Abstract
Background
Falls are a leading cause of injury and mortality in older adults, with a rapidly increasing burden in China (Zhang et al., 2022) [1]. Although short sleep duration has been linked to falls, evidence from large prospective studies remains limited, and potential mechanisms are not fully understood (Zhu et al., 2022; Zhang et al., 2025) [2,3].
Methods
We analyzed 11,476 participants aged ≥45 years from the China Health and Retirement Longitudinal Study (CHARLS, 2011–2018). Restricted cubic splines (RCS) were used to examine the non-linear association between sleep duration and fall risk. Cox proportional hazards models were applied to estimate the association between sleep duration and incident falls during follow-up. Multiple mediation analysis was conducted to quantify the mediating roles of physical function, chronic disease burden, cognitive function, and instrumental activities of daily living (IADL).
Results
Short sleep duration (<6.5 h) was associated with a higher risk of incident falls (Adjusted HR = 1.14; 95% CI, 1.07–1.22; P < 0.001). RCS analyses suggested a non-linear relationship, with increased risk mainly at shorter sleep durations, while no significant association was observed at longer durations. Mediation analysis indicated that this association was partly explained by physical function (25.2%), chronic disease burden (14.2%), cognitive function (8.2%), and IADL dependence (5.4%).
Conclusions
Short sleep duration (<6.5 h) was associated with an increased risk of falls among middle-aged and older adults in China. This association may be partly mediated by impaired physical function and greater chronic disease burden, underscoring the importance of incorporating sleep health into multifactorial fall-prevention strategies.
Keywords: Sleep duration, Fall risk, Middle-aged and older adults, Mediation analysis, Chronic disease burden, Physical function
Highlights
-
•
Short sleep duration (<6.5 h) was associated with higher incident fall risk in CHARLS.
-
•
Physical function mediated the largest proportion of the sleep–fall association (25.2%).
-
•
Chronic disease burden, cognitive function (MMSE), and instrumental activities of daily living (IADL) also partially mediated the association.
-
•
Sleep health may be integrated into multifactorial fall-prevention strategies in older adults.
1. Introduction
Falls represent a significant public health challenge for older adults globally, leading to injury, disability, and mortality [4]. In China, the aging population exacerbates this issue, with national data indicating that falls are the primary cause of injury-related fatalities among individuals aged 65 and above [1]. A meta-analysis of 23 studies revealed a high incidence of falls among individuals aged 60 and older in China, reaching 18.3%, with 37.2% resulting in moderate to severe injuries [5]. The repercussions of falls extend beyond medical costs, including fractures, head injuries, long-term disabilities, and psychological consequences such as fear and depression. These outcomes significantly diminish the quality of life for older adults, imposing substantial burdens on families and social welfare systems [6].
Interventions targeting fall risk factors have increasingly focused on sleep duration, a factor prevalent among older adults and linked to adverse health outcomes such as cardiovascular disease, cognitive impairment, and all-cause mortality [7,8]. Emerging evidence suggests a significant association between sleep duration and fall risk [9], with studies demonstrating a "U" or "J" shaped relationship, indicating increased fall risk with both insufficient (<5–6 h) and excessive (>8–9 h) sleep durations [2,3,10]. However, existing research, primarily cross-sectional, has not established a causal link between sleep duration and falls, limiting a comprehensive understanding of this association. Moreover, the pathways through which sleep duration influences fall risk remain unclear, underscoring the need to identify mediating mechanisms to inform targeted intervention strategies.
Drawing from existing biological evidence, we propose several potential mediation pathways. First, sleep deprivation may directly increase fall susceptibility by inducing daytime fatigue, reduced balance, and weakened muscle strength [11]. Second, sleep disorders often coincide with chronic conditions (e.g., diabetes, hypertension) that have been identified as independent risk factors for falls [12]. Third, sleep deprivation may impair concentration, diminish executive function, and delay reflexes, ultimately impacting overall bodily function and raising fall risk [13]. Finally, limitations in instrumental activities of daily living (IADL), which reflect challenges in performing complex daily tasks and often require increased assistance, serve as a key indicator of independent living. Impairments in IADL are commonly associated with cognitive decline, which, in turn, alters activity patterns and environmental interactions, indirectly heightening the risk of falls [14]. However, the absence of nationally representative, large-scale longitudinal studies hinders the simultaneous exploration of these mediation pathways.
This study aims to systematically investigate the association between sleep duration and future fall risk using nationally representative CHARLS data from the 2011 baseline to the 2018 follow-up. Additionally, it seeks to examine the mediating effects of physical functioning, chronic disease burden, cognitive function (MMSE), and activities of daily living (IADL). This study aims to examine the association between sleep duration and fall risk in middle-aged and older adults and to investigate the underlying mechanisms. The findings are expected to inform the design of multifaceted fall prevention strategies that integrate sleep management and physical function improvement.
2. Methods
2.1. Data sources and subjects
This study utilized data from the China Health and Retirement Longitudinal Study (CHARLS), a nationally representative prospective cohort study targeting individuals aged 45 and older in China. CHARLS conducted its baseline recruitment across 150 county units, comprising 450 villages/settlements in 28 provinces, using a multistage stratified probability sampling approach. Detailed information regarding survey methodology, implementation, and data collection can be found in the relevant literature [15]. Approval for the study protocol was obtained from the Biomedical Ethics Committee of Peking University (IRB00001052-11015), and written informed consent was obtained from all participants.Beginning with the 2011 National Baseline Survey, this study addressed the significant loss of fall outcome data in 2020 by utilizing follow-up data from 2013, 2015, and 2018, ensuring the reliability and comprehensiveness of the outcome variables.
Baseline data for this investigation were derived from the 2011 survey, which initially enrolled 17,705 individuals. Following the application of exclusion criteria, 5298 participants were excluded due to incomplete essential baseline information (e.g., age, gender, height, weight) and missing data pertinent to the sarcopenia assessment. Subsequently, 348 individuals who did not meet the inclusion criteria—including those under 45 years of age, with atypical height, or extreme weight—were excluded. Additionally, 110 participants with a self-reported history of cancer were excluded. Lastly, 473 individuals exhibiting extreme sleep durations (<1 h), missing sleep records, or incomplete data on mediators (e.g., physical function scores, MMSE, IADL) were excluded. The final sample for analysis comprised 11,476 participants (see Fig. 1).
