attracted to females
- 46 females in the “other” category
The clinic did not have data on the sexual orientation of five of the teenagers.
The second study looked at data on 420 teenagers (13 and up) referred to the Dutch clinic between 1989-2013. Their mean age at the time of assessment was 16.14 and there was no significant age difference between males and females.
The second study did not include data on sexual orientation or a control group for comparison.
“The percentage of female adolescents from Amsterdam in the first time period did not differ significantly from the percentage of female adolescents from the Toronto clinic, and the percentage of female adolescents from Amsterdam in the second time period also did not differ from the percentage of female adolescents from the Toronto clinic, both χ2(1) < 1.”
This study is a follow-up to two earlier letters to the editor about changes in the teenage population at the clinic in Toronto: Is Gender Identity Disorder in Adolescents Coming out of the Closet? and Patterns of Referral to a Gender Identity Service for Children and Adolescents (1976–2011): Age, Sex Ratio, and Sexual Orientation.
The first letter discussed a rise in teenagers referred to the Canadian clinic between 2004-2007. The second letter discussed the continued increase in referrals from 2008-2011 and raises the question of a possible change in the sex ratio in 2008-2011.
Evidence for an Altered Sex Ratio in Clinic-Referred Adolescents with Gender Dysphoria by Aitken M1, Steensma TD, Blanchard R, VanderLaan DP, Wood H, Fuentes A, Spegg C, Wasserman L, Ames M, Fitzsimmons CL, Leef JH, Lishak V, Reim E, Takagi A, Vinik J, Wreford J, Cohen-Kettenis PT, de Vries AL, Kreukels BP, Zucker KJ in J Sex Med. 2015 Mar;12(3):756-63. doi: 10.1111/jsm.12817. Epub 2015 Jan 22.
* The Canadian clinic is the Gender Identity Service, within the Child, Youth, and Family Services (CYFS) at the Centre for Addiction and Mental Health in Toronto. The clinic in the Netherlands is the Center of Expertise on Gender Dysphoria at the VU University Medical Center in Amsterdam. This may explain why we have a comparison group for the Canadian patients with gender dysphoria, but not the Dutch ones.
** “In the cohort examined in Study 1, perhaps it could be argued that, in the first time period, the greater number of biological males than biological females was an artifact of there being two prominent subtypes of GD (androphilic and nonandrophilic) in the former, whereas the latter were predominantly of only one subtype (gynephilic), but that this shifted in the second time period, with a greater number of females with a nongynephilic sexual orientation. However, the logistic regression analysis shown in Table 4 did not provide evidence for a sex × sexual orientation interaction. It only showed that a nonandrophilic or nongynephilic sexual orientation increased the odds that a proband presented in the second time period, but sexual orientation did not interact with probands’ biological sex.”