Purpose: Mental illness (MI) is a substantial and under-recognized burden among older adults in the United States. Despite this, limited research has examined how specific behavioral risk factors, including binge drinking, prescription misuse, and psychological distress, are associated with MI among adults aged 65 and older. Grounded in the stress-vulnerability model and the life course perspective, this study examines these associations using nationally representative data.
Methodology: This study analyzed national survey data to investigate the relationship between substance use, distress, and the presence of MI among individuals aged 65 and older. Using data from the 2023 National Survey on Drug Use and Health (NSDUH), the analysis was restricted to adults aged 65 and older (N = 5,060) and estimated four logistic regression models on a common specification: a complete-case weighted model, a primary Taylor-linearized survey design model (strata VESTR_C, primary sampling unit (PSU) VEREP, 50 design degrees of freedom (df)), a multiple imputation by chained equations model ((MICE), m = 20, Rubin’s rules), and a Firth penalized model, adjusting for sex, race and ethnicity, education, income, and poverty. The primary outcome was any MI past year (AMIPY): a co-occurring MI and substance use disorder (SUD) indicator was examined as a secondary outcome. Predictors were past- 30-day binge drinking, past-year prescription misuse, and the full validated six-item Kessler (K6) distress scale.
Findings: The weighted prevalence of MI was 11.68% (95% confidence intervals (CI) 10.36 to 13.01), with past-30-day binge drinking at 12.04% and past-year prescription misuse at 2.79%. In the primary design-based model, psychological distress and prescription misuse were significantly associated with MI, while binge drinking was not once the outcome was restricted to MI alone; its association appeared only against the co-occurring outcome, indicating it was carried by the SUD component. Distress had the strongest association: each one-standard-deviation (SD) increase on the full K6 distress scale corresponded to roughly 5.6 times the odds of any MI (per-SD odds ratio (OR) = 5.65, 95% CI 4.73 to 6.75). Because a one-point increment is not comparable across the three- and six-item scales, the per-SD OR is reported as the primary effect size. Because the study design is cross-sectional, causal inferences cannot be drawn; however, the associations identified suggest that targeted screening for substance misuse and psychological distress among older adults may support earlier identification of psychiatric risk and inform integrated behavioural health interventions.
Originality: This study contributes to the literature by jointly examining substance use, psychological distress, and any MI in a nationally representative sample of older adults, an analysis that has received limited empirical attention. The theoretical grounding in the stress-vulnerability model and life course perspective further distinguishes this work from predominantly descriptive analyses of older adult substance use. The central contribution is population-specific: although the links among substance use, distress, and MI are established in general adult samples, they have rarely been examined jointly in a nationally representative sample of adults aged 65 and older, and prior older-adult work has been largely descriptive or limited in scope.

PDF (Download Full text)