When a Second Disease Strikes: Depression Jumps as Chronic Illness Piles Up in Aging China

A diagnosis of a first chronic illness can be devastating, but a sweeping new longitudinal study from China suggests that the sharpest psychological blow may come later — at the moment a single condition becomes two. Analyzing four waves of nationally representative data from more than 3,600 middle-aged and older adults, researchers tracked depressive symptoms as people crossed two distinct thresholds: the shift from no chronic disease to at least one, and the progression from a single disease to full multimorbidity. Their findings, published in BMC Psychology, paint a nuanced picture in which the timing and magnitude of mental health change depend critically on which transition a person experiences.

The research team, led by Weihang Xu and Mengying Wang of Jilin University alongside colleagues, drew on the China Health and Retirement Longitudinal Study, known as CHARLS, a population-based survey that has followed nationally representative samples of Chinese adults since 2011. The investigators focused on participants aged 45 and older who had at least two valid measurements of depressive symptoms across the 2011 to 2018 survey waves. From this pool they constructed two cohorts designed to isolate the psychological consequences of specific disease-burden transitions rather than of chronic illness in general.

The first cohort captured the transition from zero chronic diseases to at least one, comprising 1,335 participants, of whom 1,200 remained stable without disease and 135 experienced the transition during the observation window. The second, larger cohort captured progression from a single chronic condition to multimorbidity, defined as two or more concurrent diseases, and included 2,357 participants, with 2,091 remaining stable and 266 making the transition. Depressive symptoms were measured at every wave using the 10-item Center for Epidemiologic Studies Depression Scale, a widely validated screening instrument that asks respondents how often they experienced feelings such as disturbed sleep, loneliness, and hopelessness during the past week.

Methodologically, the study faced a familiar challenge in observational research: people who develop chronic diseases differ systematically from those who do not, in age, body mass index, functional status, socioeconomic position, and baseline mental health. To address this, the researchers used overlap weighting, a propensity-score-based technique that reweights the stable and transition groups so their baseline covariates are balanced, giving greater weight to observations in the region where the groups overlap most. This approach is particularly effective at reducing the influence of extreme cases and produces weights with favorable variance properties compared with traditional matching or inverse-probability weighting.

On top of the weighting scheme, the team fitted segmented linear mixed-effects models, a class of statistical models that allow the trajectory of depressive symptoms to change direction at a specified point — here, the first observed wave at which a participant’s disease burden transitioned. The models estimated four distinct quantities for each cohort: a baseline difference in symptom scores between transition and stable participants, the pre-transition trend in symptoms, a level difference at the transition wave itself, and a change in slope after the transition. This segmented structure is what allowed the researchers to ask not merely whether depression differs between groups, but when in the disease course the divergence appears.

The results for the first transition — from no disease to at least one — were strikingly subtle. People who would eventually develop a chronic condition already carried higher baseline depressive symptom scores than those who remained disease-free, yet their pre-transition trajectories did not diverge, meaning symptoms were not climbing faster in the years before diagnosis. At the transition wave itself, the level difference was not statistically significant, with an estimated effect of 0.255 points on the CESD-10 scale and a 95 percent confidence interval spanning from minus 0.378 to 0.888. The primary analysis did suggest a steeper subsequent trajectory after the first diagnosis, with a slope change of 0.338 points per wave, but the authors caution that this estimate should be treated as exploratory because post-transition follow-up was limited.

The picture changed dramatically for the second transition. When participants progressed from one chronic disease to multimorbidity, their CESD-10 scores were significantly higher at the transition wave, with a level difference of 1.046 points and a confidence interval from 0.556 to 1.536 — an effect far too large to be explained by chance. Intriguingly, the post-transition slope was actually lower among those who had just crossed into multimorbidity, at minus 0.216 points per wave, suggesting that symptoms spiked at the transition and then partially stabilized or improved in relative terms. Sensitivity analyses consistently supported the elevated symptom level at the multimorbidity transition, although the slope estimates proved less stable across model specifications.

These findings carry substantial implications for how clinicians and policymakers think about mental health in aging populations. The elevated baseline scores among future transition participants hint at a selection effect: psychological distress may both precede and follow physical illness, consistent with bidirectional models in which depression increases the risk of chronic disease through behavioral and physiological pathways, while disease burden in turn erodes mental health. The absence of a significant pre-transition divergence, however, suggests that the psychological cost of accumulating a second condition is concentrated at the transition itself rather than building gradually beforehand — a pattern that could inform the timing of screening and intervention.

China provides a particularly important setting for this work. The country is aging rapidly, and the prevalence of multimorbidity among its middle-aged and older adults is rising as lifestyles change and life expectancy increases. CHARLS, administered with ethics approval from Peking University’s Biomedical Ethics Committee, offers one of the few nationally representative longitudinal windows onto this process, and the CESD-10 allows depressive symptoms to be tracked consistently across waves. The study’s funding from the National Natural Science Foundation of China and the Jilin Province Science and Technology Development Plan reflects growing institutional attention to the intersection of chronic disease and mental health in the world’s most populous aging society.

The authors are careful about the limits of their evidence. The suggested acceleration of symptoms after a first diagnosis rests on limited post-transition follow-up, and the lower post-transition slope in the multimorbidity cohort, while statistically significant in the primary model, was less robust in sensitivity analyses. Even so, the central message stands: the transition to multimorbidity marks a measurable psychological inflection point. As health systems worldwide grapple with aging populations, the study argues for treating the second diagnosis not as a routine clinical event but as a moment of heightened vulnerability, when mental health support may matter as much as the management of the diseases themselves.

Subject of Research: Depressive symptom trajectories around chronic disease and multimorbidity transitions in middle-aged and older Chinese adults

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

Article References: Depressive symptom trajectories around chronic disease burden transitions among middle-aged and older adults in China: a population-based longitudinal study. (n.d.). https://doi.org/10.1186/s40359-026-05783-w

Image Credits: AI Generated

DOI: 10.1186/s40359-026-05783-w

Keywords: depression, multimorbidity, chronic disease, CHARLS, aging, CESD-10, longitudinal study, China, mental health, depressive symptoms, older adults, epidemiology