How doctors think about their role in transgender care: a qualitative study of UK general practitioners and endocrinologists

Recruitment

We recruited trainee and qualified GPs and endocrinologists with past, present or anticipated care of transgender patients. Institutional Review Board (IRB) approval was granted (BSREC.156/20-21). We sent invitations to participate to organisations representing both specialties, and a smaller organisation representing practitioners of gender medicine. We also utilised our own social networks as medical students, and asked senior colleagues to approach established doctors in the field, using an approved advertisement. Participants were then purposively snowballed, which we deemed necessary due to the relatively small number of endocrinologists. Participants were not compensated for involvement. Informed consent was obtained electronically. We recruited 16 participants, all cisgender (we asked participants to define their gender identity at the beginning of the interview), six of whom identified as female and 10 as male (Table 1).

OPEN IN VIEWER

Table 1.

Participant demographic data.

	Age	Role	Centre	Region	Gender	Years of experience
Dr A	30s	GP Registrar	Primary Care	Large city	Female	<5
Dr B	60s	Consultant Endocrinologist	District General Hospital	Small city	Male	40+
Dr C	70s	Consultant Endocrinologist	University Teaching Hospital	Large city	Male	40+
Dr D	40s	Specialist Registrar in Endocrinology	University Teaching Hospital	Large city	Female	5–9
Dr E	30s	GP locum	Primary Care	Peripatetic	Male	5–9
Dr F	30s	Specialist Registrar in Endocrinology	University Teaching Hospital	Small city	Male	5–9
Dr G	50s	GP Partner	Primary Care	Rural	Male	20–29
Dr H	40s	GP Partner	Primary Care	Large city	Male	10–19
Dr I	60s	Consultant Endocrinologist	District General Hospital	Town	Male	40+
Dr J	60s	GP Partner	Primary Care	Large city	Female	30–39
Dr K	50s	GP Principal	Primary Care	Large city	Female	20–29
Dr L	40s	GP Partner	Primary Care	Large town	Male	10–19
Dr M	60s	Consultant Paediatric Endocrinologist	University Teaching Hospital	Large city	Male	30+
Dr N	50s	Consultant Endocrinologist	University Teaching Hospital	Small city	Male	30+
Dr O	40s	GP Partner	Primary Care	Small town	Female	20–29
Dr P	30s	Specialist Registrar in Endocrinology	University Teaching Hospital	Large city	Female	5–9

Sample size was guided by ‘information power’: the number of participants needed to reach ‘data sufficiency’ (rather than ‘saturation’). ²¹ Instead of attempting to determine the hypothetical sample size at which no new codes would emerge (saturation), an ‘information power’ approach holds that the sample size should be in inverse proportion to the quality of information held by the sample population. A more specialised field, with experienced practitioners, and a well-developed existing theoretical literature requires fewer interviewees than an investigation of an uncharacterised broad field.^21,22 In this study, nine participants reported over 20 years of experience, with their interviews offering copious material. In line with the tenets of reflexive thematic analysis, this does not necessarily mean that all possible responses have been exhausted, but rather that the ‘selected patterns’ offer enough material to extend and challenge our existing theoretical understanding of this topic.^21–23

Interview design

The interview schedule (Supplementary Appendix SA1) was initially formulated around themes from existing literature focusing on the clinical experience of transgender care, and theories of sex and gender. It was then iteratively redesigned to adapt to emerging foci. The interviews lasted from 25–60 minutes and were conducted over Microsoft Teams by the two first authors in 2021–2022, who were medical students at the time.

Data analysis

We followed the guidelines for reflexive thematic analysis as developed by Braun and Clarke. ²³ Interview transcripts were auto-generated by the video call software. Each interview was conducted by one of the two first authors; the other verified and anonymised the transcript. Transcripts were then re-read by both authors, for familiarisation with the dataset, and then inductively coded independently using NVivo for Mac. Major themes were generated by each author and then discussed to encourage validity and representativeness. Finally, themes were mapped against concerns from the existing literature. We chose themes using two criteria: (1) representativeness to the dataset, and (2) ability to contribute to prescient issues in transgender health research. Seven themes were identified, but for brevity, we have focused here on three that we believe are the most exigent (Table 2).

OPEN IN VIEWER

Table 2.

List of themes identified from the data, explaining the rationale behind selecting the three themes (starred) described in this report.

Themes identified from the data	Discussion
The psychological and technical responsibilities of the doctor*	A previously unexplored area of research that elaborates how generalists and specialists understand their complementary roles in transgender care.
Prescribing across the primary / secondary care divide*	Reported in every interview and repeatedly described as the most significant barrier to care.
Need for training to improve communication and knowledge*	Offers practical suggestions of how focused training could improve care.
Challenges of treating the non-binary patient in a binary system	–
Patient-centred vs doctor-led care: who is guiding treatment?	–
Problems emerging from limited access to care	–
Interior markers of sex and gender as more important than external appearance	–

