Title: Depression Risk Among Chinese Middle-Aged and Older Adults Based on CHARLS 2020 Survey Data.
ShuWang1
YoshihisaShirayama1✉Emailshirayam@juntendo.ac.jp
AhmadIshtiaq1
MiyokoOkamoto1
MotoyukiYuasa1
1Department of Global Health Research, Graduate School of MedicineJuntendo UniversityTokyoJapan
Author: Shu Wang1, Yoshihisa Shirayama1, Ahmad Ishtiaq1, Miyoko Okamoto1, Motoyuki Yuasa1.
Affiliations:
¹ Department of Global Health Research, Graduate School of Medicine, Juntendo University, Tokyo Japan
Corresponding author:
Yoshihisa Shirayama, Email: [shirayam@juntendo.ac.jp]
Depression Risk Among Chinese Middle-Aged and Older Adults Based on CHARLS 2020 Survey Data.
Author: Shu Wang, Yoshihisa Shirayama, Ahmad Ishtiaq, Miyoko Okamoto, Motoyuki Yuasa.
Abstract
Background:
Depression has emerged as a major public health concern among middle-aged and older adults in China. While prior research has focused on demographic and socioeconomic factors, the impact of modifiable lifestyle behaviors on depression remains insufficiently understood.
Objective:
To examine the associations between daily lifestyle behaviors—including sleep duration, physical activity, and digital or social engagement—and depression risk and subjective health among Chinese older adults.
Methods:
We conducted a cross-sectional analysis using data from the 2020 China Health and Retirement Longitudinal Study (CHARLS). Depression was defined as a CES-D10 score ≥ 10. Key independent variables included sleep duration, physical activity, alcohol consumption, smoking status, and participation in internet use and social activities. Modified Poisson regression and logistic models were employed to assess associations with depression risk, self-rated health, and health satisfaction, adjusting for age, gender, marital status, education, income, region, and urban residence.
Results:
Among 15,935 eligible middle-aged and older adults, the prevalence of depression was 45.1%. Longer sleep duration, participation in leisure and community activities, and digital engagement were associated with lower depression risk. Conversely, older age, female gender, lower income, and rural residence were associated with higher risk. Similar associations were observed for self-rated health and health satisfaction.
Conclusion:
Sleep duration, physical activity, and digital or social engagement were independently associated with depression and subjective health among older adults in China. These findings support the value of behavioral health strategies targeted toward the aging population.
Keywords:
depression Risk
mental health
Chinese middle-aged and Older Adults
lifestyle
health behaviors
A
A
Introduction
Global population ageing is accelerating [1], and China is undergoing one of the most rapid demographic transitions worldwide. With the largest older population globally, China reported over 260 million individuals aged 60 and above in 2020, accounting for 18.7% of its total population. As the cohort born during the "second baby boom" (1962–1975) enters retirement age starting in 2022, the pace of population ageing in China is expected to intensify, placing unprecedented pressure on family caregiving and the social security system [2]. China's population is projected to decline sharply from 1.4 billion in 2017 to 732 million by 2100 (95% uncertainty interval [UI]: 456–1499 million), representing a 48.0% decrease. By 2050, the total fertility rate (TFR) in China will likely remain below the replacement level of 2.1, coexisting with accelerated ageing and fundamentally reshaping the country’s demographic and economic landscape [3].
Against this backdrop of demographic transformation, depressive symptoms are relatively common among older Chinese adults [4], yet they often go undetected and untreated. Late-life depression is associated with diminished quality of life [5–6], functional decline, increased mortality, and rising healthcare expenditures. These effects may be driven by both biological and social mechanisms. Biologically, depression in older adults is often accompanied by neuroinflammation, impaired immune responses, and comorbid chronic conditions, all of which exacerbate frailty and functional deterioration. Socially, late-life depression contributes to social withdrawal, reduced health-seeking behavior, and poor adherence to treatment, leading to worse health outcomes and increased medical burden.
