Social isolation to depression: the chain mediating effect of sleep duration and frailty—evidence from a Chinese nationwide cross-sectional study

Abstract

Background

Although social isolation, sleep duration, and frailty have each been linked to depression, their potential sequential pathway—from social isolation through sleep duration and frailty to depression—has not been examined in a large, nationally representative Chinese sample. Guided by the stress–vulnerability model, this study aimed to investigate whether sleep duration and frailty sequentially mediate the association between social isolation and depression.

Methods

Data from the 2018 CHARLS wave were used. Key measures included a general information questionnaire, social isolation index, sleep duration, frailty index, and CESD-10. Spearman’s correlation was first used to explore the associations among social isolation, sleep duration, frailty, and depression. Subsequently, a chain mediation analysis was conducted using the PROCESS macro in SPSS to examine the mediating roles of sleep duration (first mediator, M1) and frailty (second mediator, M2) in the relationship between social isolation (independent variable, X) and depression(dependent variable, Y).

Results

The final analytic sample consisted of 5270 older Chinese adults, with a median age of 67 years (interquartile range [IQR]: 64–73); 47% were women. Pearson correlation analysis showed that social isolation was negatively correlated with sleep duration (r = -0.061, p < 0.01), positively associated with frailty (r = 0.138, p < 0.01), and positively associated with depression (r = 0.162, p < 0.01). Mediation analysis indicated that sleep duration (indirect effect = 0.065, 95% CI: 0.035 to 0.098), frailty (indirect effect = 0.497, 95% CI: 0.388 to 0.607), and their chain pathway (indirect effect = 0.050, 95% CI: 0.027 to 0.074) significantly mediated the relationship between social isolation and depression.

Conclusion

Promoting sleep health and identifying frailty in older adults may reduce depression and improve mental health.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12877-025-06868-8.

Background

Depression is a multifactorial disorder characterized by emotional, behavioral, and physiological symptoms that impair daily functioning and reduce quality of life [1, 2]. Globally, over 280 million people suffer from depression. According to the World Health Organization (WHO), depression was the leading cause of disability in 2015, accounting for 7.5% of all years lived with disability. It also contributes significantly to over 800,000 suicides annually [3], and imposes immense direct and indirect social costs. Depression and anxiety are estimated to cost the global economy approximately $1 trillion annually in lost productivity, placing a substantial burden on public health systems worldwide [4].

Social isolation, defined as the absence of social connections, has emerged as a major psychosocial stressor. Its prevalence is highest in young adulthood and old age [5]. While social interactions and prosocial behavior are associated with improved health outcomes, lower mortality, and enhanced well-being, social isolation increases the risk of cardiovascular diseases, stroke, cognitive decline [6, 7]. Moreover, social isolation has been shown to elevate the risk of depression, as demonstrated in a study of young adult cancer survivors, where a strong correlation between social isolation and depressive symptoms was observed [8]. Numerous studies have investigated the relationship between sleep and depression. One study noted that, in addition to reduced sleep duration, excessive sleep was linked to an increased risk of depression. The risk of depression significantly rises with sleep duration of ≥ 8 h (OR = 1.32 [95%CI 1.23, 1.41], p < 0.001), with sleep duration showing a U-shaped relationship with depression. However, such U-shaped patterns have primarily been observed in general adult populations, whereas studies focusing on middle-aged and older adults—particularly in Chinese cohorts—have reported a stronger detrimental effect of short sleep duration on depression, possibly because short sleep is more common and more physiologically disruptive in this age group [9, 10]. Frailty, a common condition in older adults, results from age-related physiological decline and is marked by fatigue, muscle weakness, malnutrition, and diminished resilience to stress [11]. It has been linked to numerous adverse outcomes including hospitalization, disability, and mortality [12]. Studies have shown that frailty is associated with depression. For example, among patients undergoing hemodialysis, frailty was significantly associated with depressive symptoms [13]. Similarly, longitudinal research has found that baseline frailty increases the risk of subsequent depression by as much as 90% [14, 15]. Research has also shown a close relationship between sleep characteristics and frailty. Deviations from the recommended 7–8 h of sleep per night are associated with poor health outcomes and increased frailty risk. Various sleep issues—including insomnia, hypersomnia, and poor sleep quality—have all been linked to higher frailty in middle-aged and older populations [16].

