目的 基于中国健康与养老追踪调查（CHARLS）数据，系统探讨中老年人群中抑郁症状与心血管疾病（CVD）发病风险的关联路径与机制，明确睡眠障碍与炎症因子的中介作用，并识别高风险亚群。

方法 本研究利用2011—2020年CHARLS全国性纵向数据，纳入基线无CVD与精神疾病史、且数据完整的45岁及以上人群（n=7,119）。通过Cox比例风险模型评估抑郁症状（CES-D-10 ≥10分）对CVD发病风险的影响；采用偏差校正Bootstrap法检验炎症（hs-CRP）与睡眠障碍的中介效应；通过亚组分析（性别、年龄、婚姻、户口）与敏感性分析（排除认知障碍、慢性疼痛）验证结果稳健性；并采用限制性立方样条分析抑郁与CVD之间的剂量反应关系。

结果 在7,119名基线无CVD及精神疾病的中老年样本中，9年随访期间共计517人新发CVD，累计发病率为7.26%。抑郁症状（CES-D-10≥10分）检出率为33.83%。Cox多因素模型（模型5）显示，抑郁症状显著增加CVD发病风险（HR=1.312，95% CI: 1.090–1.579，P=0.004），该效应在调整人口学特征、健康行为、生理健康和社会支持因素后依然稳健。中介效应分析显示，睡眠障碍在抑郁与CVD之间具有显著中介作用（间接效应=0.003，P=0.023），中介效应占总效应的15.79%；而炎症指标（hs-CRP）无显著中介效应（P=0.976）。亚组分析结果提示，抑郁对CVD风险的影响在以下群体中更显著：女性（HR=1.323，95% CI: 1.040–1.683，P=0.023）、45–64岁人群（HR=1.276，95% CI: 1.035–1.573，P=0.022）、农业户口（HR=1.344，95% CI: 1.102–1.639，P=0.003）、以及已婚人群（HR=1.342，95% CI: 1.102–1.634，P=0.003）。敏感性分析在排除基线认知障碍及慢性疼痛样本后，结论依旧显著（HR范围：1.296–1.404，P<0.05）。限制性立方样条分析显示，抑郁症状总分与CVD风险呈线性递增趋势，未发现明显平台期或反向转折，提示剂量效应持续存在。

结论 抑郁症状是中老年人群CVD发病的独立危险因素，其效应部分通过睡眠障碍介导。建议整合心理干预与心血管健康管理，重点关注高风险亚群。

Objective: This study aims to systematically investigate the association between depressive symptoms and the risk of cardiovascular disease (CVD) among middle-aged and older adults in China, based on data from the China Health and Retirement Longitudinal Study (CHARLS). Particular attention is given to exploring the mediating roles of sleep disturbances and inflammatory markers, as well as identifying high-risk subpopulations.

Methods: Using nationally representative longitudinal data from the CHARLS (2011–2020), a total of 7,119 participants aged 45 years and above, with no baseline history of CVD or mental illness and complete data, were included. The impact of depressive symptoms (CES-D-10 score ≥10) on CVD incidence was assessed using Cox proportional hazards models. The mediating effects of sleep disturbances and inflammation (measured by high-sensitivity C-reactive protein, hs-CRP) were examined using bias-corrected bootstrap methods. Subgroup analyses (by sex, age, marital status, and household registration type) and sensitivity analyses (excluding participants with cognitive impairment or chronic pain) were conducted to assess the robustness of the results. Additionally, a restricted cubic spline analysis was performed to evaluate the dose–response relationship between depressive symptoms and CVD risk.

Results: Among the 7,119 participants free from CVD and mental illness at baseline, 517 developed CVD during the 9-year follow-up period, resulting in a cumulative incidence of 7.26%. The prevalence of depressive symptoms (CES-D-10 ≥10) was 33.83%. Multivariate Cox regression (Model 5) showed that depressive symptoms significantly increased the risk of developing CVD (HR = 1.312, 95% CI: 1.090–1.579, P = 0.004), and this association remained robust after adjusting for demographic characteristics, health behaviors, physiological health, and social support. Mediation analysis revealed a significant indirect effect of sleep disturbances in the pathway between depressive symptoms and CVD (indirect effect = 0.003, P = 0.023), accounting for 15.79% of the total effect. No significant mediating effect was found for hs-CRP (P = 0.976). Subgroup analyses indicated that the association was more pronounced among women (HR = 1.323, 95% CI: 1.040–1.683), individuals aged 45–64 years (HR = 1.276, 95% CI: 1.035–1.573), those with agricultural household registration (HR = 1.344, 95% CI: 1.102–1.639), and married individuals (HR = 1.342, 95% CI: 1.102–1.634). Sensitivity analyses further confirmed the robustness of the findings (HR range: 1.296–1.404, P < 0.05) after excluding participants with cognitive impairment and chronic pain. The restricted cubic spline analysis demonstrated a linear dose–response relationship between depressive symptom severity and CVD risk, with no evidence of a plateau or reversal, indicating a persistent dose-dependent effect.

Conclusion: Depressive symptoms are an independent risk factor for CVD incidence among middle-aged and older adults, with part of the effect mediated by sleep disturbances. These findings highlight the need to integrate psychological interventions into cardiovascular disease prevention strategies, with targeted attention to high-risk subpopulations.