Structural
Intersecting socio-structural determinants of depression among men who have sex with men in the United States: A multilevel analysis of individual heterogeneity and discriminatory accuracy (MAIHDA) Jason M. Lo Hog Tian* Jason Lo Hog Tian Lo Hog Tian Lo Hog Tian Lo Hog Tian Lo Hog Tian Lo Hog Tian Lo Hog Tian Lo Hog Tian Lo Hog Tian Department of Epidemiology, Bloomberg School of Public Health, Johns Hopkins University
Depression remains a driver of poor health and wellbeing among cisgender gay, bisexual, and other men who have sex with men (MSM). However, the way that socio-structural determinants interact to shape depression is currently unknown in this population. This study used multilevel analysis of individual heterogeneity and discriminatory accuracy (MAIHDA) to quantify variation in depression risk across strata defined by key socio-structural determinants.
We analyzed cross-sectional data from 15,362 cisgender MSM aged 15+ from across the United States as part of the American Men’s Internet Survey (AMIS) collected between 2021 – 2024. Strata were defined by socio-structural determinants: age, race, region, past-year homelessness, and past-year illicit drug use. A PHQ-9 score ≥10 was used as a cutoff for significant depression symptoms. MAIHDA was used to quantify joint and individual contributions of intersecting socio-structural determinants on predicted probability of depression.
Overall, 20% (n=3,080) of participants had significant depression. In the null model, 13% of total variance was attributable to between-stratum differences; however, this was reduced by 99.6% after inclusion of the socio-structural variables, indicating that differences in depression risk were primarily due to additive effects. Homelessness accounted for the largest percentage of stratum-level variance (50%), followed by age (45%), drug use (24%), region (4%) and race (2%). Predicted depression prevalence ranged from under 10% among older, stably housed individuals without drug use to over 65% among younger individuals (<25 years old) experiencing homelessness and drug use.
There was substantial heterogeneity in predicted depression with the greatest probability characterized by people experiencing homelessness, younger individuals, and people who use drugs. Interventions prioritizing housing stability and substance use services for young MSM are likely to yield the greatest impact.
