The association between physical activity and depressive symptoms among Chinese older adults: the chain mediating roles of sleep duration and cognitive function Authors: Tongnian Yang, Jie Ren, Hui Xiao Published: 2025 ABSTRACT AbstractBackgroundDepression exhibits a high prevalence among the elderly population, posing a significant burden on public health. Although extensive research confirms that physical activity can alleviate depressive symptoms, there is insufficient evidence regarding the chain-mediated role of sleep duration and cognitive function in this process. This study aims to investigate the chain-mediating effects of sleep duration and cognitive function in the association between physical activity and depressive symptoms in Chinese older adults.MethodsData were drawn from the 2018 China Health and Retirement Longitudinal Study (CHARLS), with 4,744 adults aged 60 and older meeting inclusion criteria. Using correlation analysis, one-way ANOVA, and chained mediation models (PROCESS macro model 6), we examined the direct and indirect effects of physical activity on depressive symptoms via sleep duration and cognitive function.Results(1) There was a significant correlation between physical activity, sleep duration, cognitive function, and depressive symptoms (P < 0.01). (2) Sleep duration and cognitive function significantly mediated the association between physical activity and depressive symptoms. The three indirect pathways were physical activity → sleep duration → depressive symptoms (β = -0.111, 95% CI [-0.166, -0.061], 19.1% of total effect); physical activity → cognitive function → depressive symptoms (β = -0.173, 95% CI [-0.258, -0.088], 29.8%); and physical activity → sleep duration → cognitive function → depressive symptoms (β = -0.011, 95% CI [-0.019, -0.004], 1.9%).ConclusionSleep duration and cognitive functioning act as chain mediators between physical activity and depressive symptoms in older adults, i.e., physical activity can improve sleep duration in order to enhance cognitive functioning and, in turn, improve depressive symptoms in older adults.Keywords: Physical activity, Older adults, Depressive symptom, Sleep duration, Cognitive function Introduction Population aging is emerging as a significant global public health challenge. By 2050, the global population of older individuals is projected to double to approximately 2.1 billion [1]. In China, the elderly population has surged dramatically, increasing from 126 million in 2000 to 264 million in 2020, and is expected to reach 480 million by 2050. This will account for 30% of China’s total population and 20% of the global elderly population [2]. The substantial growth of the elderly population presents considerable challenges across various societal domains, including public health, medical care, and economic development. Research has indicated that physiological decline, poor physical health, and the sense of uselessness following career cessation pose serious threats to the mental and psychological well-being of older adults, leading to rising rates of depression [3]. Globally, depression has increased by approximately 50% from 1990 to 2017 and is projected to become the leading cause of the global disease burden by 2030 [4]. Depression in older adults represents a pressing public health concern, with a reported prevalence of 20.6% among Chinese older adults [5]. This condition significantly diminishes quality of life [6, 7] and imposes substantial economic burdens on families and society [8, 9]. In this context, physical activity has emerged as a promising non-pharmacological intervention, demonstrating consistent negative associations with depressive symptoms [10, 11]. However, the mechanisms underlying this association require further elucidation. Prior research has primarily focused on sleep duration [55] and cognitive function [57, 58] as separate mediators, neglecting their possible interconnections. Therefore, this study aims to investigate the chain mediation effect of sleep duration and cognitive function in the association between physical activity and depressive symptoms among older Chinese adults. Exploring this sequential pathway will contribute to a more comprehensive understanding of underlying mechanisms and provide evidence for developing multi-component interventions targeting sleep and cognition, thereby maximizing the mental health benefits of physical activity. The association between physical activity and depressive symptoms Physical activity is an effective means of alleviating negative emotions such as depression, with the advantages of high compliance, low adverse effects, and stable results [10]. Research has found that appropriate physical activity in older adults is effective in preventing and intervening in depression, as well as improving sleep, quality of life, and physical functioning [11]. Consistent participation in physical activities, as a way of life that directly promotes physical health, can improve the physiological functions of the elderly, including cardiovascular functions, so that they can maintain a relatively good state of physical health, thereby reducing the negative emotions such as anxiety and depression that result from the aging of the elderly’s physiological functions, and enhancing their mental health [12]. Exercise is often perceived as a difficult task for older people, and being able to develop a habit of exercising can lead to a sense of self-efficacy, which can dissipate depression [13]. Older people’s participation in physical activities enhances social interactions, gains more attention, increases the level of social support through useful social interactions, improves mental health, and reduces depressive symptoms [14]. Research has found a positive feedback mechanism between physical activity and depression, whereby older people with better mental health are more physically active, and regular participation in physical activity further improves mental health and reduces depressive symptoms [15]. Physical activity is negatively associated with the occurrence of depressive symptoms in older adults, i.e., as levels of physical activity increase, depressive symptoms typically decrease in older adults; conversely, lack of physical activity may be associated with a higher risk of depression [16]. Based on this, Hypothesis H1: Physical activity negatively predicts depressive symptoms in older adults. The mediating role of sleep duration Sleep duration generally refers to the amount of time older adults spend sleeping at night, and it has a significant impact on their mental health [17]. But not all older people will benefit, and those who will benefit most are those who get enough hours of sleep. Sleep efficiency declines over time in older people, falling by 18.6% between the ages of 40 and 100 [18]. The cumulative, synergistic effect of sleep deprivation as an important predisposing factor for cardiovascular disease, inflammation, metabolic disorders, and other disorders may further diminish the level of competence of older adults [19]. When sleep needs are not met, an individual’s mental and psychological state suffers [20]. Studies have found that lack of sleep has been shown to be a predictor of depressive symptoms and that there is a link between shorter sleep duration and an increased risk of depressive symptoms [21, 22]. Physical activity is positively correlated with sleep quality, i.e., as physical activity increases, sleep quality improves and sleep duration is optimized for older people, thus contributing to the promotion of physical and mental health [23]. The study found that an 8-week daily walking exercise program was effective in improving sleep quality in older adults [24]. In addition, physical activity mitigates the negative effects of depression by improving the quality and duration of sleep-in older adults with sleep disorders [25]. This suggests that older adults who engage in regular physical activity on a regular basis tend to exhibit better sleep quality and duration, and that adequate sleep duration reduces depressive symptoms and enhances the ability to be physically active the following day, thus creating a positive feedback mechanism. Based on this, hypothesis H2 is proposed: Sleep duration mediates the association between physical activity and depressive symptoms. The mediating role of cognitive functions Cognitive function is the mental function of the human brain to recognize and reflect objective things, including perception, attention, memory, thinking, language, and other various abilities [26]. It has been found that people with cognitive dysfunction may experience neuroinflammation and oxidative stress, causing damage to neurons or synapses in the brain, leading to slowed thinking, decreased self-care, and negative emotions such as anxiety and depression [27]. At the same time, depression can lead to negative mood, brain damage, slow response, consumption of a lot of body energy, resulting in physical and mental exhaustion, aggravated cognitive dysfunction, and even the development of dementia. Physical activity is an effective, low-cost way to improve cognitive function and reduce the risk of dementia in older adults. Physical activity reduces the risk of Alzheimer’s disease by increasing cardiorespiratory fitness and improving cognitive function in older adults [28]. Physical activity can have an impact on cognitive functioning in older adults not only through direct pathways but also indirectly through cognitive functioning on depression. A significant positive correlation was found between physical activity and cognitive function [29], with older adults who participated in high and moderate levels of physical activity having significantly better cognitive function than those who participated in low levels of physical activity. There is a close symbiotic and causal bidirectional association between cognitive function and depression, with cognitive function being negatively associated with depression in older adults, and those with cognitive dysfunction also having a relatively higher risk of depressive symptoms compared to middle-aged and older adults with normal cognitive function [30]. This leads to the conclusion that regular participation in physical activity is beneficial in reducing the reduction in the volume of grey matter in the brain and the reduction in brain volume, which will lead to better top-down self-regulation and will have a facilitating effect on cognitive functioning in older adults affected by psychological problems, thus reducing the risk of developing depression. Based on this, Hypothesis H3: Cognitive function mediates the association between physical activity and depressive symptoms. Chain mediation of sleep duration and cognitive functioning Physical activity, sleep duration, and cognitive function each had a significant effect on reducing the risk of depressive symptoms in older adults, and sleep duration and cognitive function each played a