Although demographic and clinical determinants of depression have been widely studied, the influence of modifiable lifestyle and behavioral factors remains underexplored in middle-aged and older adults. Growing evidence indicates that regular physical activity [7], adequate sleep [8], and active social engagement [9] can reduce the risk of depression and enhance overall well-being. Meanwhile, “digital engagement”—such as internet use for communication, entertainment, or daily tasks—is becoming increasingly common among middle-aged and older adults and may offer additional psychosocial benefits [10–11]. However, few studies have used nationally representative data to assess how these behavioral factors jointly affect mental and subjective health. Much of the existing research is limited by small samples, inconsistent definitions, or a focus on younger cohorts.
To address these gaps, this study draws on data from the 2020 wave of the China Health and Retirement Longitudinal Study (CHARLS) to examine the associations of sleep duration, physical activity, and digital or social engagement with depression, self-rated health, and health satisfaction among middle-aged and older adults. The findings aim to inform evidence-based behavioral strategies to support mental health and healthy ageing in an ageing society.
Materials and Methods
2.1. Data Source and Study Population
This study used data from the 2020 wave of the China Health and Retirement Longitudinal Study (CHARLS), a nationally representative survey of middle-aged and older adults more than the age of 45 years in China. The survey collected extensive information on demographic, socioeconomic, behavioral, and health-related characteristics. The 2020 wave employed a multi-stage stratified probability sampling strategy across 28 provinces, autonomous regions, and municipalities, and included structured interviews with approximately 19,000 older adults.
For this analysis, the sample was restricted to older adults with complete data on depressive symptoms, lifestyle and behavioral variables, and all relevant covariates. After excluding those with missing or implausible responses, the final analytic sample consisted of 15,935 older adults. This sample reflects the community-dwelling older population in China and has been widely applied in studies of aging and public health.
2.2. Measures
The primary outcome was depression, assessed using the 10-item version of the Center for Epidemiologic Studies Depression Scale (CES-D10) [12]. Older adults were asked how often they experienced specific depressive symptoms over the past week, rated on a four-point scale. Total scores ranged from 0 to 30, with higher scores indicating greater severity. A score of ≥ 10 was used to define clinically relevant depression risk, in line with established practice[13]. The CES-D10 has demonstrated strong validity and reliability among older adults in China. This threshold has been validated in large-scale Chinese aging cohorts, including CHARLS, and shown to provide optimal sensitivity and specificity in identifying clinically significant depressive symptoms [12–13]. Sensitivity analyses in prior studies further support the use of ≥ 10 as a reliable and stable cutoff for population-based depression screening among adults aged 45 and above.
2.2.2. Secondary Outcomes
Two additional outcomes were evaluated [14]: self-rated health and health satisfaction. Self-rated health was measured by asking older adults to rate their overall health on a five-point Likert scale, ranging from 1 (“very poor”) to 5 (“very good”). Health satisfaction was measured by asking “Are you satisfied with your health?”, with responses from 1 (“not at all satisfied”) to 5 (“extremely satisfied”). These self-reported indicators reflect perceived physical and psychological well-being and are widely used in aging-related research.
2.2.3. Key Independent Variables
The key independent variables were modifiable lifestyle and behavioral factors. Sleep duration was self-reported in minutes per day and analyzed both as a continuous variable and categorically (< 420 minutes vs. ≥420 minutes, equivalent to < 7 hours vs. ≥7 hours) [15]. Physical activity was categorized into light, moderate, or vigorous intensity based on reported frequency and type[16]. Internet use was assessed by specific purposes, including watching videos, playing online games, chatting, reading news, and using mobile payments[17]; each activity was recorded as a binary variable. Social participation encompassed leisure activities, volunteering, community involvement, and participation in training or educational programs. Smoking status was coded as current smoker or non-smoker. Alcohol use was categorized as non-drinker, occasional drinker (less than once per month), or frequent drinker (more than once per month).
