Prior depressive-symptom burden, current deviation from personal history, and incident heart disease: a three-cohort longitudinal study

<title>Abstract</title> <p>Background Depressive symptoms are associated with cardiovascular risk, but epidemiologic analyses commonly use a single baseline or current score even when repeated assessments are available. A current score may reflect long-standing symptom burden, recent departure from personal history, or both. We examined whether separating these components provides distinct information about subsequent heart-disease risk. Methods We used person-wave data from adults aged 50 years or older in the Health and Retirement Study (HRS), the Survey of Health, Ageing and Retirement in Europe (SHARE), and the English Longitudinal Study of Ageing (ELSA). Prior burden was defined as the mean of depressive-symptom scores before the current wave. Current deviation was defined as the current score minus that prior mean. Discrete-time survival models estimated associations with survey-reported incident heart disease at the next selected wave. HRS was the primary cohort, SHARE provided an independent EURO-D replication cohort, and ELSA provided a second CES-D-8 cohort. HRS–ELSA heterogeneity was examined on the raw CES-D-8 scale, followed by reweighting of HRS toward the measured ELSA baseline covariate distribution. Results The main analyses included 7,094 HRS participants with 821 events, 26,489 SHARE participants with 2,792 events, and 2,271 ELSA participants with 419 events. In HRS, both prior burden and current deviation were associated with incident heart disease, with a larger coefficient for prior burden. SHARE showed a similar ordering using EURO-D. ELSA showed no clear association for either component. HRS–ELSA heterogeneity was present for both components, and reweighting HRS to resemble ELSA on measured baseline characteristics did not materially reduce the HRS associations. HRS supporting analyses linked life-course adversity with depressive-symptom burden and incident heart disease, whereas biomarker findings were limited. Conclusions Repeated depressive-symptom histories contained cardiovascular risk information beyond a current score alone in HRS and SHARE, with stronger associations for accumulated prior burden than for current deviation. ELSA did not reproduce this pattern. Repeated symptom measures may improve the epidemiologic characterization of psychosocial cardiovascular risk, but cross-cohort validation is needed before they are used in risk stratification or primary-prevention settings.</p>