simple mediating role between physical activity and depressive symptoms in older adults. At the same time, there is a significant correlation between sleep duration and cognitive function [31]; sleep duration is a protective factor for cognitive function, and good sleep positively contributes to cognitive and memory stabilization, i.e., sleep duration influences cognitive performance [32].Research has found that sleep duration is one of the key factors in improving cognitive function, and for older adults, better quality and duration of sleep can help enhance their cognitive function [33]. Engaging in physical activity for older adults not only promotes deeper sleep and more efficient sleep, which in turn improves sleep duration, but quality sleep is essential for the brain to make the necessary repairs and consolidate memories, which can help improve cognitive functioning, including attention, memory, and decision-making skills [34]. It was further shown that enhanced cognitive functioning, in turn, promotes the ability to regulate mood and cope with stress, thereby reducing the risk of depressive symptoms [35]. Based on existing literature, we hypothesize that physical activity may enhance cognitive function by improving sleep quality and duration, which in turn could reduce depressive symptoms. However, this chain mechanism has not been empirically tested in a single model, which constitutes the primary contribution of the present study. Based on this, the present study proposes hypothesis H4: Sleep duration and cognitive function may play a chain-mediating role between physical activity and depressive symptoms in older adults. In summary, this study constructs a chain mediation model as shown in Fig. 1. Diagram of the hypothetical chain intermediary model Methods Data sources and participants The current study employs a cross-sectional approach, utilizing data from the China Health and Retirement Longitudinal Study (CHARLS) 2018. The CHARLS survey is designed to gather high-quality microdata that represent households and individuals aged 45 and above in China. These data are intended to facilitate the analysis of population aging issues in China and support interdisciplinary research on aging. The CHARLS national baseline survey was initially conducted during the period of 2011–2012, followed by biennial follow-up visits. The survey covers a sample of county-level units across mainland China, excluding Tibet. Specifically, it includes 17,708 individuals aged 45 years or older and their spouses from 10,257 households in 150 county-level units and 450 village-level units [36]. The sample is distributed across 150 county-level units (spanning 28 provincial-level units), 450 village-level units, and 10,257 households, comprising a total of 17,708 individuals aged 45 years or older and their spouses. The CHARLS project has conducted four national follow-up surveys in 2013, 2015, 2018, and 2020. In the 2018 survey, a total of 19,816 individual questionnaires were successfully completed from 11,628 households of individuals aged 45 years and older. For the present study, the final sample of older adults aged 60 years and above included 4,744 participants (see Fig. 2). Ethical approval for this study was obtained from the Ethics Review Board of Peking University (Approval No. IRB00001052-11015). The study was conducted in accordance with the ethical principles of the Declaration of Helsinki. Informed consent was obtained from all participants prior to their inclusion in the study. All methods and experimental protocols were performed in compliance with relevant guidelines and regulations and were approved by the Institutional Review Board of Peking University. Sample Screening Flowchart The China Health and Retirement Longitudinal Study (CHARLS), due to its community-based nature and emphasis on population representativeness, includes individuals with varying degrees of cognitive impairment and depression severity in its sample. Although lacking clinical diagnoses of dementia or psychiatric disorders, all participants were included in the study. To eliminate potential confounding factors, our statistical models explicitly treated cognitive function and depressive symptoms as key variables (as mediating and outcome variables, respectively), thereby fully accounting for sample variability in the analysis. This inclusive approach enhances the generalizability of our findings to broader elderly populations. According to the purpose of this study, the data inclusion criteria were as follows: (1) age 60 years and older; (2) respondents who answered complete information on physical activity status, sleep status, cognitive functioning scale, and depression scale questionnaires. (3) Providing basic information on demographic variables such as age, gender, marital status, place of residence, and self-assessed health status. Physical activity In this study, physical activity levels were classified into three categories-high, medium, and low-based on the daily physical activity section of the CHARLS questionnaire. For the purpose of analysis, the duration of physical activity was categorized as follows: 10 min to less than 0.5 h was assigned a value of 30 min; 0.5 h to less than 2 h was assigned a value of 60 min; and 2 h to less than 4 h was assigned a value of 180 min [37]. To align with the data truncation principles of the International Physical Activity Questionnaire (IPAQ), the maximum reported duration for each level of physical activity