2.2.4. Covariates
Several covariates were included to adjust for potential confounding factors. Demographic variables comprised age group (45–59, 60–74, 75–89, and ≥ 90 years), gender (male or female), and marital status (married vs. unmarried). Socioeconomic factors included education level (primary or below, secondary, and college or above) and household income quintiles. Geographic characteristics included urban or rural residence and region (Eastern, Central, Western, or Northeastern China). Additionally, we adjusted for health insurance status (public, employer-based, or none) and living arrangement (living alone or with others).
2.3. Statistical Analysis
All analyses were performed using Stata version 18.0 (StataCorp LLC, College Station, TX, USA)[18]. Descriptive statistics were used to summarize the characteristics of older adults in the sample. Continuous variables were presented as means with standard deviations, and categorical variables as frequencies and percentages.
Given the relatively high prevalence of depression, modified Poisson regression models with robust standard errors were used to estimate adjusted risk ratios (aRRs) and 95% Confidence Intervals (CIs) for the associations between each lifestyle or behavioral factor and depression. Given the high prevalence of depression in this sample, modified Poisson regression was employed in place of conventional logistic regression to estimate associations with lifestyle factors. Logistic regression tends to yield ORs that exaggerate the magnitude of associations when the outcome is common, limiting interpretability in public health contexts. In contrast, modified Poisson regression allows for the direct estimation of relative risks, which are more intuitive and policy-relevant effect measures. This approach also retains robust standard error estimation, offering both statistical reliability and conceptual clarity when modeling prevalent binary outcomes. This method provides more accurate estimates of relative risk than traditional logistic regression when outcomes are common. Independent variables included sleep duration, physical activity, digital engagement, social participation, smoking, and alcohol use. All models were adjusted for age group, gender, marital status, education level, household income quintile, geographic region, urban or rural residence, health insurance status, and living arrangement.
For the two secondary outcomes—self-rated health and health satisfaction—multivariable logistic regression models were applied to estimate ORs and 95% CIs, using the same covariate structure as in the primary analysis.
Subgroup analyses were conducted by gender and age group (< 70 years vs. ≥70 years) to assess potential effect modification. Statistical significance was determined at a two-sided p-value threshold of < 0.05. Multicollinearity was evaluated using variance inflation factors, and no concerning levels were detected. Model diagnostics and goodness-of-fit tests were performed to assess robustness.
Results
3.1. Participant Characteristics
A
A total of 15,935 middle-aged and older adults more than the age of 45 years were included in the final analysis. The sample was nearly evenly divided by gender, with 51.8% identified as female (n = 8,257) and 48.2% as male (n = 7,678). This balance provides a robust basis for gender-specific analyses of mental and behavioral health characteristics among middle-aged and older adults in China.
In terms of age structure, the majority of the sample fell into the 45–59 and 60–74 year age groups. Specifically, 47.6% of participants were aged 45 to 59 years (n = 7,585), and 44.4% were between 60 and 74 years old (n = 7,072), while only 8.0% were in the oldest group, aged 75 to 89 years (n = 1,267). This age distribution reflects the demographic composition of China’s middle-aged and older adult population and allows for meaningful subgroup analysis by age.
A
Although other sociodemographic variables such as marital status, educational attainment, residential setting, or household composition were not directly reported in this summary table, the available data provide a comprehensive baseline profile of participants’ age and gender distribution. These foundational characteristics support subsequent analyses examining the relationships between demographic traits, lifestyle behaviors, and mental health outcomes in this population. Detailed descriptive statistics are presented in Table 1.
[Table 1. Descriptive characteristics of the study population (n = 15,935)]
Table 1. Descriptive characteristics of the study population (n = 15,935)
3.2. Bivariate Associations with Depression
A
A total of 15,935 middle-aged and older adults were included in the analysis. Among them, 7,178 individuals (45.0%) met the CES-D10 threshold for depressive symptoms, whereas 8,757 individuals (55.0%) were classified as non-depressed. These figures, presented in Table 2, indicate that nearly one in every two participants in this nationally representative sample experienced clinically relevant depressive symptoms, underscoring a substantial mental health burden in later life.
