Depressive symptom trajectories around chronic disease burden transitions among middle-aged and older adults in China: a population-based longitudinal study

Chronic disease burden and depressive symptoms are closely linked, but it remains unclear whether depressive symptom trajectories differ around the first observed transition from no chronic disease to ≥ 1 chronic disease and progression from a single condition to multimorbidity. This study examined these temporal patterns among middle-aged and older adults in China. We analyzed four waves (2011–2018) of the nationally representative China Health and Retirement Longitudinal Study. Adults aged 45 years or older with at least two valid depressive symptom measurements were classified into an incident chronic disease cohort (0→≥1; n = 1,335, including 1,200 stable and 135 transition participants) and a progression-to-multimorbidity cohort (1→≥2; n = 2,357, including 2,091 stable and 266 transition participants). Depressive symptoms were assessed using the 10-item Center for Epidemiologic Studies Depression Scale. Overlap weighting balanced baseline covariates, and weighted segmented linear mixed-effects models estimated baseline differences, pre-transition trends, level differences at the first observed transition wave, and subsequent slope changes. In both cohorts, transition participants had higher baseline depressive symptom scores than stable participants, whereas pre-transition trends did not differ significantly. In the 0→≥1 cohort, no significant level difference was observed at the first observed transition wave (β = 0.255, 95% CI [− 0.378, 0.888], p = 0.430). The primary analysis suggested a steeper subsequent trajectory (β = 0.338, 95% CI [0.011, 0.665], p = 0.043), although this estimate should be considered exploratory because of limited post-transition follow-up. In the 1→≥2 cohort, CESD-10 scores were higher at the first observed transition wave (β = 1.046, 95% CI [0.556, 1.536], p < 0.001), with a lower post-transition slope (β = −0.216, 95% CI [− 0.359, − 0.074], p = 0.003). Sensitivity analyses supported the higher CESD-10 level at the 1→≥2 transition, whereas post-transition slope estimates were less stable. Progression to multimorbidity was associated with higher depressive symptom levels around the first observed transition wave. The suggested steeper trajectory after the 0→≥1 transition remains exploratory because post-transition follow-up was limited.