was capped at 21 h per week, thereby preventing potential misclassification of individuals into the high physical activity group. The metabolic equivalent (MET) values were assigned based on the intensity of physical activity: low-intensity activities were assigned a MET value of 3.3, moderate-intensity activities a MET value of 4.0, and high-intensity activities a MET value of 8.0 [38]. The physical activity energy expenditure was calculated using the following formula: physical activity score = 8.0 × total weekly duration of high-intensity activity + 4.0 × total weekly duration of moderate-intensity activity + 3.3 × total weekly duration of walking [39]. The total weekly physical activity energy consumption was derived by summing the energy expenditure of each type of physical activity. Based on the IPAQ criteria, the physical activity levels of the elderly were categorized into three groups: low (< 600 MET/week), medium (600-3,000 MET/week), and high (> 3,000 MET/week). These categories were assigned numerical values of 0, 1, and 2, respectively, for use in subsequent data analyses. Sleep duration Sleep duration is from the question option in the 2018 CHARLS questionnaire. “In the past 1 month, on average, about how many hours per night did you actually fall asleep?” According to Ouyang et al. [40], night sleep duration < 6 h was defined as sleep deprivation, and night sleep duration ≥ 9 h was defined as sleep overdose, and in conjunction with the needs of this study, night sleep duration was classified into five groups, which were < 6 h, 6–7 h, 7–8 h, 8–9 h, and ≥ 9 h, respectively. Sleep duration was categorized into five groups based on common intervals used in geriatric sleep research to capture nuanced variations in sleep patterns. However, we acknowledge that these categories may not fully align with clinical thresholds. Cognitive function Cognitive function was assessed using the Mini-Mental State Examination (MMSE), a widely recognized tool for evaluating cognitive impairment. In the CHARLS 2018 survey, cognitive functioning was measured using the complete MMSE, which comprises 19 items across five dimensions: time and place orientation (10 points), situational memory (3 points), attention and calculation (5 points), recall (3 points), and verbal ability and drawing (9 points). For scoring, correct answers were awarded 1 point, while incorrect answers received 0 points, resulting in a total possible score ranging from 0 to 30 points. Based on existing research [41], cognitive function levels can be categorized as follows: scores of 0–9 indicate severe cognitive impairment, 10–20 indicate moderate cognitive impairment, 21–26 indicate mild cognitive impairment, and 27–30 indicate normal cognitive function. These standards were specifically established and validated for the elderly population in China. While a cut-off of 24 is commonly used in Western populations, the criteria from Han et al. (2023) [41] have been demonstrated to account for educational and cultural factors prevalent in the Chinese aging cohort, thereby providing a more contextually appropriate classification for our sample. The internal consistency of the MMSE scale in this study, as measured by Cronbach’s α coefficient, was 0.739, indicating a reliable assessment of cognitive functioning among the study participants. Depressive symptom In this study, depressive symptoms were assessed using the 10-item Centre for Epidemiologic Studies Depression Scale (CESD-10). This scale comprises 10 items that inquire about the frequency of specific symptoms experienced by respondents over the past week. Participants’ responses are rated on a 4-point Likert scale, ranging from 0 (rarely or not at all) to 3 (most of the time). The items are categorized into three domains: three items related to depressive moods, five items related to physical symptoms, and two items related to positive moods (which are reverse-scored). The total score ranges from 0 to 30, with higher scores indicating more severe depression. Based on the criteria established by Huang et al. [42], respondents with a total score of 10 or higher are typically considered to have mild to moderate depressive symptoms, while those with a score of 21 or higher are classified as having major depression. The internal consistency of the CESD-10 scale in this study was evaluated using Cronbach’s α coefficient, which was found to be 0.780. This value indicates a reliable assessment of depressive symptom levels in the study population. Covariates Drawing on prior research [43, 44], we identified several potential confounding factors that may influence the association between physical activity and depression in older adults. These factors were incorporated as control variables in our analysis and encompassed three main domains: sociodemographic characteristics, socioeconomic status, and health status. Sociodemographic variables included age (categorized as 60–70 years = 1, 70–79 years = 2, and ≥ 80 years = 3), sex (male = 1, female = 2), and marital status (married = 1, unmarried/separated/divorced/widowed = 2). Socioeconomic status was assessed based on educational level (primary and below = 1, middle school = 2, high school and above = 3) and place of residence (urban = 1, urban-rural = 2, rural = 3). Health status was evaluated through indicators of smoking (yes = 1, no = 2), alcohol consumption (yes = 1, no = 2), and self-rated health (very bad = 0, bad = 1, fair = 2, good = 3, very good = 4). Statistical