Fig. 1.
Flowchart of the study.
2.2. Study variables and assessment criteria
2.2.1. Main independent variable
The primary independent variable in this study was sleep duration, assessed through the question: "How many hours did you typically sleep per night in the last month?" This question was designed to capture the participants' average nightly sleep duration [16].
2.2.2. Primary outcome variable
The primary outcome variable was falls, defined as a dichotomous variable (yes/no) based on responses to the follow-up questionnaire item: "Have you experienced a fall in the preceding two years?"
2.2.3. Mediation variables
Physical function was assessed using a composite score based on six physical performance tasks, with higher scores indicating greater impairment [13]. Chronic disease burden was defined as the total count of 13 self-reported chronic conditions [17]. Cognitive function was measured using a modified Mini-Mental State Examination (MMSE) score, with higher scores indicating better cognition [18]. IADL dependence was assessed using a five-item IADL score, with higher scores indicating greater dependence [19]. Detailed variable definitions and scoring procedures are provided in Supplementary Note 1.All exposure and mediator variables including sleep duration, physical function, chronic disease count, MMSE score, and IADL score were assessed at baseline (2011). Incident falls were ascertained during follow-up surveys in 2013, 2015, and 2018.
2.2.4. Covariates
The analysis included various potential confounding factors, all of which were assessed during the 2011 baseline survey. Demographic and sociological factors included age, sex, long-term residence, highest level of education attained, marital status, self-rated living standard, annual income, and life satisfaction. Lifestyle and health-related covariates included smoking, alcohol consumption, levels of physical activity, nap habits, social engagement, daily meal frequency, and the presence of specific chronic diseases.
Additional covariates included an assessment of sarcopenia, conducted following the 2019 guidelines from the Asian Working Group on Sarcopenia (AWGS) [20,21], with specific assessment methods detailed in the supplementary Note 2. Other covariates encompassed physical disabilities, cognitive impairment, vision and hearing impairments, speech disorders, oral health, and pain assessments (including intensity and location of recurrent pain) [22]. Depressive symptoms were evaluated using the CESD-10 scale, with a score of ≥12 points indicating the presence of depressive symptoms [23,24]. The history of falls, hip fractures, and major injuries in the previous two years was also considered. The most challenging full tandem stance test from the Short Physical Performance Battery (SPPB) was administered, requiring participants to maintain a straight-line stance for a specified duration, with the duration adjusted according to age [25]. Furthermore, BMI, waist circumference, and knee height were included as continuous variables.
2.2.5. Statistical analysis
All statistical analyses were performed using R software (version 4.3.3). Continuous variables are presented as mean ± standard deviation or median (interquartile range), while categorical variables are expressed as frequency (percentage). Group comparisons were conducted using Student's t-test, Mann-Whitney U test, or chi-square test, depending on the variable type. Restricted Cubic Spline (RCS) analysis was used to examine the nonlinear relationship between sleep duration, age, physical function score, and future fall risk [26]. The analysis identified 6.5 h as the cutoff for sleep duration, classifying participants into a short sleep group (<6.5 h) and a normal sleep group (≥6.5 h).Kaplan-Meier cumulative fall incidence curves were generated, and Log-rank tests were used to compare group differences. For the construction of a concise and robust multivariate Cox regression model, a 10-fold cross-validation Lasso-penalized Cox regression was applied to select variables from 55 baseline covariates and sleep duration groups [27]. Thirteen predictors with nonzero coefficients were included in the final model.The association between sleep duration and fall risk was evaluated using both univariate and multivariate Cox proportional hazards regression models, with results presented as hazard ratios (HR) and 95% confidence intervals (CI). To assess the robustness and heterogeneity of this association, extensive one-way subgroup analyses were conducted. Additionally, to evaluate effect modification, P for interaction was calculated by introducing an interaction term. Given that approximately 25% of participants had an MMSE score ≤10 (indicative of severe cognitive impairment) and 75% had a score ≤18 (indicative of moderate cognitive impairment) according to baseline data in Table 1, and considering that cognitively impaired participants may have provided unreliable sleep data, we performed sensitivity analyses. In these analyses, we sequentially excluded participants with more severe cognitive impairment, using MMSE cutoffs of ≤10 (severe impairment) and ≤20 (mild to moderate impairment or significant deficit) [28,29]. The association between sleep duration and fall risk was re-evaluated using Cox proportional hazards regression models to test the robustness of the findings.
Table 1.
Baseline characteristics of the normal sleep group and short sleep group.
