Socioeconomic gradients and functional vulnerability in late-life depressive symptoms: design-consistent, age-standardized evidence from HRS 2022
Description
Background Survey-design–consistent estimates of late-life depressive symptoms vary across studies due to heterogeneous thresholds, incomplete handling of complex sampling, and shifting age structures. Methods Using HRS 2022 public data, we analyzed adults aged ≥60 (non-proxy interviews). Depressive symptoms were defined by CES-D-8≥4 (primary), with sensitivity at ≥3 and ≥5. Prevalence and 95% CIs were estimated with full survey design preservation (weights/strata/PSUs; Taylor linearization). We characterized distributions by sex, education (4 levels), weighted income quintiles (Q1–Q5, with explicit “income missing”), ADL/IADL limitation, and multimorbidity (≥2). Direct age standardization aligned 2020 to the 2022 age structure. Additive interaction (IADL × multimorbidity) was summarized via RERI, AP, and S. Results In 2022, design-consistent prevalence was 12.62% (men 9.92%, women 14.87%). Clear socioeconomic gradients were observed (higher in <High School and income Q1; lowest in Q5; “NA income” remained elevated). ADL/IADL limitation and multimorbidity each marked higher burden. On the additive scale, the IADL × multimorbidity interaction was positive but statistically imprecise (RERI≈0.78, AP≈0.17, S≈1.27). After aligning age structures, the 2020–2022 difference was small (≈−0.9 percentage points) with wide CI overlap. Findings were robust to threshold choice and to weighted vs unweighted income-split constructions. Conclusions We provide a design-consistent, age-standardized national baseline for late-life depressive symptoms with socially patterned inequalities. Results support prioritizing women, lower-SES strata, and older adults with IADL limitation and multimorbidity for integrated screening and care, and motivate longitudinal tracking of equity-relevant indicators.