analysis All data are sourced from the China Health and Retirement Longitudinal Study database. After cleaning using Stata 17.0 software, the data were entered and organized in Excel 2019 before being imported into SPSS 26.0 statistical software for analysis. First, frequency statistics and composition ratios are used to describe qualitative information, while quantitative information is statistically described. Second, to further explore the mechanism underlying the association between physical activity and depressive symptoms, this study employed Pearson correlation analysis and one-way ANOVA to construct a chained mediation model. Physical activity served as the independent variable, sleep duration and cognitive function as mediating variables, and depressive symptoms as the dependent variable. This model elucidated the role of sleep duration and cognitive function in mediating the association between physical activity and depressive symptoms in older adults. Finally, the chained mediation effect analysis was conducted using the PROCESS macro in SPSS 26.0 software, with Model 6 selected. A deviation-corrected nonparametric percentile bootstrap method was employed to perform 5,000 repeated samples, calculating the 95% confidence interval for the mediating effect. If the confidence interval did not include zero, it indicated a significant mediating effect of the mediating variable. If the 95% confidence interval included zero, it indicated that the mediating effect of the mediating variable was not significant. The path analysis-based approach using PROCESS was selected for its specific suitability in testing the pre-specified simple and chain mediation effects within a regression framework, providing robust bootstrap confidence intervals for the indirect pathways of primary theoretical interest. P < 0.05 difference is statistically significant. Result Characteristics of the participants Figure 2 shows the flowchart for the screening of the subjects of this study. A final total of 4744 participants aged 60 years and older were included for mediation analyses. Of the total participants, 1990 older adults suffered from depressive symptoms (42%), and 2753 did not exhibit depressive symptoms. Among them, the prevalence of depressive symptoms was higher among older people who were male, aged 60–69, living in rural areas, married, with primary school education or less, non-smokers, non-drinkers, and with average self-rated health. Physical activity levels in older adults are more concentrated at high physical activity levels, leading to a higher prevalence of depressive symptoms in older adults with high physical activity levels. The prevalence of depressive symptoms was higher in the < 6 h and > 9 h sleep duration groups, with sleep deprivation and excessive sleep leading to an increased risk of depression. The high prevalence of mild and severe cognitive impairment suggests that cognitive impairment is an important factor in developing depressive symptoms. The basic characteristics of the study population are shown in Table 1. Table 1Basic characteristics of the study populationVariablesTotal(n=4744)Depressive symptom P Yes n1991(42.0%No n2753(58.0%)GenderMale2836(59.8%)1081(54.3%)1755(63.7%)<0.001Female1908(40.2%)910(45.7%)998(36.3%)Age (years)60-693201(67.5%)1339(67.3%)1862(67.6%)0.83270-791329(28.0%)558(28.0%)771(28.0%)≥ 80214(4.5%)94(4.7%)120(4.4%)ResidenceTown1263(26.6%)390(19.6%)873(31.7%)<0.001Urban and rural areas392(8.3%)145(7.3%)247(9.0%)Rural3089(65.1%)1456(73.1%)1633(59.3%)Marital statusMarried4023(84.8%)1646(82.7%)2377(86.3%)<0.001Single721(15.2%)345(17.3%)376(13.7%)Educational attainmentPrimary and below2857(60.2%)1370(68.8%)1487(54.0%)<0.001Junior high school1129(23.8%)408(20.5%)721(26.2%)High school and above758(16.0%)213(10.7%)545(19.8%)SmokingYes2433(51.3%)965(48.5%)1468(53.3%)0.001No2311(48.7%)1026(51.5%)1285(46.7%)DrinkingYes1858(39.2%)687(34.5%)1171(42.5%)<0.001No2886(60.8%)1304(65.5%)1582(57.5%)Self-assessed healthVery good.497(10.5%)98(4.9%)399(14.5%)<0.001Good596(12.6%)162(8.1%)434(15.8%)Fair2495(52.6%)967(48.6%)1528(55.5%)Not good922(19.4%)588(29.6%)334(12.1%)Very bad234(4.9%)176(8.8%)58(2.1%)Physical activity levelLow intensity343(7.2%)220(11.0%)123(4.5%)<0.001Medium intensity1585(33.4%)569(28.6%)1016(36.9%)High intensity2816(59.4%)1202(60.4%)1614(58.6%)Sleep duration (h)< 6737(15.5%)475(23.9%)262(9.5%)<0.0016-7724(15.3%)318(16.0%)406(14.7%)7-81089(22.9%)407(20.4%)682(24.8%)8-9949(20.0%)343(17.2%)606(22.0%)≥ 91245(26.3%)448(22.5%)797(29.0%)Cognitive functionNormalcy36(0.7%)11(0.5%)25(0.9%)<0.001Mild disorder2851(60.1%)1003(50.4%)1848(67.1%)Moderate cognitive impairment1783(37.6%)914(45.9%)869(31.6%)Severe cognitive impairment74(1.6%)63(3.2%)11(0.4%) Basic characteristics of the study population Correlation analysis Physical activity, sleep duration, cognitive function, and depressive symptom data were analyzed by Pearson correlation in older adults. The results in Figure 3 show that there was a significant positive correlation between physical activity and sleep duration and cognitive function (P < 0.01), and there was also a significant positive correlation between sleep duration and cognitive function (P < 0.01), and there was a significant negative correlation between physical activity and depressive symptoms (P < 0.01), and there was also a significant correlation between sleep duration and depressive symptoms (P < 0.01), and there was also a significant correlation between cognitive function and