The descriptive statistics in Table 2 also provide a general profile of the study population. In addition to the high prevalence of depressive symptoms, the table summarizes key health outcome variables and their distribution, offering a snapshot of both psychological status and subjective well-being at the population level. The clear contrast between the depressed and non-depressed groups highlights the potential impact of mental health challenges on a large segment of the ageing population, reinforcing the public health importance of depression screening and prevention in community settings.
[Table 2. Distribution of depressive symptoms, health satisfaction, and self-rated health]
Table 2. Distribution of depressive symptoms, health satisfaction, and self-rated health
3.3. Subgroup Analyses by Gender and Age
A
Table 3 outlines the distribution of health-related behaviors within the study population. The average reported sleep duration was 403.5 minutes (SD = 124.5). In terms of physical activity, over half of participants (54.9%) reported no exercise, 39.6% engaged in light activity, 3.9% in moderate activity, and 1.5% in vigorous activity. With respect to smoking, 26.4% identified as current smokers. Patterns of alcohol consumption indicated that 62.3% abstained, 9.9% consumed alcohol less than once per month, and 27.7% drank more frequently. Social participation varied: 39.4% reported involvement in activities with friends or relatives, 21.4% participated in recreational or leisure pursuits, 2.7% engaged in community activities, 3.4% contributed to volunteer or charity work, and 1.7% attended school or training programs. Digital engagement was also notable, with 43.5% of participants using the internet. Among these, 25.7% reported chatting online, 31.0% reading news online, 32.8% watching videos, and 5.1% playing online games. Additionally, 27.3% used mobile payment systems, 40.0% used WeChat, and 23.6% posted content on WeChat Moments.
[Table 3. Behavioral Characteristics of the Study Population]
Table 3. Behavioral Characteristics of the Study Population
Descriptive statistics summarize overall health and behavioral characteristics of the full study population (n = 15,935). Sleep duration is reported as mean minutes per day. Behavioral variables include physical activity levels, smoking and alcohol use, digital engagement, and social participation (leisure games, community involvement).
3.4. Sensitivity and Robustness Analyses
A
To evaluate the robustness of the associations observed in the primary analysis, we conducted stratified comparisons of depressive symptoms across subgroups of older adults, focusing primarily on gender differences. As shown in Table 4, the prevalence of depressive symptoms was significantly higher among female participants (52.7%) compared to their male counterparts (36.9%), with a p-value less than 0.001. This finding reinforces the gender-specific trends observed in previous research and highlights the potential value of incorporating gender considerations into mental health interventions targeting older adults. Although age categories were noted in the table, stratified estimates of depression prevalence by age group were not fully presented. Therefore, conclusions about age-related patterns in depressive symptoms remain limited at this stage of the analysis. Overall, the gender-stratified results align with patterns observed in the multivariable regression models, suggesting that the associations between behavioral factors and depression are not merely attributable to random variation or a few confounding demographic variables. These findings underscore the importance of demographic context—particularly gender—in understanding disparities in mental health outcomes among older adults.
[Table 4. Gender-stratified comparisons of depression prevalence]
Table 4. Gender-stratified comparisons of depression prevalence
3.5. Multivariable Regression Results
A
Adjusted logistic regression models were used to examine the associations between lifestyle and behavioral factors and three outcomes: depression, self-rated health, and health satisfaction. As shown in Table 5, several variables demonstrated consistent associations across all outcomes.
Middle-aged and older male participants had significantly lower ORs of depressive symptoms (OR = 0.64, 95% CI: 0.59–0.70, p < 0.001), consistent with the gender differences observed in descriptive analyses. Older adults aged 75–89 were at significantly higher risk of depression compared to middle-aged adults groups. Higher income levels were associated with reduced depression risk as well as more favorable ratings of self-rated health and health satisfaction, whereas those in the lower-middle and lowest income groups had increased ORs of reporting poor subjective health outcomes.