| Variables | Overall |
Normal Sleep (≥6.5h) |
Short Sleep (<6.5h) |
p-value |
|---|---|---|---|---|
| N = 11,476 | N = 5758 | N = 5718 | ||
| Sleep duration, (h) | 6.40 ± 1.84 | 7.87 ± 0.96 | 4.92 ± 1.22 | <0.001 |
| Age, (year) | 58.81 ± 9.13 | 57.97 ± 9.03 | 59.65 ± 9.15 | <0.001 |
| MMSE, (p25 - p75) | 14.00 (10.00 - 18.00) | 14.50 (10.00 - 18.00) | 14.00 (9.00 - 18.00) | <0.001 |
| BMI, (p25 - p75) | 23.09 (20.82 - 25.67) | 23.28 (21.06 - 25.86) | 22.90 (20.59 - 25.47) | <0.001 |
| Physical Function Score, (p25 - p75) | 9.00 (7.00 - 12.00) | 8.00 (7.00 - 11.00) | 10.00 (7.00 - 12.00) | <0.001 |
| Sex, n (%) | <0.001 | |||
| Male | 5621 (48.98%) | 2914 (50.61%) | 2707 (47.34%) | |
| Female | 5855 (51.02%) | 2844 (49.39%) | 3011 (52.66%) | |
| Physical Disabilities, n (%) | 0.06 | |||
| Yes | 341 (2.97%) | 154 (2.67%) | 187 (3.27%) | |
| No | 11,135 (97.03%) | 5604 (97.33%) | 5531 (96.73%) | |
| Brain damage/mental retardation, n (%) | 0.953 | |||
| Yes | 280 (2.44%) | 140 (2.43%) | 140 (2.45%) | |
| No | 11,196 (97.56%) | 5618 (97.57%) | 5578 (97.55%) | |
| Vision problem, n (%) | 0.005 | |||
| Yes | 716 (6.24%) | 323 (5.61%) | 393 (6.87%) | |
| No | 10,760 (93.76%) | 5435 (94.39%) | 5325 (93.13%) | |
| Hearing problem, n (%) | 0.048 | |||
| Yes | 943 (8.22%) | 444 (7.71%) | 499 (8.73%) | |
| No | 10,533 (91.78%) | 5314 (92.29%) | 5219 (91.27%) | |
| Speech impediment, n (%) | 0.249 | |||
| Yes | 28 (0.24%) | 11 (0.19%) | 17 (0.30%) | |
| No | 11,448 (99.76%) | 5747 (99.81%) | 5701 (99.70%) | |
| Number of Chronic Conditions, n (%) | <0.001 | |||
| 0 | 3784 (32.97%) | 2123 (36.87%) | 1661 (29.05%) | |
| 1 | 3496 (30.46%) | 1818 (31.57%) | 1678 (29.35%) | |
| 2 | 2197 (19.14%) | 1028 (17.85%) | 1169 (20.44%) | |
| 3 | 1115 (9.72%) | 468 (8.13%) | 647 (11.32%) | |
| 4 | 527 (4.59%) | 199 (3.46%) | 328 (5.74%) | |
| 5 | 241 (2.10%) | 82 (1.42%) | 159 (2.78%) | |
| 6 | 80 (0.70%) | 29 (0.50%) | 51 (0.89%) | |
| 7 | 26 (0.23%) | 8 (0.14%) | 18 (0.31%) | |
| 8 | 9 (0.08%) | 3 (0.05%) | 6 (0.10%) | |
| 9 | 1 (0.01%) | 0 (0.00%) | 1 (0.02%) | |
| Traffic Accident or Major Accidental Injury, n (%) | 0.978 | |||
| Yes | 1113 (9.70%) | 558 (9.69%) | 555 (9.71%) | |
| No | 10,363 (90.30%) | 5200 (90.31%) | 5163 (90.29%) | |
| Fallen Down, n (%) | <0.001 | |||
| Yes | 1781 (15.52%) | 733 (12.73%) | 1048 (18.33%) | |
| No | 9695 (84.48%) | 5025 (87.27%) | 4670 (81.67%) | |
| Fractured Hip, n (%) | 0.052 | |||
| Yes | 162 (1.41%) | 69 (1.20%) | 93 (1.63%) | |
| No | 11,314 (98.59%) | 5689 (98.80%) | 5625 (98.37%) | |
| Sarcopenia Status, n (%) | <0.001 | |||
| No Sarcopenia | 5360 (46.71%) | 2873 (49.90%) | 2487 (43.49%) | |
| Possible Sarcopenia | 4430 (38.60%) | 2188 (38.00%) | 2242 (39.21%) | |
| Sarcopenia | 1303 (11.35%) | 541 (9.40%) | 762 (13.33%) | |
| Severe sarcopenia | 383 (3.34%) | 156 (2.71%) | 227 (3.97%) | |
| Balance ability test, n (%) | <0.001 | |||
| No | 2522 (21.98%) | 1184 (20.56%) | 1338 (23.40%) | |
| Yes | 8954 (78.02%) | 4574 (79.44%) | 4380 (76.60%) | |
| Depression, n (%) | <0.001 | |||
| No | 8375 (72.98%) | 4642 (80.62%) | 3733 (65.29%) | |
| Yes | 3101 (27.02%) | 1116 (19.38%) | 1985 (34.71%) | |
| IADL, n (%) | <0.001 | |||
| 0 | 10,317 (89.90%) | 5267 (91.47%) | 5050 (88.32%) | |
| 1 | 732 (6.38%) | 339 (5.89%) | 393 (6.87%) | |
| 2 | 260 (2.27%) | 84 (1.46%) | 176 (3.08%) | |
| 3 | 102 (0.89%) | 42 (0.73%) | 60 (1.05%) | |
| 4 | 44 (0.38%) | 19 (0.33%) | 25 (0.44%) | |
| 5 | 21 (0.18%) | 7 (0.12%) | 14 (0.24%) | |
| Exercise intensity, n (%) | 0.711 | |||
| No exercising | 7030 (61.26%) | 3545 (61.57%) | 3485 (60.95%) | |
| Walking | 1102 (9.60%) | 536 (9.31%) | 566 (9.90%) | |
| Do moderate activities | 1522 (13.26%) | 769 (13.36%) | 753 (13.17%) | |
| Do vigorous activities | 1822 (15.88%) | 908 (15.77%) | 914 (15.98%) |
Note:Medians and interquartile ranges (25th and 75th percentiles) were calculated for continuous variables, and frequencies and percentages were used for categorical variables. Abbreviations: BMI (Body Mass Index), MMSE (Mini-Mental State Examination), IADL (Instrumental Activities of Daily Living). Physical Function Score, Number of Chronic Conditions, MMSE, IADL, Sarcopenia Status, Balance Ability Test, Depression, and Exercise Intensity are evaluated and defined in detail in the Methods section.Number of Chronic Conditions was calculated by summing the presence of 13 conditions: hypertension, dyslipidemia, diabetes, chronic lung disease, liver disease, heart disease, stroke, kidney disease, digestive disease, psychiatric problems, memory-related disease, arthritis/rheumatism, and asthma.
To investigate the underlying mechanisms, we conducted a multiple mediation analysis, treating sleep duration as a continuous variable to explore indirect effects across four pathways: physical function, chronic disease burden, cognitive function (MMSE), and IADL. The indirect effects were evaluated through linear regression for path a (sleep duration → mediator) and Cox regression for path b (mediator → fall risk). The proportion of indirect effects was calculated using the product method, and statistical significance was assessed using the Bootstrap method with 1000 resamples [30,31].All statistical analyses were two-sided, with statistical significance set at P < 0.05.