According to the stress–vulnerability model, health outcomes result from the interaction between inherent vulnerability and external stressorss [17]. Frailty reflects a reduction in physiological reserve and adaptability to stressors [18]. while deviations in sleep duration indicate physiological instability and potential disruption of neurobiological mechanisms, which may manifest in mood and cognitive disturbances [19, 20]. According to the stress-vulnerability model, frailty and sleep duration can be viewed as key components of an individual’s intrinsic vulnerability. Additionally, social isolation, as a chronic external stressor, exerts a profound impact on frailty and sleep duration by limiting an individual’s social support network, increasing feelings of loneliness, and reducing opportunities for social interaction.

Depression, as a multifactorial syndrome, is driven by a complex interplay of genetic, neurobiological, psychological, and environmental factors, including neurotransmitter dysregulation, neuroinflammation, altered stress responses, as well as adverse life events such as social isolation and chronic stress [21]. Despite ongoing research advances, the precise pathophysiological mechanisms underlying depression remain incompletely understood, which limits the development of personalized prevention and treatment strategies. Social isolation, sleep duration, and frailty have all been demonstrated to be closely associated with late-life depression; however, the mechanisms through which these factors influence depression lack systematic investigation. Specifically, social isolation may disrupt circadian rhythms, resulting in reduced sleep duration and quality, which in turn accelerates the progression of frailty and ultimately increases the risk of depressive symptoms. Nonetheless, empirical evidence supporting this mechanism in large-scale Chinese older adult populations remains scarce. Grounded in the stress–vulnerability model, this study leverages nationally representative data from the China Health and Retirement Longitudinal Study (CHARLS) to systematically examine the independent and interrelated mechanisms by which social isolation, sleep duration, and frailty affect depression. The aim is to elucidate the critical links among social environment, physiological vulnerability, and emotional disorders, thereby providing theoretical foundations for culturally tailored precision interventions to improve mental health outcomes among Chinese older adults. Our hypothesized model is depicted in Fig. 1 and comprises the following pathways:

Fig. 1.

Hypothesized model diagram of the mediating role of sleep duration and frailty in the relationship between social isolation and depression

Theoretical hypothesis

• Social isolation may disrupt an individual’s sleep patterns, leading to insufficient sleep, which in turn affects mental health, particularly depression.

• Social isolation may exacerbate the risk of depression by increasing levels of frailty.

• Social isolation may sequentially affect sleep duration and frailty, ultimately leading to depression.

Study population and research methodology

General information questionnaire

A self-designed general information questionnaire, developed with reference to the CHARLS codebook and prior studies, was used to collect participants’ demographic and lifestyle data. Age was calculated from the reported birth year and treated as a continuous variable. Gender was classified as male or female. Residence, based on the household registration (hukou) variable, was categorized into urban community, rural village, or special zone. Education level was grouped into four categories: (1) no formal education or incomplete primary school, (2) elementary or middle school, (3) high school or vocational school, and (4) college degree or above. Marital status was dichotomized as having a spouse (currently married or partnered) or not (widowed, divorced, separated, or never married). Smoking status was derived from self-reported current smoking behavior and classified as current smoker (still smoking) or non-smoker (quit or never smoked). Drinking status was based on alcohol consumption frequency over the past year and categorized as drinker (≥ 1 time/month) or non-drinker (< 1 time/month or none). Health status was assessed by self-rated health (“very good,” “good,” “fair,” “poor,” “very poor”) and recoded into good (“very good” and “good”), fair (“fair”), and poor (“poor” and “very poor”), consistent with previous CHARLS studies [22–24].

Social isolation index

The social isolation index used in this study was adapted from previous studies. The index included the following items: marital status (currently unmarried), frequency of contact with adult children (in person, by phone, or by email), living in a rural area, living alone, and engagement in social activities (e.g., interactions with friends, participation in community-related organizations, and involvement in voluntary or charitable work) less than once a month [25]. The index score ranges from 0 to 5, with higher scores indicating greater social isolation.

 Sleep duration

Total sleep duration over a 24-hour period was calculated by summing the duration of nighttime sleep and daytime naps. Nighttime sleep duration and nap duration were assessed by the following questions: “In the past month, how many hours of actual sleep did you get each night (on average)?” and “In the past month, how long did you nap after lunch?” Nap duration refers to the total amount of sleep achieved after lunch each day, while total sleep duration refers to the overall sleep duration each day. The reliability of self-reported nighttime sleep and daytime naps has been validated in sleep epidemiology research [26].