Longer sleep duration was inversely associated with depression; older adults who slept more than 420 minutes (7 hours) per night were less likely to report depressive symptoms (OR = 0.60, 95% CI: 0.80–0.98). Digital behaviors, such as participation in online gaming and the use of mobile payment services, were linked to better mental health outcomes. In contrast, frequent alcohol consumption and physical inactivity were associated with higher depression risk and poorer subjective health.
Social engagement also emerged as a significant protective factor. Older adults who participated in leisure activities, community service, or training programs were more likely to report better overall health and lower levels of depression.
[Table 5. Adjusted odds ratios from multivariable logistic regression models]
Table 5. Adjusted odds ratios from multivariable logistic regression models
3.6. Extended or Additional Models
A
To further evaluate the consistency of observed relationships, additional multivariable models were constructed incorporating key behavioral and demographic covariates. These extended analyses confirmed the robustness of earlier findings and provided additional insight into subgroup-specific effects, as shown in Table 6.
In univariate comparisons, gender differences in depression remained pronounced, with depressive symptoms reported by more than half of female older adults and just over one-third of males. These differences were statistically significant and persisted after adjustment. In the fully adjusted model, male gender continued to be strongly associated with reduced ORs of depression (OR = 0.64, 95% CI: 0.58–0.69, p < 0.001), closely mirroring the results from the primary model.
The extended models also re-evaluated digital behavior and social participation in relation to mental health. Online chatting, mobile payment use, and participation in leisure activities all maintained significant protective associations, even when adjusted for additional socioeconomic variables. No evidence of multicollinearity or model instability was detected across these specifications. Although the adjusted models demonstrated consistent associations for most variables, certain results warrant further clarification. For instance, while the unadjusted prevalence of depressive symptoms was lower among individuals who posted on WeChat Moments (41.1%) compared to non-posters (44.1%), the adjusted ORs ratio slightly exceeded 1 (AOR = 1.03, p < 0.001), suggesting no substantial protective effect after accounting for covariates. This pattern may reflect confounding or possible reverse causality, whereby individuals with better mental health are more likely to engage digitally. Additionally, variables such as smoking and light alcohol consumption were not significantly associated with depression risk in multivariable models, indicating the potential influence of social norms or residual confounding in this population.
[Table 6. Univariate and multivariate associations with depressive symptoms]
Table 6. Univariate and multivariate associations with depressive symptoms
4. Discussion
Our findings indicate that four key lifestyle behaviors—sleep duration, physical activity, digital engagement, and social participation—were independently associated with depression risk and subjective health. These associations remained robust across age and gender groups, highlighting actionable targets for mental health promotion in later life. This study revealed a high prevalence of depression risk, as assessed by CES-D10 screening, among middle-aged and older adults in China, with notable disparities across demographic, socioeconomic, and behavioral subgroups. Depression was more commonly reported among women, individuals with lower educational attainment or income, and those residing in rural areas. These findings are consistent with prior research based on the China Health and Retirement Longitudinal Study (CHARLS). For instance, a large-scale meta-analysis involving over 260,000 participants reported that depression prevalence declined slightly in the early phase of the COVID-19 pandemic but increased significantly thereafter (P for trend < 0.001), with overall prevalence shifting from 25.8% before the pandemic to 23.8% during the pandemic. Higher prevalence was observed among women, rural residents, individuals with low education levels, and those living alone [19]. Another study using data from the 2018 CHARLS found that depressive symptoms were more prevalent among rural adults aged 80 and above (15.7%) compared to their urban counterparts (12.3%), with disparities primarily attributable to differences in income, marital status, physical activity, and self-rated health status[20].
Several modifiable lifestyle factors demonstrated robust associations with depressive symptoms. Short sleep duration was significantly associated with elevated depression risk, emphasizing the often underappreciated role of sleep in supporting mental well-being among older adults. A 2020 CHARLS-based study found significant interaction effects between sleep duration and limitations in activities of daily living (ADLs), suggesting that short sleep combined with poor functional status can synergistically elevate depression risk, whereas interventions aimed at optimizing sleep and improving ADL function may help prevent depression among middle-aged and older adults [21]. Another CHARLS study reported that urban older adults who engaged in 9,000–12,000 METs of recreational physical activity per week experienced a 71.7% reduction in depression risk (OR = 0.283), while rural middle-aged and older adults performing more than 6,000 METs of non-recreational physical labor had significantly elevated risk (OR = 2.224) [22].