3. Results
3.1. Baseline characteristics
Participants were categorized by sleep duration into a normal sleep group (≥6.5 h, n = 5758) and a short sleep group (<6.5 h, n = 5718) based on Restricted Cubic Spline (RCS) analysis (Fig. 2). As detailed in Table 1, the two groups differed significantly across most baseline characteristics.The remaining baseline characteristics are provided in Supplementary Table S1.
Fig. 2.
Sleep Duration and Fall Risk: RCS Analysis Note:This plot shows the relationship between sleep duration and fall risk based on RCS analysis.
Compared to the normal sleep group, individuals with short sleep were generally older, more likely to be female, and reported lower socioeconomic status (e.g., education, life satisfaction, and self-rated living standard). Their health profile was notably poorer, marked by a higher burden of chronic conditions, more severe physical limitations, and a greater prevalence of pain, depressive symptoms, and sarcopenia. Functionally, the short sleep group also demonstrated lower performance in balance tests and instrumental activities of daily living (IADL), and had a higher incidence of prior falls at baseline.
3.2. Kaplan-Meier analysis of incident falls by sleep duration group
The Kaplan-Meier analysis revealed a significant association between sleep duration and fall risk after approximately 7 years of follow-up (log-rank test, p < 0.0001). Survival curves for individuals with insufficient sleep consistently showed lower values compared to those with normal sleep patterns, indicating a higher cumulative incidence of falls. The divergence between the two curves was apparent early in the follow-up period (2013) and continued to widen over time, suggesting a more rapid increase in fall risk among the sleep-deprived group. Risk table analysis indicated a consistently lower number of participants available for follow-up in the sleep deficit group compared to the normal sleep group at each time point. In conclusion, the Kaplan-Meier analysis strongly supports the notion that shorter sleep duration is a significant risk factor for an increased likelihood of future falls (see Fig. 3).
Fig. 3.
Kaplan-Meier Curves for Incident Falls by Sleep Duration Group Note:This plot shows the Kaplan-Meier curves for incident falls by sleep duration group. The short sleep group (<6.5 h) has a significantly higher fall risk compared to the normal sleep group (≥6.5 h) (log-rank test, P < 0.0001). The curves diverge over time, indicating a faster increase in fall risk for those with insufficient sleep. The number at risk table shows fewer participants in the short sleep group at each time point.
3.3. Covariate screening and Lasso-penalized Cox regression results
To construct a multivariable Cox regression model while avoiding overfitting, we performed Lasso regression on 55 baseline covariates together with the sleep-duration grouping. The optimal regularization parameter (λ = 0.0228) was selected via 10-fold cross-validation. The Lasso procedure retained 13 predictors with non-zero coefficients: marital status, sex, arthritis/rheumatism, history of falls in the past two years, pain level, self-rated living standard, age, sarcopenia status, depressive symptoms, lower-limb pain, pain distribution, sleep group, and physical function score (see Supplementary Fig. S1A and S1B). These variables were carried forward into the subsequent multivariable Cox regression analyses.
3.4. Univariate and multivariate Cox regression analysis
Both univariate and multivariate Cox regression analyses revealed an independent association between shorter sleep duration (<6.5 h) and an elevated risk of falls during the follow-up period (HR = 1.14; 95% CI, 1.065–1.219; P < 0.001). RCS analysis categorized physical function scores into three groups: the high-function group (<9 points) as the reference, the intermediate-function group (9–13 points), and the low-function group (≥14 points) (overall P < 0.001, nonlinear P < 0.001) (see Supplementary Figure S2). Compared with the high-function group, the intermediate-function group was associated with an increased risk of falls (HR = 1.128; 95% CI, 1.042–1.221; P = 0.003), and the low-function group showed a further elevated risk (HR = 1.180; 95% CI, 1.056–1.318; P = 0.004). The findings indicated a gradual increase in fall risk with worsening physical function limitations. Other risk factors included increasing age, female sex, depressive symptoms, a history of falls, and arthritis or rheumatism. Notably, pain-related variables (including pain intensity, distribution, and leg pain) and sarcopenia were not statistically significant after multivariable adjustment (Table 2).Detailed Cox regression results are shown in Supplementary Table S2. Sensitivity analyses were conducted by separately excluding participants with MMSE scores ≤10 and ≤ 20, using multivariable Cox proportional hazards regression models, and the association between sleep duration and fall risk remained consistent (Supplementary Tables S3 and S4).
Table 2.
Univariate and multivariate Cox regression for fall risk.
| Variables | Mean (SD/)Number(%) | HR (95% CI) | P-value | Adjusted HR (95% CI) | P-value |
|---|---|---|---|---|---|
| Age | 58.81 (9.13) | 1.018(1.014 – 1.021) | 0 | 1.01(1.005 – 1.015) | 0 |
| Sex | |||||
| Male | 5621 (49.0) | ||||
| Female | 5855 (51.0) | 1.386 (1.298 – 1.481) | 0 | 1.259 (1.174 – 1.35) | 0 |
| Arthritis or rheumatism | |||||
| Yes | 3860 (33.6) | ||||
| No | 7616 (66.4) | 0.657 (0.615 – 0.702) | 0 | 0.82 (0.764 – 0.881) | 0 |
| Fallen Down | |||||
| Yes | 1781 (15.5) | ||||
| No | 9695 (84.5) | 0.432 (0.401 – 0.466) | 0 | 0.516 (0.477 – 0.558) | 0 |
| Physical Function Score Categories | |||||
| High Physical Function:<9 | 5476 (47.7) | ||||
| Moderate Physical Function: 9-13 | 4477 (39.0) | 1.483 (1.38 – 1.594) | 0 | 1.128 (1.042 – 1.221) | 0.003 |
| Low Physical Function:≥14 | 1523 (13.3) | 2.036 (1.856 – 2.232) | 0 | 1.18 (1.056 – 1.318) | 0.004 |
| Depression | |||||
| No | 8375 (73.0) | ||||
| Yes | 3101 (27.0) | 1.621 (1.514 – 1.735) | 0 | 1.14 (1.054 – 1.234) | 0.001 |
| Sleep Duration Group | |||||
| Normal Sleep (≥6.5h) | 5758 (50.2) | ||||
| Short Sleep (<6.5h) | 5718 (49.8) | 1.335 (1.25 – 1.425) | 0 | 1.14 (1.065 – 1.219) | 0 |
Note: This table presents the results of univariate and multivariate Cox regression analyses for fall risk. The multivariate model included all 13 variables selected from the univariate analysis to assess their independent associations with fall risk. Variables that were not statistically significant in the multivariate model—including pain severity, pain distribution, marital status, and self-rated standard of living—are presented in Supplementary Table S2.
