https://orcid.org/0009-0001-4509-769X
Figures
Abstract
Background
The association between nighttime sleep duration and osteoarthritis (OA) remains ambiguous. Chinese older adults exhibit distinct sleep patterns, as well as genetic predispositions and rapidly changing lifestyles, which may have shaped the unique epidemiology of OA. However, most existing evidence is based on Western populations. This study aimed to investigate this association and the potential moderating role of BMI in middle-aged and older Chinese adults.
Methods
Data of wave 1 (2011) and wave 4 (2018) were obtained from the nationally representative China Health and Retirement Longitudinal Study (CHARLS). Univariate and multivariate logistic regression models were utilized to explore the association of sleep duration and OA with sleeping 7–9 h as reference group. Additionally, to further explore the potential combined effect of sleep and BMI, interaction terms were added into the model. Restricted cubic spline was also used to explore the non-linear correlation between sleep duration and OA.
Results
Out of 6,825 participants, 1,396 were diagnosed with OA. After multivariable adjustment, the odds ratios (OR) for OA were 1.39 (95% CI 1.20–1.60; P < 0.001) for individuals with sleep duration (<6 h/night) and 1.27 (95% CI 1.20–1.60 P = 0.003) for individuals with sleep duration (6–7 h/night). The association between sleep duration and OA followed a U-shaped pattern, with 7.5 h acting as an inflection point. Significant interactions were found in overweight individuals, with both short (OR = 1.41, P = 0.042) and long (OR = 2.71, P = 0.006) sleep durations increasing OA risk.
Citation: Long L, Wang Q, Fang B, Cao L, Xu Y, Li G (2025) Non-linear association of sleep duration with osteoarthritis among middle-aged and older adults: Results from a prospective cohort study in China. PLoS One 20(10): e0335552. https://doi.org/10.1371/journal.pone.0335552
Editor: Qian Wu, The First Affiliated Hospital of Soochow University, CHINA
Received: April 7, 2025; Accepted: October 12, 2025; Published: October 31, 2025
Copyright: © 2025 Long et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: CHARLS data are freely available upon registration at: http://charls.pku.edu.cn/en. We have uploaded the minimal dataset necessary to replicate the study’s findings to the public Figshare repository and provide the following link/DOI for access: 10.6084/m9.figshare.29652194.
Funding: B.F. was supported by the Zhejiang Provincial Administration of Traditional Chinese Medicine under the Zhejiang Provincial Traditional Chinese Medicine Science and Technology Program (Project No.: 2024ZL828).
Competing interests: The authors have declared that no competing interests exist.
Abbreviations: BMI, body mass index; CHARLS, China Health and Retirement Longitudinal Study; CI, confidence interval; CRP, C-reactive protein; DM, diabetes mellitus; IL-6, interleukin-6; OA, osteoarthritis; OR, odds ratio; SD, standard deviation; TNF-α, tumor necrosis factor-alpha; WBC, white blood cell count; WHO, World Health Organization
Introduction
Osteoarthritis (OA), a prevalent degenerative joint disease characterized by articular cartilage deterioration, pain, and functional impairment, poses a significant global health burden. OA affects over 303 million individuals worldwide and disproportionately impacts middle-aged and older adults. Its prevalence is expected to rise due to aging populations and increasing obesity rates [1]. In China, rapid urbanization, significant lifestyle transitions, and an increasing obesity rate have notably increased the burden and altered the risk factor profile for OA compared to previous decades, highlighting the need for targeted epidemiological insights [2]. Additionally, Chinese older adults exhibit unique sleep patterns, including a higher prevalence of daytime napping compared to Western populations (approximately 60% vs. 25%, respectively) [3]. Furthermore, genetic and epigenetic differences in OA susceptibility among Asian populations compared to Europeans underscore the necessity for region-specific studies [4]. Despite these distinctive characteristics, most existing epidemiological studies investigating sleep duration and OA primarily focus on Western cohorts, limiting generalizability to the Chinese population. Beyond its economic costs, OA significantly reduces quality of life through chronic pain and mobility limitations, necessitating deeper insights into modifiable risk factors [3].
