Are there gender-based mental health inequalities in primary care in England?

Problem

Existing evidence suggests a high prevalence of mental health conditions amongst trans and gender diverse (TGD) populations. The leading explanation for worse mental health outcomes is minority stress theory, with TGD populations experiencing structural invalidation of identity, prejudice, and discrimination. For some TGD people, gender dysphoria may also increase the risk of poor mental health. TGD patients may also face barriers to appropriate mental health care. However, poor recording of gender in medical records and most surveys limits research into gender-based health inequalities.

Approach

We worked with four LGBTQ+ and mental health charities to refine the research questions, analysis, framing, and language. We made use of changes to gender recording in the 2021 and 2022 GP Patient Survey (GPPS) to estimate gender-based inequalities in mental health in England. We focused on two GPPS outcomes: reporting a long-term mental health condition, and reporting that mental health needs were not recognised or understood at a GP appointment. We used age-adjusted logistic regression with gender (female/male/non-binary/prefer to self-describe/prefer not to say) interacted with cis/trans identity (cisgender/transgender/prefer not to say), then computed predicted probabilities for each gender group within each cis/trans group. We added groups of potential mediators as covariates, including variables capturing health, socioeconomic status, appointment factors, and healthcare professional (HCP)-patient communication.

Findings

Of 1,533,478 respondents, 2,687 (0.2%) were non-binary, 2,353 (0.2%) self-described their gender, and 9,524 (0.6%) preferred not to say. 7,994 (0.5%) respondents were transgender and 12,611 (0.8%) preferred not to say their cis/trans identity.

We found wide age-adjusted inequalities in the probability of reporting a long-term mental health condition, ranging from 8-12% amongst male and female patients who were cisgender or preferred not to say their cis/trans identity, to 48% amongst non-binary transgender patients. Probabilities tended to be higher amongst transgender groups, with the worst outcomes for non-binary trans patients. Mediation analysis suggested these inequalities may be partially explained by socioeconomic and other health inequalities, but remained mostly unexplained by these factors.

Male and female cisgender patients had the lowest age-adjusted probability (17%) of not having their mental health needs met. The probability was significantly higher amongst all other groups, reaching 28-30% amongst groups who preferred not to say their cis/trans identity and trans patients who self-described their gender. Mediation analysis suggested inequalities may be largely explained by differences in the quality of communication and relationships between HCPs and TGD patients.

Consequences

Our results provide the first national estimates of mental health inequalities for TGD groups in England. We find evidence of wide inequalities, consistent with previous targeted surveys and qualitative evidence. Results from mediation analysis suggest training primary care staff to improve HCP-patient communication and relationships could reduce inequalities in unmet mental health needs for TGD patients.