Depression at the intersection of race/ethnicity, sex/gender, and sexual orientation in a nationally representative sample of US adults: a design-weighted intersectional MAIHDA

Abstract

This study examined how race/ethnicity, sex/gender, and sexual orientation intersect under interlocking systems of oppression to socially pattern depression among US adults. With cross-sectional data from the 2015-2020 National Survey on Drug Use and Health (n = 234 722), we conducted a design-weighted, multilevel analysis of individual heterogeneity and discriminatory accuracy (MAIHDA) under an intersectional framework to predict past-year and lifetime major depressive episodes (MDEs). With 42 intersectional groups constructed from 7 race/ethnicity, 2 sex/gender, and 3 sexual orientation categories, we estimated age-standardized prevalence and excess or reduced prevalence attributable to 2-way or higher interaction effects. Models revealed heterogeneity across groups, with prevalence ranging from 1.9% to 19.7% (past-year) and 4.5% to 36.5% (lifetime). Approximately 12.7% (past year) and 12.5% (lifetime) of total individual variance was attributable to between-group differences, indicating key relevance of intersectional groups in describing the population distribution of depression. Main effects indicated, on average, that people who were White, women, gay/lesbian, or bisexual had greater odds of MDE. Main effects explained most between-group variance. Interaction effects (past year: 10.1%; lifetime: 16.5%) indicated another source of heterogeneity around main effects average values, with some groups experiencing excess or reduced prevalence compared with main effects expectations. We extend the MAIHDA framework to calculate nationally representative estimates from complex sample survey data using design-weighted, Bayesian methods. This article is part of a Special Collection on Mental Health.

Keywords: Bayesian analysis; United States; adult; depression; gender; health surveys; intersectional framework; multilevel analysis; prevalence; probability sampling; race; sex; sexual minorities; sexual orientation.

Figure 1

Measures of discriminatory accuracy of intersectional groups by area under the receiver operator characteristic curve (AUC). Values describe the discriminatory accuracy of intersectional groups to classify cases vs. noncases in the baseline model (ie, age-adjusted model with no social position covariates) of each outcome. A random classifier (ie, random chance in classifying cases vs. noncases) is simulated by the dashed line with an AUC of 0.5, and the solid line curve represents the baseline model observed values.

Figure 2

Age-standardized prevalence of past-year major depressive episode among US adults, National Survey on Drug Use and Health (NSDUH) 2015-2020 (n = 234 374). CrI, credible interval; NAAN, Native American or American Indian; NHPI, Native Hawaiian or Pacific Islander. Estimates are design-weighted to account for the NSDUH complex sampling design and standardized to the US adult population age distribution to account for age differences across intersectional groups.

Figure 3

Age-standardized prevalence of lifetime major depressive episode among US adults, National Survey on Drug Use and Health (NSDUH) 2015-2020 (n = 234 722). CrI, credible interval; NAAN, Native American or American Indian; NHPI, Native Hawaiian or Pacific Islander. Estimates are design-weighted to account for the NSDUH complex sampling design and standardized to the US adult population age distribution to account for age differences across intersectional groups.