depressive symptoms (P < 0.01). The results suggest that further mediation effects analyses can be conducted. Correlation analysis of variables Analysis of variances To further explore the mechanisms underlying the role of physical activity in influencing depression and cognitive functioning in older adults. This study examined the effects of different intensities of physical activity on depressive symptoms and cognitive functioning in older adults through one-way ANOVA. The results showed that the more physically active the older adults were, the fewer depressive symptoms and better cognitive function they had. The medium- and high-intensity activity group had significantly fewer depressive symptoms and better cognitive function than the low-intensity group, with significant differences (P < 0.001). Provides solid evidence for the test of mediating effects (as shown in Table 2). Table 2Analysis of variancesVariablesLow physical activity group(n = 343)Moderate physical activity group(n = 1585)High physical activity group(n = 2816)F P Depressive symptom12.63 ± 6.858.53 ± 5.989.31 ± 6.3560.24< 0.001Cognitive function18.63 ± 5.7821.40 ± 3.1920.97 ± 3.3287.58< 0.001 Analysis of variances Mediation effect test As can be seen from the regression results in Table 3, physical activity was a significant positive predictor of sleep duration (β = 0.146, t = 4.523, P < 0.001); Physical activity was a significant positive predictor of cognitive function (β = 0.435, t = 5.262, P < 0.001); Physical activity was a significant negative predictor of depressive symptoms (β = −0.285, t = −2.022, P < 0.05); Sleep duration was a significant positive predictor of cognitive function (β = 0.183, t = 4.940, P < 0.001); Sleep duration was a significant negative predictor of depressive symptoms (β = −0.760, t = −12.006, P < 0.001); Cognitive functioning was a significant negative predictor of depressive symptoms (β = −0.397, t = −16.075, P < 0.001). Table 3Regression analysis of physical activity on depressive symptomsVariablesSleep durationCognitive functionDepressive symptomβtβtβtPhysical activity0.1464.523***0.4355.262***−0.285−2.022*Sleep duration0.1834.940***−0.760−12.006***Cognitive function−0.397−16.075***R0.0660.1080.294R20.0040.0120.087F20.453***27.865***149.818****P < 0.05; ***P < 0.001 Regression analysis of physical activity on depressive symptoms *P < 0.05; ***P < 0.001 The results of the mediated effects analysis in Table 4 show that the direct effect value of physical activity on depressive symptoms was − 0.285, and the direct effect did not contain 0 at the 95% confidence interval (95% CI [−0.562, −0.009]), indicating that the direct effect of physical activity on depressive symptoms was significant, with an effect value of 49.2% of the total effect. Physical activity produced a total effect value of −0.579 for depressive symptoms with a 95% confidence interval not including 0 (95% CI [−0.867, −0.292]), indicating a significant total effect of physical activity on depressive symptoms. Table 4Analysis of chained intermediation effectsEffectBoot SEBoot LLCLBoot ULCLEfficiency ratio(%)Direct effect−0.2850.141−0.562−0.00949.2Physical activity → sleep duration → depressive symptoms (indirect effect 1)−0.1110.027−0.166−0.06119.1Physical activity → cognitive function → depressive symptoms (indirect effect 2)−0.1730.043−0.258−0.08829.8Physical activity → sleep duration → cognitive function → depressive symptoms (indirect effect 3)−0.0110.004−0.019−0.0041.9Total indirect effect−0.2940.057−0.409−0.18250.8Total effect−0.5790.147−0.867−0.292100 Analysis of chained intermediation effects From the results of the chained mediation model analyses in Fig. 4; Table 4, it can be concluded that physical activity negatively predicted depressive symptoms and that sleep duration and cognitive functioning mediated the effect, with a total indirect effect value of −0.294, with 95% confidence intervals that did not include 0 (95% CI [−0.409, −0.182]), which accounted for 50.8% of the total effect. Physical activity affected depressive symptoms in older adults mainly through the following 3 mediating pathways: physical activity → sleep duration → depressive symptoms (indirect effect 1), with an indirect effect value of −0.111, 95% CI [−0.166, −0.061], accounting for 19.1% of the total effect; physical activity → cognitive function → depressive symptoms (indirect effect 2), with an indirect effect value of −0.173, 95% CI [−0.258, −0.088], accounting for 29.8% of the total effect; physical activity → sleep duration → cognitive function → depressive symptoms (indirect effect 3), with an indirect effect value of −0.011, 95% CI [−0.019, −0.004], accounting for 1.9% of the total effect. The 95% confidence intervals for all 3 mediating effects did not include 0, indicating that the pathway mediation effect was significant. It further indicates that research hypothesis H4 is valid. The chain mediation model is shown in Fig. 4. Chained mediation effect model Discussion Association between physical activity and depressive symptoms The results of this study indicate a significant negative association between physical activity and depressive symptoms in older adults, which aligns with existing literature [16, 45]. Higher levels of physical activity are consistently associated with fewer depressive symptoms [46, 47]. The mechanisms by which physical activity affects depression are mainly categorized into physiological and psychological aspects. In terms of physiological mechanisms, physical