Digital behaviors such as mobile payment use and online communication were inversely associated with depression, indicating that digital inclusion is linked to greater emotional connection, autonomy, and a sense of control among older adults. This aligns with findings from Journal of Medical Internet Research (2025), which showed that regular digital engagement was linked to lower risks of both sleep disturbances and depressive symptoms (relative risk for poor sleep: 0.57 in women and 0.61 in men) [23]. A separate longitudinal analysis using CHARLS data found that older adults who engaged in at least three forms of digital social activity had significantly higher ORs of depression remission over a two-year period (OR = 1.24–1.36) [24].
Social participation also emerged as a particularly strong protective factor for both mental and subjective health. Participation in leisure activities, volunteering, and community events was associated with reduced depression risk and greater health satisfaction. These benefits may be related to the psychological value of structured routines, a sense of purpose, and interpersonal connectedness—all of which have been associated with lower levels of loneliness and emotional distress. Supporting this, a CHARLS-based study showed that older adults who lacked sufficient weekly physical activity had a depression prevalence of 31.6%, markedly higher than the 18.9% observed in active individuals, reinforcing the importance of physical activity in mental health promotion among aging populations [25]. More recently, a 2024 study using CHARLS data confirmed that community participation—such as volunteering or involvement in social organizations—was independently associated with lower of depressive symptoms, with stronger effects observed at higher frequencies of participation (OR range: 0.78–0.95) [26].
A
While digital engagement may partially compensate for the lack of face-to-face interaction, authentic offline social relationships remain essential for psychological resilience in later life. Collectively, these findings highlight the need to shift public health strategies beyond clinical treatment toward addressing upstream social determinants of mental health. Aging-focused community programs should incorporate interventions to promote sleep hygiene, encourage physical activity, and expand equitable access to digital tools. Digital inclusion policies—particularly targeting women and rural older adults—may yield additional mental health benefits. Moreover, tailoring behavioral interventions by gender and age group may further enhance their precision and effectiveness. Although smoking and light alcohol use were not significantly associated with depression in our findings, these results warrant further reflection. Several explanations may account for the lack of statistical significance, including potential underreporting, social desirability bias, or residual confounding. It is also possible that cultural norms regarding tobacco and alcohol use in this age group dilute measurable associations. Future research using more detailed behavioral exposure metrics and stratified analyses may help clarify these complex relationships.
A major strength of this study lies in its use of nationally representative data and the comprehensive assessment of lifestyle and behavioral factors. Nevertheless, several limitations should be noted. One important concern is the cross-sectional design, which restricts our ability to make causal inferences. While the observed associations offer valuable insights for public health strategy, they do not establish temporal precedence. In addition, the reliance on self-reported data may introduce recall and reporting biases.
To address these issues, future research should adopt longitudinal designs that can clarify the temporal dynamics between lifestyle behaviors and mental health. Further work is also needed to explore the quality and context of digital and social engagement. Finally, intervention-based studies will be essential to assess whether lifestyle-oriented strategies can effectively improve mental health outcomes among middle-aged and older adults.
5. Conclusion
Depression is common among middle-aged and older adults in China and is associated with short sleep duration, physical inactivity, limited digital engagement, and low social participation. These modifiable behaviors represent actionable targets for mental health promotion. Interventions that support healthy sleep, encourage physical and social activity, and expand digital inclusion may help reduce depression risk and enhance subjective well-being in this population.
A
Data Availability
The data used in this study are from the China Health and Retirement Longitudinal Study (CHARLS) 2020 wave, which is publicly available to researchers upon application. The CHARLS datasets can be accessed at: http://charls.pku.edu.cn/en.