3.5. Univariate Subgroup Cox Regression analysis results
Subgroup analyses assessed the consistency of the association between short sleep duration (<6.5 h) and fall risk across key population strata (Fig. 4). Age was categorized based on a restricted cubic spline–derived threshold (58 years; Supplementary Figure S3). The elevated fall risk associated with short sleep was largely consistent across most subgroups, including those defined by sex, age, depressive symptoms, and balance ability (all P for interaction >0.05). However, significant effect modification was observed for sarcopenia status (P-interaction = 0.023) and physical function (P-interaction = 0.01). Specifically, the association was attenuated and became statistically non-significant among individuals with established sarcopenia or severe sarcopenia, and similarly among those with low physical function.
Fig. 4.
Univariate Subgroup Cox Regression for Fall Risk by Sleep Duration Note:This forest plot shows the univariate Cox regression analysis for fall risk by sleep duration (<6.5 h) in subgroups. Short sleep duration significantly increases fall risk in most subgroups, including males, those without sarcopenia, and individuals with better physical function. The effect is attenuated in individuals with sarcopenia or low physical function, as indicated by significant interaction terms (P-interaction = 0.023 for sarcopenia, P-interaction = 0.01 for physical function).
3.6. Mediation analysis results
The results of the multiple mediation analysis are summarized in Table 3. The total effect of sleep duration on fall risk was significant (β = −0.155; 95% CI, −0.180 to −0.130; P < 0.001), where β denotes the standardized total effect estimated from the mediation model, suggesting that longer sleep duration is associated with a lower risk of falls. Four significant mediating pathways were identified as potential links between sleep duration and fall risk. Physical function was the strongest mediator (proportion mediated: 25.2%), followed by chronic disease burden (14.2%), cognitive function (8.2%), and IADL dependence (5.4%). In accordance with the variable specifications, a longer sleep duration was significantly associated with better status for all mediators (path a; β represents standardized linear regression coefficients): a lower physical function impairment score (β = −0.166), fewer chronic conditions (β = −0.139), a higher MMSE score (β = 0.091), and lower IADL dependence (β = −0.081) (all P < 0.001). (see Supplementary Fig. S4A–S4D).
Table 3.
Multiple mediation analysis of the association between sleep duration and fall risk.
| Mediator | Path a: Sleep → Mediator |
Path b: Mediator → Falls |
Indirect Effect (a × b) |
Proportion Mediated (95% CI) |
|---|---|---|---|---|
| β (95% CI); P-value |
HR (95% CI); P-value |
β (95% CI) | % (95% CI) | |
| (Linear Regression) | (Cox Regression) | |||
| Physical Function | −0.166 (−0.189, −0.143); <0.001 | 1.265 (1.218, 1.314); <0.001 | −0.039 (−0.045, −0.033) | 25.2% (19.8, 31.5) |
| Chronic Conditions | −0.139 (−0.162, −0.116); <0.001 | 1.171 (1.126, 1.218); <0.001 | −0.022 (−0.027, −0.017) | 14.2% (10.1, 18.9) |
| Cognitive Function (MMSE) | 0.091 (0.068, 0.114); <0.001 | 0.870 (0.838, 0.903); <0.001 | −0.013 (−0.017, −0.009) | 8.2% (5.3, 11.8) |
| IADL | −0.081 (−0.104, −0.058); <0.001 | 1.109 (1.066, 1.154); <0.001 | −0.008 (−0.011, −0.005) | 5.4% (3.2, 8.3) |
Note: Path a coefficients (β) were derived from linear regression models and represent standardized changes in each mediator per 1-SD increase in sleep duration. Path b effects are presented as hazard ratios (HRs) from Cox regression, reflecting the change in fall hazard per 1-SD increase in the mediator. Indirect effects (a × b) are reported as standardized β coefficients. Higher Physical Function and IADL scores indicate greater impairment/dependence, whereas higher MMSE scores indicate better cognitive function; a greater number of chronic conditions indicates higher disease burden. Negative β values for Physical Function, IADL, or chronic conditions suggest better health status with longer sleep, while a positive β for MMSE suggests better cognitive function.
4. Discussion
This nationally representative longitudinal study based on CHARLS examined the association between sleep duration and incident falls in middle-aged and older adults over a seven-year follow-up period. We found that short sleep duration (<6.5 h) was associated with a higher risk of future falls. RCS analyses further supported a non-linear pattern, with risk increasing mainly at shorter sleep durations, while no statistically significant association was observed at longer sleep durations (>8 h). These findings are consistent with prior evidence suggesting a “U-shaped” or “J-shaped” relationship between sleep duration and fall risk [32,33].
Multiple mediation analysis suggested that several health domains may help explain the observed association between sleep duration and falls. Physical function explained the largest proportion of the indirect effect, highlighting the potential importance of mobility and functional limitations in fall susceptibility. Poor sleep has been linked to reduced muscle strength, impaired balance, and functional decline, which may increase vulnerability to falls [34]. Sleep disruption may also affect neuromuscular recovery and coordination, thereby contributing to functional impairment [35,36]. In addition, chronic disease burden represented another notable pathway, which may reflect the cumulative impact of multimorbidity on physiological reserve and frailty [37,38]. Sleep-related alterations in metabolic and inflammatory processes may further contribute to chronic disease development and progression, potentially increasing fall risk through reduced overall resilience [39,40].