Frailty index (FI)

The frailty index (FI) used in this study consisted of 32 items, which have been widely employed to assess frailty in various populations. The FI score ranges from 0 to 1, with the score representing the ratio of deficits present to the total number of deficits considered. A higher FI score indicates greater frailty. This frailty index has been extensively validated in previous research (Table S1) [27].

Center for epidemiologic studies depression Scale-10 (CESD-10)

The level of depression in this study was measured using the CESD-10, a shortened version of the Depression Assessment Scale developed by the National Institute of Mental Health to quickly screen individuals for depressive symptoms. The CESD-10 consists of 10 items, with a total score ranging from 0 to 30; higher scores indicate more severe depressive symptoms. The CESD-10 is widely used in epidemiological surveys and mental health research and is particularly suitable for screening depressive symptoms in the general population.

Statistical methods

Analyses used R 4.0 and SPSS 25.0. Categorical variables were summarized as n (%) and compared by χ² tests (Fisher’s exact test when expected counts < 5); continuous variables as mean ± SD with independent-samples t-tests when approximately normal, otherwise as median (Q1, Q3) with Wilcoxon rank-sum tests. Pearson correlation analysis was conducted to explore the relationships between variables. The process plug-in (model 6) in SPSS was used to test the chain mediation model, with a 95% confidence interval obtained via the bootstrap method, using 5000 repeated samples. The significance level was set at α = 0.05.

Results

General information on the population included in the study

The 2018 CHARLS initially included 19,816 participants. After excluding 8,799 participants under the age of 60, 11,017 participants aged 60 and above remained. Next, participants with missing data on social isolation (4170 participants), sleep duration (0 participants), frailty (42 participants), and depression (1,535 participants) were excluded. The final sample for analysis consisted of 5,270 participants. To maximize the sample size, missing demographic information was imputed using multiple imputation with the ‘mice’ package in R. Among the 5270 included participants, the median age was 67 years (IQR 64–73). Male accounted for 2793 (53%) and female for 2477 (47%). Regarding education, 2516 (48%) had no formal education or incomplete primary school, 2160 (41%) had elementary or middle school, 502 (9.5%) had high school or vocational education, and 92 (1.7%) held a college degree or higher. A total of 3551 (67%) had a spouse, while 1719 (33%) did not. For smoking status, 801 (15%) were current smokers, while 4469 (85%) had either quit or never smoked; for drinking status, 1738 (33%) reported drinking and 3532 (67%) did not. The screening process and basic demographic characteristics of the study population are illustrated in Fig. 2, and additional demographic information is provided in Table 1. We also compared the baseline characteristics of included and excluded participants (Table S2) and observed some differences in demographic variables, such as age, gender, and education level.

Fig. 2.

Flowchart for the screening of research subjects

Table 1.

Basic information on the included study subjects

Characteristic	N = 5270	Percentage
Age (years), Median (Q1, Q3)	67 (64, 73)	——
Gender
- Male	2793	53%
- Female	2477	47%
Education Level
- No formal education or incomplete primary school	2516	48%
- Elementary or Middle school	2160	41%
- High school or Vocational school	502	9.5%
- College degree or higher	92	1.7%
Marital Status
- Having a Spouse	3551	67%
- Not Having a Spouse	1719	33%
Smoking Status
- Still have	801	15%
- Quit or never smoked	4469	85%
Drinking Status
- Drink	1738	33%
- None	3532	67%
Residence
- Urban Community	1497	28%
- Rural Village	3761	71%
- Special Zone	12	0.2%
Health Status
- Good	1143	22%
- Fair	2538	48%
- Poor	1589	30%
Household Income (RMB), Median (Q1, Q3)	17,614 (4420, 50,190)	——
Number of chronic disease
− 0	2378	45%
− 1–2	2227	42%
- ≥3	665	13%

Values are presented as median (Q1, Q3) for continuous variables or n (%) for categorical variables

Results of univariate analysis and correlation analysis

Independent samples t-tests were conducted for categorical variables with two categories, and one-way analysis of variance (ANOVA) was used for variables with three categories. The results indicated that residence, education level, marital status, gender, smoking status, health status, drinking status, number of chronic disease and household income were all significantly associated with depression in the older adults (p < 0.01). Correlation analyses revealed that social isolation was negatively correlated with sleep duration (r = −0.061, p < 0.01), positively correlated with frailty (r = 0.138, p < 0.01), and positively correlated with depression (r = 0.162, p < 0.01). Sleep duration was negatively correlated with both frailty (r = −0.211, p < 0.01) and depression (r = −0.257, p < 0.01), while frailty was positively correlated with depression (r = 0.568, p < 0.01). Regression analyses yielded consistent findings (Table 2; Fig. 3).

