Achille et al, 2020

Summary

This is a report of preliminary results from an ongoing longitudinal study of patients aged 9-25 referred for endocrinology treatment for gender dysphoria at a single New York hospital.  Between 2013-2018, the 50 subjects each completed a set of depression and quality of life questionnaires on three occasions, at approximately six monthly intervals, with the first set of questionnaires completed prior to any endocrine intervention. Of the 50 subjects, 12 received puberty suppression, and 11 received puberty suppression and cross-sex hormones.

Results

The authors’ conclusion is that endocrine intervention “may” improve mental health in transgender youths. However, like so many in its genre, the study is of low quality:

• There is no control group.

• The sample size is small, particularly when stratified by sex (33 female subjects, 17 male).

• The follow up is short (12 months, albeit the study is ongoing).

• 45 out of 50 subjects were also receiving counselling, and about a third were on psychiatric medications. While there was an attempt to control for these factors through regression analysis, the authors note that the availability of other supports, and regular visits to a multidisciplinary medical team, may have influenced the results (p. 4).

• There is no data about the age of onset of gender dysphoria: a point relevant to current debates about the spike in adolescent females presenting for gender treatment, and a new cohort of adolescent rapid onset gender dysphoria.

In addition, the results – which are preliminary only – are somewhat mixed.

Eleven out of twelve models did not reach statistical significance (Table 4). Effect sizes were “notably larger” (p. 3) for male to female than female to male. It is interesting that this point does not raise any enquiry about the suitability of the medical pathway for adolescent females. Rather, the authors suggest it could be due to the fact that testosterone takes 6-12 months to take effect – i.e. the “answer” is “more drugs”.

In common with similar literature there is a tendency to overstate the positive effects for soundbites. For example, the “Conclusions” section states that:

“transgender children and adolescent [sic] are a high-risk population for suicide and depression. Our preliminary results show negative associations between depression scores/suicidal ideation and endocrine intervention…These results align with previous work in the Netherlands and the UK.”

The mention of suicidal ideation is misleading in circumstances where a study found no significant changes in suicidal ideation. Five out of fifty subjects were reportedly suicidal at baseline and three subjects suicidal at the end of 12 months (Table 3). Regression models for suicidal ideation were not estimable on this tiny sample (p. 3).

The study is too small and the intervention protocol is too varied to draw conclusions.

Tags