Authors' reply

between the brains of females and males’. But a lack of reliable neurobiological pathology is true of most psychiatric disorders, for which psychiatrists routinely prescribe drugs and other physical treatments, quite often coercively. Again, ‘As a pure subjective experience, [gender identity] may be overwhelming and powerful but is also unverifiable and unfalsifiable’. Psychiatric disorders are mainly diagnosed on the basis of what patients report about their ‘subjective experience’, so the requirement that transgender patients must provide substantial additional ‘verification’ of their experiences also suggests that the authors have adopted a double standard. Do they propose that patients with depression or post-traumatic stress disorder demonstrate that their problems are ‘falsifiable’ before they can receive treatment? The authors attempt to distance themselves from ‘conversion therapy’, but many gender dysphoric patients will not find their arguments convincing. They claim ‘there is little evidence’ that transgender conversion therapy ‘is taking place in the UK’, but the 2018 National LGBT Survey found that 13% of UK trans respondents ‘had been offered’ conversion therapy, compared with 7% of ‘cisgender’ respondents. Conversion therapy for homosexuality is closely associated with psychoanalysis. The American Psychiatric Association removed homosexuality from its list of disorders in 1973, with strong opposition from psychoanalysts. It took nearly three decades for the London-based International Psychoanalytical Association (IPA) to act similarly, in 2002. It seems likely that the IPA continues to tolerate the view that homosexuality is a disorder, treatable by psychoanalysis. The authors allude to ‘complex intrapsychic conflicts’ but fail to explain what they mean by this or provide a reference. This suggests an undeclared allegiance to psychoanalysis. Some British psychoanalysts appear to see transgender patients as a growth opportunity. The gender critical views of the ex-Tavistock London psychoanalyst Marcus Evans have been quoted by the BBC and the BMJ, while journalists have failed to scrutinise his implied claim that psychoanalysis can provide valid clinical opinion. Such scrutiny is especially necessary given the problematic relation of psychoanalysis to homosexuality and its wider history of evading scrutiny of its claims to therapeutic efficacy, validity and safety. Psychiatry should retain its gatekeeping role for transgender patients seeking physical treatments or legal gender change, but extreme gender critical views, which at present appear to include special pleading for psychoanalysis, would undermine the consent necessary for that role to be effective.

We thank Dr MacFarlane for his response 1 and welcome his comments about various providers and pharmaceutical agencies as well as freedom of speech.
We did not review the neuroscience of brain sex differences but draw interested readers' attention to a recent review 2 and accessible analyses of available research. 3,4 We agree that the art of psychiatric diagnosis depends on the clinician accepting the truth of the patient's own experience. In the absence of objective diagnostic tests, believing and trusting the patient's own subjective narrative is central to the doctorpatient relationship. However, this starts to lose coherence when the doctor must readjust their own understanding of material reality in order to accommodate another's subjective belief. Declaring that 'Despite having the body of a man, I am in all other respects a woman' supposes some inherent essence of gender that many would reject. Reorganising psychiatry to give primacy of gender identity over sex risks breaching the necessary boundaries that exist to maintain the safety and dignity of individuals, groups of people and society more generally.
Without a clear definition of conversion therapy, it is not possible to know the extent of the practice in the UK. Proponents of affirmative care have argued that conversion therapy is anything that might act as a barrier to medical transition. 5,6 It would follow that attempts to assess and treat coexistent mental illness, or even the process of making an accurate initial diagnosis of gender dysphoria, could be described as conversion therapy rather than the basic standard of clinical care that would apply for any other presentation. 7 The authors remain opposed to any treatment model designed to coercively alter the sexual orientation of bisexuals, gay men or lesbians. It is crucial to distinguish between sexual orientation and gender identity when the latter comes with an expectation for complex, irreversible medical interventions, described as 'affirmative'. If sustained, long-term benefits of medical and surgical transition could be clearly and independently demonstrated, it would be appropriate to offer these interventions early, but the evidence is not convincing. 8 Therefore it is reasonable to exercise therapeutic caution, especially in light of growing concerns about complications and regret, 9 particularly in younger patients.
Given government moves to criminalise conversion therapy in medical settings, the nature of 'barriers to treatment' must be clearly described. 9 New laws will need detailed supplementary guidance for the benefit of patients, doctors and the criminal justice system. We propose that organisations representing clinicians should help legislators make explicit that neutrally framed therapeutic or exploratory work is not conversion therapy, irrespective of how an individual ultimately feels about their own identity.
In the absence of evidence-based guidelines underpinned by solid research, we cannot make recommendations about treatment pathways, and do not advocate one particular model over any other. We draw readers' attention to the unexplained increase in referral numbers, the higher numbers of children and young people seeking interventions, and the shift in sex ratio, 10 as such demographic changes are significant and deserving of research and explanation. Doctors should 'first do no harm'. The bar for informed consent to life-changing, irreversible medical and surgical interventions is necessarily high. 9 Enhanced service provision and new care pathways should be informed by robust research in this patient group.

Freedom to think should not mean freedom from evidence and experience
Evans made a number of inaccurate statements in his originally published article, which subsequently resulted in the publication of a corrigendum. Some of the inaccurate statements and implications for scholarship and clinical practice are referred to in our letter, which was first submitted on 23 August 2020. Gender identity is a pertinent and timely topic, given the current moral panic around transgender individuals. Evans 1 states several fallacies that risk increasing the stigma towards transgender and non-binary individuals. Throughout, motivated reasoning that seeks to describe transgender individuals as inherently disordered is apparent. This manifests in the very limited selection of evidence cited and in unsupported claims made contrary to the bulk of existing evidence.
It is also important to put the letter in context. The consensus among the World Psychiatric Association, 2 Royal College of Psychiatrists, American Academy of Child and Adolescent Psychiatry and American Psychiatric Association is that psychological treatments to 'suppress or revert gender diverse behaviours are unscientific and unethical'. The reports from these groups follow detailed review of the current literature.
Evans relies heavily on personal accounts and experiences reported by media organisations. Although each individual story deserves hearing, understanding and respect, there is an inherent danger to highlighting cherry-picked examples from blog posts or newspapers. These sources often have political standpoints or biases. The quoted Christian Institute, for instance, writes on their website that transgender ideology 'seeks to completely destroy the distinction between men and women that God in his wisdom has created', and the Sunday edition of the quoted Times has been forced to correct a number of inaccuracies in articles about the Tavistock after intervention from the press regulator.
The existence of a few examples of adverse outcomes does not lend support to the idea that affirming care is harmful. In any area of medicine or psychiatry there are always patients who regret treatment, but provided informed consent is obtained and the proportion remains low, that risk is deemed acceptable. What little legitimate research into detransition exists