‘I know my language isn’t adequate’: training needs in transgender medicine

Many doctors felt that they had not received enough training in transgender medicine. Our respondents used alternative sources of information: academic research, expert patients, YouTube videos and younger family members. Dr G reported that he would ‘ask my children what it means’. Dr L noted that ‘I’ve got kids and they’re very, very open about this stuff’. This reinforced the sense of transgender care as a generational issue amid changing social norms. Our respondents deferred their understanding of transgender issues to their patients and wanted their use of terminology to be patient-led (‘common courtesy’ according to Dr G). While this approach shows less political dogmatism than some studies report, it risks turning patients into ‘reluctant educators’. ²⁴

There were three clusters of thought about what training might improve transgender care. First, many doctors wanted a general level of training to be offered to all staff. This would include measures to reduce unconscious bias, present correct terminology (including pronoun usage) and dealing with inherently gendered situations. For example, binary gendered wards lead to transgender patients being placed in side rooms away from their specialty, which evocatively renders them ‘outliers’ in hospital nomenclature. Second, some endocrinologists advocated general medical training to support transgender patients with physical comorbidities. Dr F had ‘heard experiences of people who went to see a cardiologist and they said, “‘I'm transgender’” and the cardiologist said, “I've got no idea what you're talking about. Please go home and book another appointment with someone else”’. Dr M imagined a ‘trans man with abdominal pain’:

How do you sensitively take a history and make a diagnosis in a situation like that, you know, with respecting the person’s gender identity? … Clearly something like pregnancy is a differential diagnosis, for instance. All these things are challenging, and all doctors need to know how to deal with that.

Third, some GPs requested training on peer support and organisations to which patients could be signposted. A few GPs thought that patient perspectives and case studies would provide more exposure to transgender voices, while those who anticipated being asked to continue private prescriptions requested more pharmacological training. Although two experienced GPs were happy with the current provision of training, reporting that they saw relatively few transgender patients, most doctors were keen for further learning opportunities. The endocrinologists-in-training described the difficulty of gaining clinical experience in gender clinics ‘with enough frequency to understand what’s happening’ (Dr D). Many of our respondents, echoing studies from North America, thought lack of training was compromising care. ²⁶

‘Second violin, really’: specialty roles in transgender care

Every doctor interviewed described the importance of a multi-disciplinary team in managing gender transitions, and few respondents saw themselves as leaders. Two main roles emerged from the interviews, which we labelled ‘technicians’ and ‘supporters’. An endocrinologist described his role as looking after the patients’ ‘best interests in hormonal terms. In that sense, I’m almost a technician’ (Dr C). Another endocrinologist described his position as ‘second violin’, with mental health teams leading the orchestra (Dr M). This had implications for their theorisation of their patients’ gender roles. One endocrinologist questioned whether he was involved in gender ‘transition’: ‘transition … implies to me that they’re in the process of changing, whereas most of those I’ve seen have been fairly established in their gender roles’ (Dr C). He saw his role as the biochemical confirmation of a pre-existing, settled, gender decision.

By contrast, many GPs figured their roles as ‘supporters’ and signposters: they saw patients at an earlier stage of gender incongruence, where they may not have a full understanding of their own gender identity. Dr J, who regularly saw transgender patients, stated that she ‘luckily’ does not have to make ‘the decision’ of where a patient’s gender identity should be treated to, so she could be ‘guided by them’. Dr K saw this as intrinsic to the field: ‘it doesn’t feel odd [for GPs] to be guided by a patient’s narrative’. When GPs were asked to maintain hormone regimens, they found this to be ultra vires. One GP described this in trenchant terms: ‘You wouldn't ask a plumber to deal with your electrics, or you would be very stupid … and the plumber would be very foolish too’ (Dr H). This division of labour – into emotional and technical work – reflects Isabel Menzies’ thesis that the division of the patient into multiple body systems defends against clinical anxiety. ²⁷ Rather than any individual doctor taking on the overwhelming responsibility of remodelling someone’s sex, the clinicians helping patients consider their gender identity sequestered their role from those implementing the change.

Psychiatry had a particularly fraught role. Some scholars have called for the removal of gender identity treatment from the purview of mental healthcare.^6,8 However, no doctor we interviewed was willing to treat without the input of a mental health professional. This was not because our respondents saw gender incongruence as a psychiatric disorder. Instead, doctors emphasised the difficulty of dealing with the psychological consequences of biographical disruption. Dr E, a GP who had served many transgender patients, thought psychiatric services were needed ‘from start to finish’ to ‘help with stigma’ post-transition. Dr K, a GP, reiterated she did not ‘feel that the person has a pathology, just that [psychological services should be] available to have support’. Dr K explicitly located the pathology in social stigma and suggested that psychological support would not be necessary if society ‘was kinder and more open’. One endocrinologist thought psychiatrists earned their place because they were ‘more trained’ in being ‘less judgemental’ than other specialties (Dr I).