Table 2.

Correlation analysis between social isolation, sleep duration, frailty, depression

Variable	Social Isolation	Sleep Duration	Frailty	Depression
Social Isolation	1	−0.061	0.138	0.162
Sleep Duration	−0.061	1	−0.211	−0.257
Frailty	0.138	−0.211	1	0.568
Depression	0.162	−0.257	0.568	1

Fig. 3.

mediating role of sleep duration and frailty in the relationship between social isolation and depression

Mediating effect test

On the basis of correlation analysis and linear regression, in order to test the mediating role of sleep duration and frailty between social isolation and depression, this study conducted a mediation effect test based on the Bootstrap method using the PROCESS macro in SPSS, and 95% confidence intervals were computed using a sample size of 5,000 Bootstrap samples. We found that, among older adults, sleep duration mediated the association between social isolation and depression, with an effect value of 0.065 (95% CI: 0.035–0.098), accounting for 5.324% of the total effect. Frailty also served as a mediator, with an effect value of 0.497 (95% CI: 0.388–0.607), explaining 40.704% of the total effect. Moreover, sleep duration and frailty exerted a chain mediating effect, with an effect value of 0.050 (95% CI: 0.027–0.074), contributing 4.095% of the total effect. The direct effect of social isolation on depression was 0.609 (95% CI: 0.443–0.775), accounting for 49.877% of the total effect, while the total indirect effect was 0.612 (95% CI: 0.494–0.733), explaining 50.123% of the total effect. The total effect of social isolation on depression was 1.221 (95% CI: 1.020–1.422). See Fig. 3 for a diagram of the mediation model and Table 3 for a decomposition of the mediating effects.

Table 3.

Decomposition of effects of social isolation on depression in older adults

	Effect value	95% confidence interval		Relative mediation effect (%)
		BootLLCI	BootULCI
Social isolation- sleep duration-depression	0.065	0.035	0.098	5.324%
Social isolation- frailty-depression	0.497	0.388	0.607	40.704%
Social isolation- sleep duration-frailty-depression	0.050	0.027	0.074	4.095%
Ind1 - Ind2	−0.432	−0.545	−0.320	-
Ind1 - Ind3	0.016	0.001	0.033	-
Ind2- Ind3	0.448	0.338	0.558	-
Direct effect	0.609	0.443	0.775	49.877%
Total indirect effect	0.612	0.494	0.733	50.123%
Total effect	1.221	1.020	1.422	1

To address potential confounders, we built an adjusted mediation model based on the univariate results and re-estimated the mediation analysis after including residence, education level, marital status, gender, smoking status, self-rated health, drinking status, number of chronic diseases, and household income as covariates. The results indicate that the key mediating pathway did not undergo any substantive change; its effect size and statistical significance were broadly consistent with those in the unadjusted model. These findings provide strong evidence for the robustness of this mediating relationship, suggesting that it is independent of the demographic and health-related factors included as covariates (Figure S1, Table S3). Given that FI construction typically requires < 5% item-level missingness, we performed a sensitivity analysis: item-level missingness was computed, items with > 5% missingness were excluded, and the FI was recomputed using the retained items. Relative to the primary analysis, the direction and statistical significance of the key effects were materially unchanged (Figure S2, Table S4).

Discussion

Based on relevant literature and the stress-vulnerability model, this study systematically examined the associations between social isolation, sleep duration, frailty, and depression in older adults. The results revealed that sleep duration mediated the relationship between social isolation and depression, while frailty also played a mediating role in this association among older adults. Specifically, both sleep duration and frailty were found to mediate the link between social isolation and depression in older adults. These findings highlight potential intervention targets for alleviating depression in older adults, providing both a theoretical foundation and empirical evidence for improving their mental health.

