Inequities in access to mental health care at intersections of race/ethnicity, gender identity, and gender modality: An application of Multilevel Analysis of Individual Heterogeneity and Discriminatory Accuracy (MAIHDA)

Transgender populations, and particularly transgender women of color, are burdened by high rates of both mental illness and healthcare avoidance. Despite this, research examining the patterns of non-use of mental healthcare within this community is lacking, and has further suffered due to the difficulty of collecting large, population-based samples of transgender adults in general, and transgender people of color in particular. We therefore use data from the Census Bureau’s new Household Pulse Study in conjunction with the novel Multilevel Analysis of Individual Heterogeneity and Discriminatory Accuracy (MAIHDA) method to assess the prevalence of unmet mental healthcare needs at critical intersections of gender modality, gender identity, and race/ethnicity. Participants were 699,843 U.S. adults who completed the Household Pulse Survey between July 2021 to May 2022. Participants were sorted into intersectional strata by race and ethnicity (Asian, Black, White, Hispanic, and mixed/another race), gender modality and sexuality (cisgender heterosexual, cisgender LGB+, and transgender) and gender identity (dichotomized here as masculine vs feminine). We then fit a multilevel log-binomial regression model with strata and participant-level random intercepts to assess the main effects, and calculated stratum-level residuals as a measure of intersectional effects. Unmet mental health needs were more prevalent among transgender people (PR=2.63, 95% CI= 2.22, 3.11), cisgender LGB+ people (PR=2.25, 95% CI=1.93, 2.63), women (PR=1.36, 95% CI= 1.25, 1.48), and people who self-identified as Black (PR=1.56, 95% CI=1.31, 1.86), Hispanic (PR=1.81, 95% CI=1.51, 2.15) mixed/another race (PR=2.11, 95% CI=1.75, 2.53). The most negative intersectional effects were for transfeminine people and for non-Black cisgender heterosexual men. The largest positive intersectional effects were for Black cisgender women and for transmasculine people. These findings suggest that although racism, transphobia, and misogyny might each independently increase the prevalence of unmet mental health needs, race and gender modality additionally impact the degree to which various groups are impacted by gender norms when seeking mental healthcare. Intersectional effects are thus essential to evaluate when studying mental healthcare use and non-use among transgender people of color.