Sleep, a cornerstone of physiological homeostasis, plays a pivotal role in systemic health. Current guidelines recommend 7–9 hours of nightly sleep for adults, yet nearly one-third of individuals report suboptimal sleep duration [5]. Both short (<7 hours) and long (>9 hours) sleep durations have been linked to adverse outcomes, including cardiovascular diseases, metabolic dysfunction, and premature mortality [6]. Emerging evidence suggests sleep disturbances may also influence OA pathogenesis. Mechanistically, disrupted sleep elevates pro-inflammatory cytokines such as interleukin-6 (IL-6) and tumor necrosis factor-alpha (TNF-α), which accelerate cartilage degradation and synovial inflammation [7]. Epidemiological studies reveal a U-shaped association between sleep duration and OA risk, with minimal risk observed at 7–8 hours [8,9]. However, these findings derive predominantly from Western cohorts, limiting generalizability to Asian populations with distinct lifestyle and genetic profiles [4].
Notably, obesity—quantified by body mass index (BMI), is a key driver of systemic inflammation and metabolic dysregulation [10]. Visceral fat secretes adipocytokines (e.g., leptin, adiponectin) that promote joint catabolism, while excess mechanical load on weight-bearing joints exacerbates structural damage [10]. Both insufficient and prolonged sleep correlate with elevated BMI, suggesting bidirectional pathways linking sleep, adiposity, and OA [11,12]. Previous studies in U.S. cohorts have shown that 12.1% of the association between sleep duration and OA is mediated by waist circumference [13]. However, a study on nighttime sleep duration and knee OA in a Chinese population lacked in-depth analysis of the moderating role of BMI [2].
To address these gaps, we leveraged data from a prospective cohort of middle-aged and older adults in China to investigate: (1) the relationship between sleep duration and OA incidence, and (2) the moderating role of BMI.
Method
Study population
We utilized data from the CHARLS [14], a nationally representative longitudinal study of Chinese adults aged 45 and above. The baseline national census of CHARLS was fielded in 2011, including about 10,000 households from 17,500 individuals in 150 counties/districts and 450 villages/resident committees through multistage stratified probability-proportionate-to-size sampling. The individuals were followed up every 2 or 3 years by a face-to-face computer-assisted personal interview. More details about objectives, design, and methods can be found in a previous study [14].
The inclusion criteria were: 1) individuals aged at least 45 years old in the baseline survey; 2) having data regarding OA, nighttime sleep duration, and BMI; 3) reported having OA absence in the baseline; 4) being followed up till 2018; 5) and having OA data in 2018. After exclusion, 6,825 respondents were included in the analysis. The detailed process of participant selection is shown in Fig 1.
Abbreviations: CHARLS, China Health and Retirement Longitudinal Study; BMI, body mass index.
The CHARLS had received ethical approval from the Institutional Review Board of Peking University (approval number: IRB00001052-11,015) [14] Written informed consents were obtained from all participants before attending the survey. The CHARLS was performed in line with the Declaration of Helsinki. This study was conducted following the Strengthening the Reporting of Observational Studies in Epidemiology guideline.
Assessment of sleep duration
The data on nighttime sleep duration was collected in the 2011 baseline survey. Each participant was asked “During the past month, how many hours of actual sleep did you get at night (average hours for one night)?”. We further categorized the self-reported nighttime sleep duration into four groups: < 6 h, 6 to < 7 h, 7–9 h, and > 9 h[5]. Sleep duration of 7–9 h was treated as the reference group.
Assessment of arthritis
New-onset arthritis was identified based on self-reported data; when the interviewer asked, “Have you been diagnosed with arthritis by a doctor?” and the respondents answered “Yes,” they were classified as arthritis patients. Participants who had arthritis in 2011 were excluded, and if the patient was diagnosed with arthritis after that until the follow-up period in 2018, he or she was included in the study under our definition of a patient with new-onset arthritis.
Assessment of covariates
The process of selecting covariates is grounded in substantial reasoning and prior scholarly works [15–17]. Covariates included age, gender, current marriage status (married or single), education level (no formal education, elementary school or lower, middle school, and high school or above), household income per capita (log transformed), smoking status, drinking status were collected through face-to-face interviews. BMI was categorized according to the World Health Organization (WHO) classification [18]: underweight (<18.5 kg/m2), normal weight (between 18.5 and less than 25 kg/m2), overweight (between 25 and less than 30 kg/m2), and obesity (≥30.0 kg/m2). Waist circumference was accessed using a soft measure tape while standing [14]. The diagnoses of hypertension and diabetes mellitus (DM) were based on self-reported medical history. The high sensitivity C-reactive protein (CRP) was measured by the immunoturbidimetric assay and the white blood cell (WBC) was measured by automated analyzers within 141 min [14].