activity is primarily responsible for modulating cytokine production, increasing neurotransmitter release, inhibiting inflammatory pathways, maintaining mitochondrial function, decreasing apoptosis in hippocampal neuronal cells, and mediating microRNA expression, which improves depressive symptoms [48]. Psychologically, engagement in physical activity fosters a sense of achievement, self-efficacy, and self-worth, while also serving as a distraction from daily stressors, thereby improving mood and reducing depressive symptoms [49, 50].The improvement of depressive symptoms by physical activity is not the result of the action of a single mechanism, but a process of interaction and synergistic regulation of multiple mechanisms that work together to effectively inhibit or slow down the development of depressive symptoms. Therefore, encouraging and supporting older people to participate in an appropriate amount of physical activity is important for improving their mental health and quality of life. The mediating role of sleep duration This study found that sleep duration partially mediated the association between physical activity and depressive symptoms, accounting for 19.1% of the total effect. This suggests that physical activity not only directly reduces depressive symptoms but also does so indirectly by improving sleep duration. Regular physical activity is known to enhance sleep quality and optimize sleep patterns, particularly in older adults who often experience sleep disturbances [51, 52]. Both insufficient sleep (< 6 h) and excessive sleep (≥ 9 h) were associated with a higher risk of depressive symptoms in our sample, consistent with previous findings that sleep duration has a U-shaped association with depression [53–56]. Therefore, appropriate and regular physical activity may help regulate sleep duration, thereby improving mental health in older adults. It is recommended that the older population with poorer sleep be actively involved in appropriate physical activities, thereby reducing the appearance of depressive symptoms. Therefore, sleep duration, as a bridge between physical activity and depressive symptoms in older adults, is an important factor in establishing good mental health. The mediating role of cognitive functions The results demonstrated that cognitive function served as a significant mediator between physical activity and depressive symptoms in older adults, which aligns with previous research [57, 58]. Regular physical activity was positively associated with cognitive performance, helping to preserve cognitive abilities and slow age-related decline [59, 60]. Physical activity may improve cognitive function in several ways, such as promoting neurogenesis, angiogenesis, and synaptic plasticity; reducing inflammatory responses; and reducing oxidative stress-induced cellular damage [61]. Furthermore, better cognitive functioning contributes to mental health by strengthening emotional regulation, building stress resilience, and enhancing self-confidence, thereby reducing vulnerability to depressive symptoms. These findings underscore the role of cognitive function as a protective factor in the mental health of older adults, partially explaining how physical activity reduces depressive symptoms. Chain intermediation The results of this study showed that the total effect value of physical activity to improve depressive symptoms in older adults was − 0.579. Among them, the effect value of physical activity directly negatively predicting depressive symptoms was − 0.285, i.e., the total indirect effect value was − 0.294, indicating that the amount of mediated effect was larger than the amount of direct effect, in which the 3rd chained mediated pathway accounted for 1.9% of the total effect, i.e., Physical Activity→Sleep Duration→Cognitive Functioning→Depressive Symptoms, and the study proved that Hypothesis 4 was established. Physical activity may not only significantly and negatively predict depressive symptoms in older adults through the separate mediation of sleep duration and cognitive functioning but may also jointly influence depressive symptoms in older adults through the chained mediation of sleep duration and cognitive functioning. On the one hand, there is a certain correlation between sleep duration and cognitive function, and previous studies have confirmed that regular and sufficient sleep duration can enhance an individual’s cognitive function, especially the consolidation of memory and the repair of the brain, which have important roles [62]. On the other hand, older adults participate in social activities in the form of physical activity by shifting to new social roles, integrating into new social circles [63, 64], increasing positive emotional support and interpersonal interactions, relieving stress, improving the quality of sleep, optimizing the duration of sleep, and improving cognitive functioning [65], which in turn affects the level of perceived health of the individual and ultimately affects the depressive symptoms of older adults [66]. At the same time, the results of the study reflect the virtuous circle of “individual-physical health-mental health,” emphasizing the mutually beneficial association between the individual and physical and mental health. This further confirms that older people’s social participation through physical activity not only improves their physical and mental health but also allows them to better adapt to social changes, integrate into the new socialization