Scientific oversight falling between responsible institutions should worry us all
Following Cohen and Barnes’ investigative article,(1) it is positive that the BMJ is hosting an open medical debate about the effects of puberty blockers on young people with gender dysphoria. We agree that some of the problems identified were relatively minor misdemeanours that many researchers in constrained circumstances could commit,(2) but contend that the troubles of this GIDS service study(3) are symptomatic of a wider collective failure to ensure good science is applied to this important area. A more recent examination of the evidence has independently determined that it is still ‘experimental’.(4) Key institutions need to work together effectively in order to determine whether, and when, we should prescribe puberty blockers, or cross sex hormones, to children and young people identifying as transgender. Current patients cannot give fully informed consent in the absence of precise estimates of benefits and harms that could be life-long, except in research trials that admit the uncertainties.
The ethics of research conduct belongs to the Health Research Authority (HRA); the quality of science is primarily the responsibility of funders and sponsors (here UCL). However, the two are rarely completely unrelated. The HRA investigation concluded that although its oversight via annual reports is lax, other procedural issues were not serious.(5) We disagree with the conclusion that problems around the recording of declarations of interest were not important. More critically, the HRA report did not try to determine if its first committee (which ‘pushed back’ about the science) was right, preferring only to defend the applicants’ resubmission without addressing concerns, and its second committee’s decision to allow the small evaluation to proceed without considering, or testing feasibility for, a randomised trial. Effectively the HRA absolves itself from responsibility for judging the science in deliberations about ethics. Thus the avoidance of bad science (which is inherently unethical) relies entirely upon the sponsor’s usually ‘light-touch’ oversight (a problem also found with the AgeX trial).(6) In the absence of independent peer review during competitive funding, Universities must rely on the Principal Investigator. In the event of the inevitable occasional error, this weakness exposes individual employees to potential criticism.
Whilst many paediatricians prescribe ‘off-label’, halting natural puberty in gender dysphoria is entirely unrelated to previous indications. Prescribers, and psychologists who don’t prescribe, should not suggest that safety data can be transferred from triptorelin used for restoring correct timing in precocious puberty. Here, its use has many far reaching consequences so a more stringent research design was required. By gaining permission to study individuals without controls, and by not studying the whole pathway, we were never likely to learn much about the benefits or harms. Empiricism is simply not a good enough approach. Whilst RCTs might be difficult, they are not impossible, and simply have not been tried. Back in 2010, an opportunity for a UK pilot feasibility trial was missed when puberty blockers were not routine. It remains an option to only prescribe within the context of a trial, maybe robustly comparing an ‘earlier’ vs ‘later’ timing, with or without ‘earlier’ or ‘later’ cross-sex hormones or supportive psychological therapy. It is also important that the changing demographic of those identifying as transgender is taken into account. This new cohort of mainly natal females, often with more recent onset, and with high rates of trauma and developmental disorder may respond differently.
A further troubling aspect was the lack of justification for moving from a small, uncontrolled cohort with unpromising, unpublished results to the scaling up and widespread implementation of care which has not been evaluated. While GIDS has traditionally been more conservative about offering medical interventions than the US, Holland and Canada, following a trend is inadequate reason for the NHS to commission such expansion. The new, much larger NIHR funded study is again designed only to observe, not test, effects,(7) similar to a US study of different populations (i.e. comparing ‘apples’ with ‘pears’).(8) The absence of the development and evaluation of psycho-social therapy for the many individuals assessed by GIDS as suitable for hormonal therapy but with co-existing mental health and developmental problems is a further matter deserving investigation.
We agree it is troubling that clinicians are fearful about engaging in this area. (2) We need generalists and specialists, researchers, methodologists, public health experts, parents and individuals with lived experience (of desisting and detransitioning as well as those feeling the benefit) to work together in constructive disputation. We conclude that the real significance of the BMJ report(1) lies in illuminating part of the greater global issue of how commissioners, service providers and research institutions have failed to create a prospective framework with adequate controls to evaluate the impact of interventions in this vulnerable group who deserve equitable standards of research. The UK has the potential to create the best service for gender questioning young people in the world. To do so, key institutions and leaders will need both to embrace equipoise and to work together effectively - a modus operandi which has transformed the care of children with cancer.
1. Cohen D, Barnes H. Gender dysphoria in children: puberty blockers study draws further criticism. BMJ 2019; 366 doi: https://doi.org/10.1136/bmj.l5647
2. Stott S, Ford T. Puberty blockers study: much ado about nothing? https://www.bmj.com/content/366/bmj.l5647/rr-1 (10 October 2019)
3. http://gids.nhs.uk/our-early-intervention-study [last accessed 20 October 2019]
4. Henegan C, Jefferson T. Gender-affirming hormone in children and adolescents
5. https://www.hra.nhs.uk/about-us/governance/feedback-raising-concerns/inv... [last accessed 20 October 2019]
6. Bewley S, Payne M, Blennerhassett M. Analysis. Cost of extending the NHS breast screening age range in England. Br Med J 2019;364:l1293
7. Longitudinal outcomes of gender identity in children. ISRCTN98564473
8. Olson-Kennedy J, Chan YM, Garofalo R, Spack N, Chen D, Clark L, Ehrensaft D, Hidalgo M, Tishelman A, Rosenthal S. Impact of Early Medical Treatment for Transgender Youth: Protocol for the Longitudinal, Observational Trans Youth Care Study. JMIR Res Protoc. 2019 Jul 9;8(7):e14434. doi: 10.2196/14434.
Competing interests: No competing interests