A
Author Contribution
SW conceived and designed the study, performed data cleaning and statistical analysis, and drafted the initial manuscript.YS supervised the study design and analysis, contributed to the interpretation of the results, and critically revised the manuscript for important intellectual content.AI assisted with data analysis, contributed to the literature review, and participated in drafting and revising the manuscript.MO contributed to the study design, interpretation of findings, and critical review of the manuscript.MY provided methodological guidance, contributed to interpretation, and reviewed the manuscript for accuracy and clarity.All authors read and approved the final manuscript and agree to be accountable for all aspects of the work.
Acknowledgements
Not applicable.
Electronic Supplementary Material
Below is the link to the electronic supplementary material
Reference
1.Kemoun P, Ader I, Planat-Benard V, et al. A gerophysiology perspective on healthy ageing. Ageing Res Rev. 2022;73:101537. 10.1016/j.arr.2021.101537.
2.Chen X, Giles J, Yao Y, et al. The path to healthy ageing in China: a Peking University-Lancet Commission. Lancet. 2022;400(10367):1967–2006. 10.1016/S0140-6736(22)01546-X.
3.Vollset SE, Goren E, Yuan CW, et al. Fertility, mortality, migration, and population scenarios for 195 countries and territories from 2017 to 2100: a forecasting analysis for the Global Burden of Disease Study. Lancet. 2020;396(10258):1285–306. 10.1016/S0140-6736(20)30677-2.
4.Zhou P, Wang S, Yan Y, et al. Association between chronic diseases and depression in the middle-aged and older adult Chinese population-a seven-year follow-up study based on CHARLS. Front Public Health. 2023;11:1176669. 10.3389/fpubh.2023.1176669. Published 2023 Jul 20.
5.Zhang P, Wang L, Zhou Q, et al. A network analysis of anxiety and depression symptoms in Chinese disabled elderly. J Affect Disord. 2023;333:535–42. 10.1016/j.jad.2023.04.065.
6.Gambaro E, Gramaglia C, Azzolina D, Campani D, Molin AD, Zeppegno P. The complex associations between late life depression, fear of falling and risk of falls. A systematic review and meta-analysis. Ageing Res Rev. 2022;73:101532. 10.1016/j.arr.2021.101532.
7.Kim H, Yoo J, Han K, Jeon HJ. Physical activity and cardiovascular health in depression: Links between changes in physical activity and cardiovascular risk. Gen Hosp Psychiatry. 2022;78:35–41. 10.1016/j.genhosppsych.2022.07.002.
8.Plante DT. The Evolving Nexus of Sleep and Depression. Am J Psychiatry. 2021;178(10):896–902. 10.1176/appi.ajp.2021.21080821.
9.Zhu S, Kong X, Han F, et al. Association between social isolation and depression: Evidence from longitudinal and Mendelian randomization analyses. J Affect Disord. 2024;350:182–7. 10.1016/j.jad.2024.01.106.
10.Alber CS, Krämer LV, Rosar SM, Mueller-Weinitschke C. Internet-Based Behavioral Activation for Depression: Systematic Review and Meta-Analysis. J Med Internet Res. 2023;25:e41643. 10.2196/41643. Published 2023 May 25.
11.Liu H, Fan X, Luo H, et al. Comparison of Depressive Symptoms and Its Influencing Factors among the Elderly in Urban and Rural Areas: Evidence from the China Health and Retirement Longitudinal Study (CHARLS). Int J Environ Res Public Health. 2021;18(8):3886. 10.3390/ijerph18083886. Published 2021 Apr 7.
