1 - (SYM 50) Under the Same Umbrella: Differences in Depression, Anxiety, and Stress Experiences Between Sexual and Gender Minority Groups

Trans and non-binary (TNB) people have a higher prevalence of depression and anxiety (Su et al., 2016) compared to lesbian, gay, and bisexual individuals (LGB). This may be due to greater TNB exposure to minority stress (Testa et al., 2015). However, it is not known whether the relationships between depression, anxiety, and stress in TNB individuals are like those observed in LGB individuals. Lindley and Bauerband (2022) found support for measurement invariance (configural, metric, scalar) between TNB and LGB samples on the Depression, Anxiety, Stress Scale -21 (DASS-21; Lovibond & Lovibond, 1996). Our previous work has identified an additional well-fitting 2-factor DASS-21 structure in TNB participants (Micanovic, in preparation), which collapsed the Anxiety/Stress factors. Here we report tests of measurement invariance using traditional and alternative, TNB-validated DASS-21 structures to examine the phenomenology of depression, anxiety, and stress in TNB people and across TNB and LGB groups. 

We used DASS-21 data from an online cross-sectional study of U.S. LGBTQ+ adults with 271 TNB participants (M_age = 32.88) and 530 LGB participants (M_age = 28.15). For the DASS-21 3-factor solution, all model fits (CFI > .90, RMSEA < .70) were good, and chi-squared difference tests offered evidence for scalar and configural invariance. There was no evidence for metric invariance across TNB vs. LGB groups (CFI = .910, RMSEA = .075, p < .05). Specifically, the latent stress factor for TNB participants was more strongly characterized by stress items (trembling [original DASS-21 item 07], close to panic [15], heartbeat awareness [19], scared without reason [20]) and one anxiety item (can’t wind down [01]). The anxiety latent factor for LGB participants was more strongly characterized by feeling intolerant [14].  With regards to the novel, TNB-informed, 2-factor solution, all model fits (CFI > .80, RMSEA = .077-.085) were adequate. Similarly, there was evidence for scalar/configural invariance, but not metric invariance (CFI = .87, RMSEA = .084). For TNB participants, the latent stress/anxiety factor was again more strongly characterized by stress items (close to panic [15], heartbeat awareness [19], scared without reason [20]). Together, these results provide evidence that in TNB people, stress and anxiety are more bound up with physiological symptoms, relative to LGB people. Relevance to the minority stress model and clinical interventions with TNB clients presenting with mood/anxiety symptoms will be discussed.