Cognitive function and IADL dependence were identified as additional mediating pathways. Prior studies have reported associations between sleep duration and cognitive decline [41], and sleep-related neurobiological changes may influence cognition and functional performance [36,42,43]. Functional impairment, particularly limitations in IADL, may increase fall susceptibility by reducing independent living capacity and affecting daily activity patterns [11,14,44]. Moreover, sleep disruption may compromise balance regulation through impaired sensory integration and motor coordination [45]. Sleep medication use may also contribute to fall risk via adverse effects such as dizziness or orthostatic hypotension [46,47]. Although these mediating pathways provide useful mechanistic hypotheses, mediators were assessed at baseline in this study; thus, the mediation findings should be interpreted as statistically compatible pathways rather than definitive causal mechanisms.
Notably, subgroup analyses demonstrated that the association between short sleep duration and falls was generally consistent across key strata such as sex, age, depressive symptoms, and balance ability. However, significant effect modification was observed for sarcopenia status and physical function. The association was attenuated and became statistically non-significant among individuals with established sarcopenia (including severe sarcopenia) and among those with poor physical function. One possible explanation is that in frailer subpopulations, underlying vulnerability and functional impairment may play a dominant role in fall risk, thereby diminishing the relative contribution of sleep duration. Clinically, these findings suggest that sleep duration may be a more informative and potentially actionable risk marker for fall prevention among individuals without advanced sarcopenia and those with preserved functional capacity, whereas frailer individuals may benefit more from comprehensive multifactorial interventions targeting mobility and overall frailty in addition to sleep optimization [48].
Several limitations should be considered. First, both exposure and outcome were self-reported. Sleep duration was assessed using a single-item question, which may introduce reporting bias compared with objective measurements. Falls were collected using a 2-year recall period without external verification, and fall frequency or severity was not captured, which may have led to outcome misclassification and potential attenuation of effect estimates. Second, although we adjusted for multiple covariates, residual confounding remains possible due to unmeasured factors such as specific sleep disorders, medication use (especially hypnotics), vestibular dysfunction, visual impairment, and environmental hazards. Third, mediation analyses were limited by the concurrent assessment of sleep duration and mediators at baseline; therefore, the identified indirect effects should be interpreted as plausible pathways rather than definitive causal mechanisms. Fourth, sleep quality could not be included due to substantial missingness. Finally, as an observational study, our findings indicate associations but do not establish causality, and further studies with repeated objective sleep measures and interventional designs are warranted.
5. Conclusions
This study suggests that short sleep duration (<6.5 h) is associated with an increased risk of incident falls among middle-aged and older adults in China. Sleep health should be incorporated into multifactorial fall-prevention strategies. This association may be partly mediated by impaired physical function and greater chronic disease burden, with additional contributions from cognitive decline and IADL dependence. Further studies are warranted to clarify temporal ordering and verify the underlying mechanistic pathways.
CRediT authorship contribution statement
Leqin Bai: Writing – original draft, Data curation, Conceptualization. Qi Kong: Writing – review & editing, Investigation, Formal analysis. Jianbo Cai: Software, Project administration, Methodology. Wenkai Wang: Visualization, Validation, Supervision. Weirong Lai: Methodology, Formal analysis, Data curation. Lijing Luo: Investigation, Formal analysis, Data curation. Qianwei Sun: Writing – review & editing, Validation, Supervision. Dongwen Su: Visualization, Supervision, Resources.
Consent to participate
All participants in this study provided written informed consent prior to their participation. The study's purpose, procedures, potential risks, and benefits were thoroughly explained to each participant in a language they understood. Participants were informed that their involvement was entirely voluntary, and they had the right to withdraw from the study at any time without penalty or loss of benefits. Confidentiality of participants' data was maintained throughout the study, and personal identifying information was anonymized during analysis and publication phases.
Ethics statement
This study was approved by the Ethics Committee of Peking University (IRB00001052-11015). Ethical review and approval were obtained in accordance with local legislation and institutional requirements. Written informed consent was obtained from all participants prior to their participation in the study, and the confidentiality of participants' data was ensured throughout the research process.
Clinical trial number
Not applicable.
Publisher's note
All statements expressed in this article are solely those of the authors and do not necessarily reflect the views of their affiliated institutions, the publisher, the editors, or the reviewers. Any product that may be evaluated in this article, or any claim made by its manufacturer, is neither guaranteed nor endorsed by the publisher.
Funding
No funding was received for this study.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
We extend our gratitude to the China Health and Retirement Longitudinal Study (CHARLS) research team, the field team, and all respondents for their valuable time and efforts dedicated to the CHARLS project.
Footnotes
Supplementary data to this article can be found online at https://doi.org/10.1016/j.sleepx.2026.100178.
Contributor Information
Leqin Bai, Email: 928720282@qq.com.
Qi Kong, Email: 4116350@qq.com.
Jianbo Cai, Email: 342933120@qq.com.
Wenkai Wang, Email: 293563483@qq.com.
Weirong Lai, Email: 15195477778@163.com.
Lijing Luo, Email: 450753760@qq.com.
Qianwei Sun, Email: 15950670080@163.com.
Dongwen Su, Email: 405433780@qq.com.
Appendix A. Supplementary data
The following is the Supplementary data to this article:
Data availability
The datasets generated and analyzed during the current study are available from the corresponding author upon reasonable request. Comprehensive details regarding the study design, data collection procedures, and implementation protocols can be found on the official CHARLS website (https://charls.pku.edu.cn).