12.Fu H, Si L, Guo R. What Is the Optimal Cut-Off Point of the 10-Item Center for Epidemiologic Studies Depression Scale for Screening Depression Among Chinese Individuals Aged 45 and Over? An Exploration Using Latent Profile Analysis. Front Psychiatry. 2022;13:820777. Published 2022 Mar 14. 10.3389/fpsyt.2022.820777
13.Pan Y, Pikhartova J, Bobak M, Pikhart H. Reliability and predictive validity of two scales of self-rated health in China: results from China Health and Retirement Longitudinal Study (CHARLS). BMC Public Health. 2022;22(1):1863. Published 2022 Oct 5. 10.1186/s12889-022-14218-1
14.Wang T, Han W, Wang C, et al. Interaction effects of sleep duration and activities of daily living on depressive symptoms among Chinese middle-aged and older adult individuals: evidence from the CHARLS. Front Public Health. 2025;13:1547329. 10.3389/fpubh.2025.1547329. Published 2025 Mar 12.
15.Wang J. The longitudinal relationship between leisure activities and depressive symptoms among older Chinese adults: an autoregressive cross-lagged analysis approach. BMC Public Health. 2024;24(1):763. 10.1186/s12889-024-18293-4. Published 2024 Mar 12.
16.Yang HL, Zhang S, Cheng SM et al. A study on the impact of Internet use on depression among Chinese older people under the perspective of social participation. BMC Geriatr. 2022;22(1):701. Published 2022 Aug 24. 10.1186/s12877-022-03359-y
17.Lin X, Liu S, Hu Z, Xu H. Social activities and long-term depressive-symptoms trajectories among middle-aged and older adults in China: a population-based cohort study. Front Psychiatry. 2023;14:1131084. Published 2023 Aug 16. 10.3389/fpsyt.2023.1131084
18.Daly C, Soobiah C. Software to Conduct a Meta-Analysis and Network Meta-Analysis. Methods Mol Biol. 2022;2345:223–44. 10.1007/978-1-0716-1566-9_14.
19.Zhao X, Du X, Bai S, Zheng P, Zhou X, Wang Z. Differences in depression prevalence among older adults in China before and during the COVID-19 pandemic: a systematic review and meta-analysis. PeerJ. 2025;13:e19251. 10.7717/peerj.19251. Published 2025 Apr 11.
20.Hu C, Jiang Q, Yuan Y, et al. Depressive symptoms among the oldest-old in China: a study on rural-urban differences. BMC Public Health. 2024;24(1):3604. 10.1186/s12889-024-21069-5. Published 2024 Dec 30.
21.Wang T, Han W, Wang C, et al. Interaction effects of sleep duration and activities of daily living on depressive symptoms among Chinese middle-aged and older adult individuals: evidence from the CHARLS. Front Public Health. 2025;13:1547329. 10.3389/fpubh.2025.1547329. Published 2025 Mar 12.
22.Wang R, Tan J, Su Z et al. Dose-response relationship between different physical activity types and depressive symptoms in middle-aged and elderly people: a study of Chinese urban-rural differences. BMC Public Health. 2025;25(1):1931. Published 2025 May 26. 10.1186/s12889-025-23128-x
23.Li X, Liu J, Huang N, Zhao W, He H. Association Between Internet Use and Sleep Health Among Middle-Aged and Older Chinese Individuals: Nationwide Longitudinal Study. J Med Internet Res. 2025;27:e71030. 10.2196/71030. Published 2025 Apr 16.
24.Qi Y, Zhang C, Zhou M, Zhang R, Chen Y, Li C. Association between social media use and depressive symptoms in middle-aged and older Chinese adults. Transl Psychiatry. 2024;14(1):430. Published 2024 Oct 11. 10.1038/s41398-024-03142-0
25.Li S, Zhang J, Yang Y. Correlation between the physical activity volume and cognitive and mental capacity among older adult people in China: a cross-sectional study based on the 2020 CHARLS database. Front Public Health. 2024;12:1462570. 10.3389/fpubh.2024.1462570. Published 2024 Nov 20.
26.Zhang X, Xue M, Zhang Z et al. Impact of social, familial and personal factors on depressive symptoms in middle-aged and older adults from the national CHARLS cohort. BMC Public Health. 2024;24(1):2669. Published 2024 Sep 30. 10.1186/s12889-024-20159-8