Social isolation is recognized as a major public health concern due to its significant threat to physical and mental health [28]. Numerous epidemiological studies have examined the complex relationships among social isolation, sleep duration, and depression. For instance, a prospective cohort study of 344,956 participants from the UK Biobank demonstrated an additive interaction between social isolation and cardiovascular-renal-metabolic health in elevating the risk of depression [29]. Another study involving 275 American Indian adults explored associations between social networks, loneliness, existential isolation (the feeling of being alone in one’s experiences), and sleep health. It found that existential isolation was the only statistically significant predictor of sleep health, with higher levels of existential isolation linked to poorer sleep quality [30]. A Chinese study of 734 individuals revealed that poorer sleep quality significantly increases the risk of isolated depression in adults over 18 [31]. Furthermore, another study highlighted that insomnia symptoms, particularly prevalent among women during the menopausal transition, contribute to elevated levels of anxiety and depression [32]. In the present study, sleep duration was identified as a mediating factor between social isolation and depression in older adults. Social isolation directly influences depression levels and indirectly affects them by altering sleep duration. This phenomenon may be explained by the heightened emotional distress caused by social isolation, which disrupts normal sleep patterns, leading to insomnia or hypersomnia. Sleep deprivation exacerbates the overactivation of the stress response, fueling anxiety and negative emotions [33, 34]. Cultural and traditional beliefs in China further complicate the understanding of depression, its management, and social interactions among adults with mental health issues. For example, many Chinese individuals may suppress emotions and avoid self-reporting stress and depressive symptoms due to stigma [31]. These findings underscore the importance of increasing social support and managing sleep duration to alleviate depressive symptoms in the Chinese middle-aged and older population. In clinical practice, providing sufficient social support can be achieved through group therapy, physical exercise programs, internet and social media training, community activities, emotional support interventions, and intergenerational communication initiatives. Such efforts can effectively reduce social isolation among this demographic [35]. Future research should explore the long-term effects of various sleep duration on the mental health of older adults. This will provide a foundation for developing individualized sleep intervention plans aimed at improving sleep quality and, in turn, mitigating depression.

The present study found that frailty played a mediating role between social isolation and depression in older adults, suggesting that frailty could be a critical target for interventions aimed at alleviating depression. Previous studies have also explored the association between frailty, social isolation, and depression. A systematic review of older adults demonstrated that frailty significantly increased the risk of depression (OR = 1.90; 95% CI: 1.55–2.32, I² = 0%) [36]. Additionally, a Mendelian randomization study investigated the relationship between frailty and common psychiatric disorders (bipolar disorder, major depressive disorder, schizophrenia, and suicide or other deliberate self-harm) using genetic data. The findings indicated that causal associations were more pronounced between frailty and major depression as well as suicide or intentional self-harm. Several mechanisms may explain these associations. First, frailty is strongly linked to a chronic inflammatory state, characterized by elevated levels of inflammatory markers such as IL-6 and CRP, which may disrupt the balance of neurotransmitters (e.g., serotonin and dopamine) in the brain, thereby increasing the risk of depression [37, 38]. Chronic inflammation may also impair the function of the hippocampus and prefrontal cortex, which are regions closely associated with emotion regulation [39]. Furthermore, frail adults often experience physical inactivity, a significant decline in quality of life, socioeconomic deprivation, and co-morbidities, all of which may heighten their vulnerability to life stressors. This elevated vulnerability, in turn, increases their risk of developing mental health disorders [40]. Despite being a severe public health issue, frailty is both preventable and treatable, making it a reversible condition. Identifying older adults most at risk for frailty is thus a public health priority [16]. Currently, there is no universally accepted gold standard for frailty screening, and existing tools differ in their conceptual frameworks and applicability across settings. Given these limitations, future efforts should utilize contextually appropriate validated screening tools to identify frail individuals and employ comprehensive clinical assessments to elucidate the characteristics and underlying mechanisms of frailty [41]. Targeted management strategies should include integrated care plans that comprehensively address key contributors to frailty. These include optimizing polypharmacy to reduce medication-related risks, treating sarcopenia to improve physical function, managing the underlying causes of weight loss and fatigue, and ensuring adequate social support to promote adherence to care [42].