Statistical analysis
Continuous variables were presented as means ± standard deviations (SDs and categorical variables were described as frequencies and proportions. Differences between groups were compared by independent sample t-text for continuous variables and chi-square test for categorical variables. As this analysis compares traits across independent groups (not within subjects over time), independence assumptions for t-tests/chi-square were satisfied.
In the correlation analysis, we used the multivariate logistic regression model to estimate the odds ratios (OR) with 95% confidence interval (95% CI) for the association of nighttime sleep duration with incident OA. We used three models in the analyses, of which Model 1 was the crude model. Model 2 adjusted for age and gender. Model 3 additionally adjusted current marriage status, education level, household income per capita, smoking and drinking status, BMI, waist circumference, hypertension, DM, CRP and WBC level based on model 2.
The study also examined the non-linear association between sleep duration and OA using restricted cubic spline. The saturation value of the connection between sleep duration and OA was determined using threshold effect analysis. We also conducted subgroup analysis by dividing the participants by sex and age in order to investigate the correlation between sleep duration and OA in various subgroups. Data were analyzed using R (version 3.4.3). P < 0.05 was chosen as the level of significance for the analysis.
Results
Baseline characteristics of study population
After applying the inclusion and exclusion criteria, a total of 6,825 participants were included in the study. The mean age of participants without OA was 57.6 ± 8.8 years, while for those with OA, it was 58.2 ± 8.7 years. Among participants with OA, 43.8% were female, while 56.2% were male. Individuals with OA more often reported sleeping < 6 h compared to participants without OA. A summary of the baseline characteristics of the subjects is provided in Table 1.
Associations between sleep duration and OA
Table 2 presents the analysis of the association between sleep duration and OA. The ORs 95% CI for OA is reported for different sleep durations, compared to appropriate sleep duration (7–9 h/night). In model 1, which remained unadjusted for any variables, the duration of sleep exhibited a statistically significant association with OA as a categorical variable. Participants who slept <6 h had a 1.48-fold higher risk of developing OA compared to those who slept 7–9 h (OR = 1.48, 95% CI: 1.28–1.70, p < 0.001). Conversely, there was no noteworthy variance in OA incidence between participants who slept for > 9 h and those who slept for 7–9 h (OR = 0.95, 95% CI [0.69–1.31] p = 0.756). In Model 3, after accounting for all covariates, participants who slept <6 h had a 1.39-fold higher risk of developing OA compared to those with appropriate sleep durations (OR = 1.39, 95% CI: 1.20–1.60, p < 0.001).
There were significant interactions among subgroups of sleeping <6 hours with overweight status (OR = 1.41, P = 0.042) and sleeping >9 hours with overweight status (OR = 2.71, P = 0.006). No significant interactions were observed in the underweight and obesity groups across different sleep durations (Table 3).
Analysis of threshold effects and smooth curve fitting
The threshold analysis identified 7.5 h as the inflection point, with a negative association below this point (β = −0.014, p < 0.05) and a positive association above it (β = 0.032, p < 0.05).
Fig 2(A) depicts a U-shaped association between the overall occurrence of OA and the duration of sleep. Non-linear correlation between sleep duration and OA in the categorized BMI groups are presented in Fig 2(B-E). Among individuals with different BMI levels, sleep duration influenced changes in OA occurrence in distinct ways. For those who were overweight, the U-shaped curve was more pronounced, whereas the obesity group exhibited an inverted U-shaped curve. In normal-weight individuals, the occurrence of OA gradually decreased as sleep duration increased. However, this variation appeared non-significant in the underweight group.
(A) overall sample, (B) underweight, (C) normal weight, (D) overweight, and (E) obesity. Abbreviations: OA, osteoarthritis; BMI, body mass index.
