The study arrives against the backdrop of what the researchers describe as deep population aging in China, where depressive symptoms among rural older adults have become a prominent challenge for grassroots elderly health governance. Persistent urban–rural disparities in symptom prevalence have been reported, and rural older adults face constraints that their urban counterparts largely do not: limited access to medical resources, thin formal social support, and the emotional absence that can follow when younger generations migrate to cities for work. Although rural elders accumulate a certain amount of daily productive physical activity through farming and household labor, their participation in structured, regular exercise remains generally low, making this population a distinctive test case for the exercise–mood relationship.

To recruit the sample, the team combined multistage stratified random sampling with quota sampling. Drawing on microdata from China’s seventh National Population Census, they stratified the country into eastern, central and western regions, randomly selected three provinces per region—Shandong, Zhejiang and Fujian in the east; Henan, Hubei and Anhui in the center; and Sichuan, Shaanxi and Gansu in the west—and then randomly sampled counties tiered by per-capita GDP, townships and administrative villages. The final field sample spanned 108 villages in 54 townships across 27 counties. Within each village, eligible residents aged 60 to 69 were drawn from household-registration rosters, with quotas set for the 60–62, 63–65 and 66–69 age groups cross-classified by sex to mirror the census composition of the rural older population.

Of 900 individuals invited to participate, 872 completed the questionnaire and 100 were excluded after eligibility screening, leaving 772 participants in the final analysis—85.78 percent of those invited. The sample was nearly balanced by sex, with 405 women and 367 men, and evenly distributed across regions. Participants with severe cognitive or mental disorders, acute critical illness, advanced cancer or severe cardiovascular or cerebrovascular disease were excluded, as were those who had regularly used anxiolytic or antidepressant medication, received formal psychological counseling, or taken part in standardized long-term exercise interventions in the preceding three months. Questionnaires were administered face to face during home visits by trained investigators, and responses flagged as rushed, internally contradictory or excessively uniform were screened out.

Four standardized instruments anchored the measurements. Physical exercise was captured with the Chinese version of the Physical Activity Rating Scale (PARS-3), which combines exercise intensity, frequency and duration into a continuous volume score from 0 to 100. Depressive symptoms were assessed with the 10-item Center for Epidemiologic Studies Depression Scale (CES-D 10), scored from 0 to 30, with the authors careful to note that this measures symptom severity rather than serving as a clinical diagnosis of depression. Health anxiety was measured with the 18-item Short Health Anxiety Inventory, which probes worry about illness likelihood and fear of illness consequences, and general self-efficacy with the 10-item General Self-Efficacy Scale. All four scales showed acceptable internal consistency in this sample, with Cronbach’s alpha values ranging from 0.790 to 0.851.

The headline result was a significant negative cross-sectional correlation between physical exercise and depressive symptoms, with a total association estimate of −0.0793 (95 percent confidence interval: −0.0992 to −0.0595). When the researchers decomposed this association, the direct link between exercise and symptoms—estimated at −0.0356—accounted for 44.89 percent of the total, while the combined indirect association through health anxiety and self-efficacy accounted for the remaining 55.11 percent. In the adjusted regression model, which controlled for age, sex and region, physical exercise and self-efficacy were each inversely associated with depressive symptom scores, whereas health anxiety was positively associated. The model explained a modest share of the variance, with an adjusted R-squared of 0.166.

The most striking pattern concerned health anxiety, defined in the literature as the tendency to interpret bodily sensations or changes as signs of serious illness. The specific indirect association running through health anxiety was the largest of the three pathways, estimated at −0.0266 and accounting for 33.54 percent of the total association. The theoretical logic is that regular exercise may coincide with slower age-related physical decline, better self-reported health and fewer somatic complaints, which in turn may accompany lower subjective health worry—and that persistent health anxiety, by trapping individuals in repetitive worry and excessive attention to bodily signals, may co-occur with the low mood and diminished interest characteristic of depressive symptoms. The authors point out that rural older adults, with comparatively limited health literacy and poorer access to primary care, may be especially prone to negative attributions of bodily discomfort.

Self-efficacy—the belief in one’s ability to accomplish tasks and cope with adversity—formed a second, smaller pathway, with a specific indirect association of −0.0115, or 14.50 percent of the total. Drawing on Bandura’s Social Cognitive Theory, the authors suggest that mastery experiences from improved physical function and achieved exercise goals, along with informal social support generated in group exercise settings, may coincide with stronger perceived coping capacity, which in turn is associated with problem-focused coping and fewer depressive symptoms. A third, serial pathway—running from exercise through lower health anxiety to higher self-efficacy and finally to fewer symptoms—was the smallest, at −0.0057, or 7.07 percent of the total. All three indirect associations were statistically significant by bootstrap analysis with 5,000 resamples, with confidence intervals excluding zero.

The authors are explicit about what these numbers cannot show. Because all variables were measured at the same time, the ordering of health anxiety before self-efficacy is a model assumption rather than a demonstrated temporal sequence, and the analyses identify statistical indirect associations, not causal mediation. Reverse causation is plausible: depressive symptoms may themselves be associated with less exercise, greater health anxiety and lower self-efficacy. Residual confounding by socioeconomic circumstances, physical health and social support cannot be ruled out, particularly since only age, sex and region were available for adjustment. All core variables were self-reported in a single interview, inviting recall, social-desirability and common-method bias, although Harman’s single-factor test suggested the first unrotated component accounted for only 20.26 percent of variance. The authors also note that the robustness of some significance estimates to village-level clustering remains uncertain due to incomplete reporting of the clustering procedure for the indirect-association confidence intervals.

Generalizability is likewise bounded. The findings apply to rural Chinese adults aged 60 to 69 and may be affected by the participation and exclusion criteria, and the use of an overall exercise score leaves open whether purposeful exercise differs from productive agricultural labor in its association with mood. The serial pathway’s statistical significance, the authors caution, should not be equated with clinical importance, and no clinical thresholds were evaluated. They call for multi-wave longitudinal studies measuring all four constructs repeatedly, and for randomized exercise trials with repeated assessments of health anxiety, self-efficacy and depressive symptoms, to test whether exercise changes these psychological variables and whether psychological changes precede symptom changes. Until such evidence arrives, the study stands as a detailed cross-sectional map of how exercise, health worry and coping confidence travel together in a rural aging population—one that flags health anxiety as the most prominent node in the observed network of correlations, while stopping well short of prescribing exercise as a treatment for depression.

Subject of Research: Cross-sectional associations between physical exercise, health anxiety, self-efficacy and depressive symptoms among rural older adults in China

Article Title: Physical exercise and depressive symptoms among rural older adults in China: a cross-sectional study of serial correlational linkages via health anxiety and self-efficacy

Article References: Jiao, J., & Chen, X. (2026). Physical exercise and depressive symptoms among rural older adults in China: a cross-sectional study of serial correlational linkages via health anxiety and self-efficacy. Frontiers in Psychology, 17, Article 1940756. https://doi.org/10.3389/fpsyg.2026.1940756

Image Credits: AI Generated

DOI: 10.3389/fpsyg.2026.1940756

Keywords: physical exercise, depressive symptoms, health anxiety, self-efficacy, rural older adults, China, cross-sectional study, population aging, mental health, mediation analysis, CES-D 10, psychology of aging