It has been reported that approximately 50% of the older adults suffers from sleep disorders. In Korea, 32.4% of older adults in the Korean community experience sleep disorders [43], while 61.8% do not achieve the recommended sleep duration set by the National Sleep Foundation [44]. Insufficient sleep is strongly associated with adverse outcomes such as frailty, highlighting the critical role of proper sleep in maintaining the overall health and well-being of older adults. Research on community-dwelling older adults has revealed that good subjective sleep health is linked to lower odds of frailty, particularly among older women, even after adjusting for socio-demographic characteristics and health disparities. Conversely, the same study noted that men reported poor sleep continuity and deviations in sleep duration, which were associated with an increased risk of mortality [44, 45]. Although the origins of frailty may stem from earlier stages of life, studies suggest that healthier lifestyles in late adulthood and old age can mitigate frailty risks. For instance, individuals experiencing sleep difficulties (adjusted β × time = 0.20, 95% CI: 0.10–0.31) showed a faster increase in frailty from late middle age to old age, whereas those with improved sleep (adjusted β × time = −0.10, 95% CI: −0.23–0.01) demonstrated a slower increase in frailty over the same period [46]. This evidence supports the notion that sleep is a key factor influencing frailty. Building upon this foundation, the present study further explored the role of sleep duration and frailty in the relationship between social isolation and depression. The findings indicate that sleep duration and frailty play a chain mediating role in this relationship. Specifically, social isolation adversely affects sleep duration, which subsequently increases frailty, ultimately heightening the risk of depressive symptoms. These results suggest that interventions aimed at improving sleep quality and reducing frailty may serve as effective strategies to mitigate the psychological impact of social isolation and reduce the prevalence of depression. Firstly, interventions targeting sleep quality, such as Cognitive Behavioral Therapy for Insomnia (CBT-I), have demonstrated significant efficacy in addressing sleep disorders and mood symptoms. By improving sleep patterns, these interventions can effectively alleviate mood disturbances induced by sleep deprivation, thereby reducing the risk of depression [47]. Secondly, interventions that enhance physical health, such as regular physical activity and nutritional programs, can help decrease frailty and associated depressive symptoms. For example, aerobic exercise and strength training not only improve physical health but also boost self efficacy and alleviate the low mood linked to frailty [48]. Finally, enhancing social support is another critical intervention avenue. Studies have shown that strengthening social support networks can alleviate the relationship between social isolation and depression. Community engagement initiatives and fostering supportive relationships can reduce loneliness and social isolation, thereby lowering the risk of depression [49].

Conclusion

The present study identified that sleep duration mediates the relationship between social isolation and depression in older adults. Frailty was also found to mediate this relationship, and a chain mediation effect of sleep duration and frailty was observed between social isolation and depression. These findings suggest that addressing both sleep duration and frailty could be critical for mitigating the psychological effects of social isolation and reducing the prevalence of depression among older adults.

Limitations

This study has several limitations. Firstly, its cross-sectional design prevents us from inferring causality. While associations between variables can be identified, determining cause and effect requires that the cause occurs before the effect. Since cross-sectional studies measure all variables simultaneously, it is not possible to confirm the direction of these relationships. Future studies should employ longitudinal or experimental approaches to better clarify causal links. Secondly, the study relied on self-reported measures, which may introduce recall bias. Future research should incorporate objective measurement methods to ensure greater accuracy and reliability of the data. Thirdly, this study focused solely on sleep duration and its effects on the older adults. Future research should expand to include additional aspects of sleep, such as sleep quality (e.g., difficulty falling asleep, frequency of nighttime awakenings) and sleep structure (e.g., the proportion of rapid-eye-movement sleep). Fifthly, while efforts were made to control for confounding variables, the study may not have accounted for all potential confounders, which could influence the observed relationships. This limitation underscores the need for future studies to employ more robust methods to better control for these factors and enhance the reliability of their conclusions. Finally, due to factors such as missing data for key variables, non-response from certain demographic groups, attrition bias may have occurred. These exclusions could introduce systematic differences between included and excluded participants, which may limit the generalizability of our findings to the broader population. Additionally, since recruitment was restricted to China, the observed pathway from social isolation to depression should be interpreted with caution. To confirm their generalizability, these findings require validation through replication in more diverse populations and settings.

Authors’ contributions

Bin Ma served as principal author contributed to conception and draft of the manuscript; Mengyuan Zhang contribute to the data collection and analysis; Yuanmin Jia contribute to data analysis; Dongqing Zhang contribute to the revise, correction and validation; Ou Chen contributed to validation of the article and final approval.

Funding

National Natural Science Foundation of China (82172543); Natural Science Foundation of Shandong province(ZR2024MH071).

Data availability

The data set used in this research is available from the corresponding author upon a reasonable request.

Declarations

Ethics approval and consent to participate

This study was approved by the Biomedical Ethics Council of Beijing University. The IRB approval number for the main household survey, including anthropometrics, is IRB00001052-11015; the IRB approval number for biomarker collection is IRB00001052-11014. All participants gave written informed consent.

Consent for publication

Not applicable.

Conflict of 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.

Competing interests

The authors declare no competing interests.