Other interviewees proposed that psychiatrists were needed as transgender people might be at greater risk of psychopathology. Some attached this to the ‘psychological comorbidities that go with trying to figure out what gender you are’ (Dr F). It was suggested that transgender people may have histories of abuse or family breakdown: one GP noted that many transgender patients had changed their surnames, which she took as a sign of disrupted familial relations. She imagined that changing one’s sex involves ‘incredibly drastic’ changes to the body and new ways of interacting socially,^28,29 and she noted the ‘courage it takes to … and the distress that must have been underlying to want to go down that route’ (Dr J). It is noteworthy that although many transgender patients acknowledge co-existing psychological pressures from gender variance, many studies have reported a suspicion of mental health involvement, such as feeling pressure to conceal distress to present as a ‘perfect patient’. ²⁵ In one study, a transgender man described avoiding going ‘to the GP for medication for mental health [as it might] risk the referral [not] being put through’. ³⁰ A concern about underlying psychological pathology slows down care by mandating psychological assessment prior to treatment. The relationship between comorbid psychological distress and gender diversity, if it exists, is an argument for swifter treatment. Many doctors interviewed noticed the remission of mental health concerns after medical affirmation therapy.

‘How am I gonna get my Zoladex?’: prescribing hormone therapy

Many GPs were concerned about hormone therapy prescription practices, especially in the transition between primary and secondary care. The reasons included gender clinics expecting primary care doctors to absorb the prescription costs, GPs lacking experience in prescribing these medications off-licence, differences between primary and specialist care formularies and GPs’ mistrust of assessments performed at private clinics. GPs were cautious about the legal responsibility of prescribing medication for unfamiliar uses: ‘Even though I might have a professor of gender identity medicine giving me advice, if it's ultimately my prescription … then I am liable’ (Dr H). The lack of best practice guidance compounded this anxiety. Dr E was concerned that testosterone therapy could increase risk of cardiac issues. There is ‘nothing to say how to treat people … I looked it all up that night and there’s very little data on cholesterol levels in … men taking oestrogen, women taking testosterone’.

Some interviewees thought that any personal discomfort on the part of the GP should not affect prescribing practices: ‘It's a bit like the GPs who used to refuse to prescribe contraceptive pills … You were obliged to signpost the patient to someone who would’ (Dr H). More formal signposting might counteract the phenomenon, noted by many interviewees, of transgender patients resorting to word-of-mouth to find a willing doctor, creating clusters of demand in certain surgeries. One GP’s surgery had formulated internal guidelines for continuing hormones to encourage consistency between the partners.

Doctors who were willing to prescribe emphasised the ‘sense of relief’ (Dr B) that their patients expressed:

I will tell them right at the onset that I’d be perfectly happy to give you the female hormone, the male hormone and supervise you, but I would like to know a lot more about – a bit more about – your background as it were. So I put them at ease, because usually what happens is that they have to do a presentation to me, as it were, that's what they feel, about ‘why I want to be male’ or ‘why I want to be female’ but I say, right, OK, you've made your mind up. That's fine, I can supervise you now. (Dr B, endocrinologist)

Many researchers have discussed the pressures on transgender patients to present a coherent narrative during medical consultations, which may ride roughshod over the idiosyncrasies of their gender identity.^14,15 Our respondents showed how pressures in service provision, such as delays in prescribing, may contribute to this. Doctors should be aware that patients’ anxieties about the provision of their hormone therapy may lead to these self-misrepresentations.

Limitations

This study offers a novel account of cisgender clinicians (although two described themselves as gender-fluid) with varying degrees of interest and experience in transgender care. We were unable to recruit any gender minority clinicians, despite advertising through LGBT medical networks, as the pool of endocrinologists, in particular, is limited. Transgender practitioners would be likely to have particularly rich insights, and would be able to account for the effect of clinical pressures on those seeking and providing care. Future work that explores these experiences might help bridge the binary between the provider and the patient.

Our participants were all prepared to treat transgender patients. We note that some doctors who were invited refused due to lack of interest in, or suspicion of, the field. Many spoke to us of colleagues who used delaying tactics or poor practice to avoid treating transgender patients, but none admitted to doing this themselves: a study designed to elicit doctors’ reasons for refusing to treat transgender patients could make a crucial contribution. This study also predominantly recruited clinicians working in urban centres, although did have respondents working in rural general practice. Transgender patients living in rural areas report facing more difficulty transferring their care to different doctors or hospitals. ³⁵ Further research targeting rural doctors, who see fewer transgender patients but may not be able to use referral to opt-out of transgender care, would offer insights into improving services for patients with the fewest options.

While this study helps characterise doctors’ understandings of working with ‘transgender patients’, it is important to note that this is a heterogenous community, including non-binary, gender variant and gender fluid people, some of whom will never seek medical treatment. Fluidity is particularly central in children, for whom gender identity often evolves with other aspects of social development. While any non-cisgender patients can face misunderstanding, misgendering and hostility, the constituent groups may face different challenges, depending on the type of medical care they seek. Our endocrinologists mostly reported feeling comfortable with the hormonal technicalities of treating non-binary adults, but they found the process not adequately accounted for by existing guidelines or hospital structure. This research elucidates the difficulties for clinicians of managing both the psychological and pharmacological aspects of gender transition. Further research that helps improve care for children questioning their gender identity, and for non-binary and gender fluid people of all ages, some of whom may want improved care without needing hormonal treatment, emerged as particularly urgent.