References
- 1.Zhang K., Qi J., Zuo P., et al. The mortality trends of falls among the elderly adults in the mainland of China, 2013-2020: a population-based study through the national disease surveillance points system. Lancet Reg Health West Pac. 2022;19 doi: 10.1016/j.lanwpc.2021.100336. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Zhu W., Lin H., Zhang J., et al. Real-world association of self-reported sleep duration and quality with falls among older adults: a representative nationwide study of China. Sleep Med. 2022;100:212–218. doi: 10.1016/j.sleep.2022.08.019. [DOI] [PubMed] [Google Scholar]
- 3.Zhang H., Tang S., Kong L., et al. Association between sleep duration and hip fracture risk among the older adults: a cross-sectional study based on the NHANES. BMC Muscoskelet Disord. 2025;26(1):478. doi: 10.1186/s12891-025-08721-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Guirguis-Blake J.M., Perdue L.A., Coppola E.L., et al. Interventions to prevent falls in older adults: updated evidence report and systematic review for the US preventive services task force. JAMA. 2024;332(1):58–69. doi: 10.1001/jama.2024.4166. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Xu Q., Ou X., Li J. The risk of falls among the aging population: a systematic review and meta-analysis. Front Public Health. 2022;10 doi: 10.3389/fpubh.2022.902599. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Davis J.C., Robertson M.C., Ashe M.C., et al. International comparison of cost of falls in older adults living in the community: a systematic review. Osteoporos Int: J Establ Result Coop Between Eur Found Osteoporos Natl Osteoporos Found USA. 2010;21(8):1295–1306. doi: 10.1007/s00198-009-1162-0. [DOI] [PubMed] [Google Scholar]
- 7.Li J., Cao D., Huang Y., et al. Sleep duration and health outcomes: an umbrella review. Sleep Breath. 2022;26(3):1479–1501. doi: 10.1007/s11325-021-02458-1. [DOI] [PubMed] [Google Scholar]
- 8.Knechel N.A., Chang P. The relationships between sleep disturbance and falls: a systematic review. J Sleep Res. 2022;31(5) doi: 10.1111/jsr.13580. [DOI] [PubMed] [Google Scholar]
- 9.Canever J.B., Zurman G., Vogel F., et al. Worldwide prevalence of sleep problems in community-dwelling older adults: a systematic review and meta-analysis. Sleep Med. 2024;119:118–134. doi: 10.1016/j.sleep.2024.03.040. [DOI] [PubMed] [Google Scholar]
- 10.Zhu C., Sun J., Huang Y., Lian Z. Sleep and risk of hip fracture and falls among middle-aged and older Chinese. Sci Rep. 2024;14(1) doi: 10.1038/s41598-024-74581-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Gao Q., Hu K., Yan C., et al. Associated factors of sarcopenia in community-dwelling older adults: a systematic review and meta-analysis. Nutrients. 2021;13(12):4291. doi: 10.3390/nu13124291. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Ramos A.R., Wheaton A.G., Johnson D.A. Sleep deprivation, sleep disorders, and chronic disease. Prev Chronic Dis. 2023;20 doi: 10.5888/pcd20.230197. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Zhao X., Han L., Ding H., Li C. Associations between sleep duration trajectories and physical dysfunction among middle-aged and older Chinese adults. BMC Public Health. 2025;25(1):2600. doi: 10.1186/s12889-025-23870-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Jiang J., Zhou Q., Zhang C., Cong K. Activities of daily living mediate the association between balance and falls in middle aged and older adults. Sci Rep. 2025;15(1) doi: 10.1038/s41598-025-14608-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Zhao Y., Hu Y., Smith J.P., et al. Cohort profile: the China health and retirement longitudinal study (CHARLS) Int J Epidemiol. 2014;43(1):61–68. doi: 10.1093/ije/dys203. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Essien S.K., Feng C.X., Sun W., et al. Sleep duration and sleep disturbances in association with falls among the middle-aged and older adults in China: a population-based nationwide study. BMC Geriatr. 2018;18(1):196. doi: 10.1186/s12877-018-0889-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Jiang C.H., Zhu F., Qin T.T. Relationships between chronic diseases and depression among middle-aged and elderly people in China: a prospective study from CHARLS. Curr Med Sci. 2020;40(5):858–870. doi: 10.1007/s11596-020-2270-5. [DOI] [PubMed] [Google Scholar]
- 18.Yao Y., Wang K., Xiang H. Association between cognitive function and ambient particulate matters in middle-aged and elderly Chinese adults: evidence from the China health and retirement longitudinal study (CHARLS) Sci Total Environ. 2022;828 doi: 10.1016/j.scitotenv.2022.154297. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Zhu X., Wang Y., Luo Y., Ding R., et al. Bidirectional, longitudinal associations between depressive symptoms and IADL/ADL disability in older adults in China: a national cohort study. BMC Geriatr. 2024;24(1):659. doi: 10.1186/s12877-024-05248-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Yeung S.S.Y., Reijnierse E.M., Pham V.K., et al. Sarcopenia and its association with falls and fractures in older adults: a systematic review and meta-analysis. J Cachexia Sarcopenia Muscle. 2019;10(3):485–500. doi: 10.1002/jcsm.12411. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Gao K., Cao L.F., Ma W.Z., et al. Association between sarcopenia and cardiovascular disease among middle-aged and older adults: findings from the China health and retirement longitudinal study. eClinicalMedicine. 2022;44 doi: 10.1016/j.eclinm.2021.101264. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Sun Q., Shen L., Liu H., et al. Development and validation of a predictive model for sarcopenia risk in older chinese adults based on key factors. BMC Geriatr. 2025;25(1):464. doi: 10.1186/s12877-025-06104-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Wang S., Yu M., Huang W., et al. Longitudinal association between ADL disability and depression in middle-aged and elderly: national cohort study. J Nutr Health Aging. 2025;29(2) doi: 10.1016/j.jnha.2024.100450. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Zhou L., Ma X., Wang W. Relationship between cognitive performance and depressive symptoms in Chinese older adults: the China health and retirement longitudinal study (CHARLS) J Affect Disord. 2021;281:454–458. doi: 10.1016/j.jad.2020.12.059. [DOI] [PubMed] [Google Scholar]