process, and enhance their quality of life and sense of well-being. Limitations and outlook First, due to limitations in data availability and model complexity, this study established baseline associations using only cross-sectional data from the 2018 wave and cannot establish causality. Future research should employ longitudinal or experimental study designs to determine causality. Second, all data in this study were derived from voluntary retrospective reporting, which may be subject to recall bias or reporting bias. Although the five-category sleep duration classification and MMSE cognitive threshold serve distinct purposes for detailed analysis and localized adaptation, both exhibit discrepancies from universal clinical standards. This inconsistency may compromise the direct comparability of results, necessitating caution when interpreting findings from studies conducted over specific time periods. Future research should also explore integrating other measurement methods, such as using accelerometers to measure physical activity levels and employing sleep monitors to record sleep duration, in order to obtain more comprehensive and objective results. Finally, while sleep duration was included as a mediator in our model, our data suggest its association with depressive symptoms and cognitive function may be non-linear (U-shaped). Treating it as a linear variable in the mediation analysis, while statistically conventional, may oversimplify this complex association. Future research using nonlinear mediation models could provide further insights into these intricate associations. Conclusion Significant correlations were found between physical activity, sleep duration, cognitive function, and depressive symptoms. Significant correlations were found between physical activity, sleep duration, cognitive function, and depressive symptoms. Physical activity significantly and negatively predicts depressive symptoms in older adults and is an important intervening variable for improving depression in older adults. Physical activity significantly and negatively predicts depressive symptoms in older adults and is an important intervening variable for improving depression in older adults. Sleep duration and cognitive functioning independently mediate the association between physical activity and depressive symptoms in older adults. Sleep duration and cognitive functioning independently mediate the association between physical activity and depressive symptoms in older adults. Sleep duration and cognitive functioning act as chain mediators between physical activity and depressive symptoms in older adults, i.e., physical activity can improve sleep duration in order to enhance cognitive functioning and, in turn, improve depressive symptoms in older adults. Sleep duration and cognitive functioning act as chain mediators between physical activity and depressive symptoms in older adults, i.e., physical activity can improve sleep duration in order to enhance cognitive functioning and, in turn, improve depressive symptoms in older adults. Acknowledgements We thank the China Health and Retirement Longitudinal Study (CHARLS) team for providing data. Thanks, are also due to the CHARLS participants for their contributions. Authors’ contributions Tongnian Yang: Conceptualization, Methodology, Formal analysis, Writing – Original Draft.Jie Ren: Supervision, Writing – Review & Editing, Project administration.Hui Xiao: Data Curation, Validation, Visualization.All authors read and approved the final manuscript. Funding This study received no external funding. Data availability The datasets used and analyzed in this study are publicly available from the China Health and Retirement Longitudinal Study (CHARLS) repository, hosted by Peking University. The data can be accessed at: http://charls.pku.edu.cn/en. Users must register and agree to CHARLS’ terms of use before accessing the data. Declarations Ethics approval and consent to participate Ethical approval for this study was obtained from the Ethics Review Board of Peking University (Approval No. IRB00001052-11015). The study was conducted in accordance with the ethical principles of the Declaration of Helsinki. Informed consent was obtained from all participants prior to their inclusion in the study. All methods and experimental protocols were performed in compliance with relevant guidelines and regulations and were approved by the Institutional Review Board of Peking University. Consent for publication Not applicable. This study uses secondary data from the China Health and Retirement Longitudinal Study (CHARLS), a de-identified, publicly available dataset. No individual person’s data (including personal details, images, or videos) are included in this manuscript, and therefore consent for publication is not required. Competing interests The authors declare no competing interests. Footnotes The original version of this article was revised: the authors identified an error in Table 1 and Figure 2. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Change history 2/3/2026 A Correction to this paper has been published: 10.1186/s40359-026-04043-1 References Associated Data Data Availability Statement The datasets used and analyzed in this study are publicly available from the China Health and Retirement Longitudinal Study (CHARLS) repository, hosted by Peking University. The data can be accessed at: http://charls.pku.edu.cn/en. Users must register and agree to CHARLS’ terms of use before accessing the data.