- 25.Lin Y., Zhong S., Sun Z. Association between balance ability and cardiovascular disease onsets among middle-aged and older adults: an observational cohort study from the China health and retirement longitudinal study. Front Public Health. 2025;13 doi: 10.3389/fpubh.2025.1436520. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Austin P.C., Fang J., Lee D.S. Using fractional polynomials and restricted cubic splines to model non-proportional hazards or time-varying covariate effects in the cox regression model. Stat Med. 2022;41(3):612–624. doi: 10.1002/sim.9259. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Daneshvar A., Mousa G. Regression shrinkage and selection via least quantile shrinkage and selection operator. PLoS One. 2023;18(2) doi: 10.1371/journal.pone.0266267. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Folstein M.F., Folstein S.E., McHugh P.R. "Mini-mental state". A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975 Nov;12(3):189–198. doi: 10.1016/0022-3956(75)90026-6. [DOI] [PubMed] [Google Scholar]
- 29.Li H., Jia J., Yang Z. Mini-mental state examination in elderly Chinese: a population-based normative study. J Alzheimers Dis. 2016 May 7;53(2):487–496. doi: 10.3233/JAD-160119. [DOI] [PubMed] [Google Scholar]
- 30.Liu X., Xia X., Hu F., et al. The mediation role of sleep quality in the relationship between cognitive decline and depression. BMC Geriatr. 2022;22(1):178. doi: 10.1186/s12877-022-02855-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Liang C., Shi L., Li B., He Z. The mediating role of sarcopenia in the association between physical activity and falls among Chinese older adults: a cross-sectional study. Healthcare. 2023;11(24):3146. doi: 10.3390/healthcare11243146. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Ma T., Shi G., Zhu Y., et al. Sleep disturbances and risk of falls in an old Chinese population-rugao longevity and ageing study. Arch Gerontol Geriatr. 2017;73:8–14. doi: 10.1016/j.archger.2017.07.003. [DOI] [PubMed] [Google Scholar]
- 33.Kuo H.K., Yang C.C.H., Yu Y.H., et al. Gender-specific association between self-reported sleep duration and falls in high-functioning older adults. J Gerontol. 2010;65A(2):190–196. doi: 10.1093/gerona/glp143. [DOI] [PubMed] [Google Scholar]
- 34.Zhiqing Z., Huan L., Mengqi C., et al. vol. 47. 2024. (Construction of a nomogram model for fall influencing factors and risk in elderly inpatients). [Google Scholar]
- 35.Y M., Pw S. Poor sleep and risk of falls in community-dwelling older adults: a systematic review. J Appl Gerontol : Off J South Gerontol Soc. 2018;37(9) doi: 10.1177/0733464816681149. [DOI] [PubMed] [Google Scholar]
- 36.Fjell A.M., Sørensen Ø., Wang Y., et al. Is short sleep bad for the brain? Brain structure and cognitive function in short sleepers. J Neurosci. 2023;43(28):5241–5250. doi: 10.1523/JNEUROSCI.2330-22.2023. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Wang L.M., Guan Y.Q. Association between sleep status and prevalence of major chronic diseases. Zhonghua Liuxingbingxue Zazhi. 2020;41(8):1237–1241. doi: 10.3760/cma.j.cn112338-20200529-00788. [DOI] [PubMed] [Google Scholar]
- 38.Lu C., Liao B., Nie J., et al. The association between sleep duration and chronic diseases: a population-based cross-sectional study. Sleep Med. 2020;73:217–222. doi: 10.1016/j.sleep.2020.05.005. [DOI] [PubMed] [Google Scholar]
- 39.Wang S., Li Q., Hu J., et al. Association of multimorbidity patterns and order of physical frailty and cognitive impairment occurrence: a prospective cohort study. Age Ageing. 2025;54(4):afaf101. doi: 10.1093/ageing/afaf101. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Wei J., Wu H., Zheng Y., et al. Proteomic signatures underlying sleep, circadian activity patterns, and major chronic diseases. J Pineal Res. 2025;77(4) doi: 10.1111/jpi.70067. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Ma Y., Liang L., Zheng F., et al. Association between sleep duration and cognitive decline. JAMA Netw Open. 2020;3(9) doi: 10.1001/jamanetworkopen.2020.13573. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Sabatini S., Ukoumunne O.C., Ballard C., et al. Cross-sectional and longitudinal associations between subjective sleep difficulties and self-perceptions of aging. Behav Sleep Med. 2022;20(6):732–761. doi: 10.1080/15402002.2021.1994405. [DOI] [PubMed] [Google Scholar]
- 43.Aakre J.A., Schulz J., Ballard C., et al. Self-reported sleep fragmentation and sleep duration and their association with cognitive function in PROTECT, a large digital community-based cohort of people over 50. Int J Geriatr Psychiatr. 2023;38(11) doi: 10.1002/gps.6022. [DOI] [PubMed] [Google Scholar]
- 44.Idalino S.C.C., Cândido L.M., Wagner K.J.P., et al. Association between sleep problems and functional disability in community-dwelling older adults. BMC Geriatr. 2024;24(1):253. doi: 10.1186/s12877-024-04822-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Andrade Junior M.C.D., Stefanini R., Gazzola J.M., et al. Individuals with peripheral vestibulopathy and poor quality of sleep are at a higher risk for falls. Brazilian J Otorhinolaryngol. 2021;87(4):440–446. doi: 10.1016/j.bjorl.2019.10.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Min Y., Kirkwood C.K., Mays D.P., Slattum P.W. The effect of sleep medication use and poor sleep quality on risk of falls in community-dwelling older adults in the US: a prospective cohort study. Drugs Aging. 2016;33(2):151–158. doi: 10.1007/s40266-015-0339-9. [DOI] [PubMed] [Google Scholar]
- 47.Kobayashi K., Ando K., Nakashima H., et al. Relationship between use of sleep medication and accidental falls during hospitalization. Nagoya J Med Sci. 2021;83(4):851–860. doi: 10.18999/nagjms.83.4.851. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Gschwind Y.J., Kressig R.W., Lacroix A., et al. A best practice fall prevention exercise program to improve balance, strength/power, and psychosocial health in older adults: study protocol for a randomized controlled trial. BMC Geriatr. 2013;13:105. doi: 10.1186/1471-2318-13-105. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The datasets generated and analyzed during the current study are available from the corresponding author upon reasonable request. Comprehensive details regarding the study design, data collection procedures, and implementation protocols can be found on the official CHARLS website (https://charls.pku.edu.cn).