Subgroup analysis
Subgroup analysis indicated that the association between sleep duration and OA differed by sex and age group (Table 4). In males, compared to those with a sleep duration of 7–9 hours, individuals with short sleep duration (<6 hours) had a 30% higher likelihood of developing OA (OR = 1.30, 95% CI [1.04–1.62], p = 0.019). Among females, both short and insufficient sleep duration were significantly associated with an increased risk of OA. Those sleeping less than 6 hours had a 45% higher OA risk (OR = 1.45, 95% CI [1.20–1.76], p < 0.001), while those sleeping between 6 and 7 hours had a 31% increased risk (OR = 1.31, 95% CI [1.06–1.63], p = 0.012). However, prolonged sleep duration (>9 hours) was not significantly associated with OA in both males and females. Similar results like main models were also observed in the subgroup analysis by age.
Discussion
Through this cross-sectional analysis, we found short sleep (<7h) was associated with a higher incidence of OA compared to normal night-time sleep duration (7-9h). Additionally, we further identified a non-linear, U-shaped association between sleep duration and the risk of developing OA in middle-aged and older Chinese adults, and this association exhibits significant variations across different body weight subgroups. A sleep duration of approximately 7.5 hours appears optimal for reducing the risk of OA.
Our findings demonstrate a significant association between sleep duration and OA, consistent with previous studies. Importantly, several Asian cohort studies support this relationship. For example, Cho et al. analyzed 8,918 Korean adults aged ≥50 years and found that both short (≤6 hours/day) and long (≥9 hours/day) sleep durations were associated with higher OA prevalence, although the association with long sleep lost significance after adjustment. However, subgroup analysis showed that participants with both radiological OA and joint pain had increased odds of OA with both short and long sleep durations (OR=1.32 and 1.41, respectively) [19]. In a Chinese study, Zhou et al. (2024) observed that those with <6 hours or >9 hours of sleep had higher incidence of knee OA over four years, though the trend was stronger for short sleep duration [2]. These Asian studies closely align with our findings and suggest the possibility of a U-shaped association in regional populations.
In contrast, several European studies [8,20] observed similar U-shaped trends in Western cohorts, but often based on different lifestyle baselines. For instance, Jung et al., in a sample of 11,540 participants, reported increased OA prevalence with both <6 and ≥8 hours of sleep, though the association for long sleep did not reach significance [8]. These discrepancies may reflect cultural and behavioral differences. Notably, approximately 60% of older Chinese adults take regular daytime naps, compared to about 25% in the U.S., potentially affecting nighttime sleep duration measurements [21]. Further research is warranted to clarify the impact of prolonged sleep on OA development in different populations.
Our analysis also indicates that BMI moderates the association between sleep duration and OA risk. Previous work using the NHANES database found that 12.1% of the association between sleep duration and OA was mediated by waist circumference [13]. In our study, visualization of the BMI subgroups revealed that, although a U-shaped association was evident in the overall population, the curve was more pronounced in the overweight subgroup, whereas an inverted U-shaped relationship was observed in the obese subgroup. Normal-weight individuals exhibited a gradual decrease in OA incidence with increasing sleep duration, while the underweight group showed no significant variation. The inverted U-shaped curve in obese individuals is intriguing and may be explained by several factors: (1) the potential masking of the true effect by an underlying inflammatory state, whereby individuals with moderate sleep duration may be in a “balanced” condition; and (2) the higher prevalence of sleep disorders such as sleep apnea in overweight populations [22], which may complicate the relationship between reported sleep duration and actual sleep quality. Additionally, the obesity rate among Chinese older adults (approximately 9%) is significantly lower than that in the United States (41%) and Europe (20%−30%) [23]. This may partially explain why the effect of obesity in our study was less pronounced than that reported by Ma et al. [13]. These findings underscore the importance of developing targeted OA prevention and management strategies that consider variations in body weight.
Regarding underlying mechanisms, a prevailing explanation involves the activation of pro-inflammatory pathways and oxidative stress induced by extreme sleep durations, leading to increased expression of cytokines such as interleukin-6 (IL-6) and tumor necrosis factor-α (TNF-α) [24]. These cytokines are critical mediators in the degradation of cartilage matrix and the pathogenesis of OA [25]. Moreover, the relationship between poor sleep quality and OA may be bidirectional and complex, especially in the context of metabolic disorders such as obesity, type 2 diabetes [26], and hypertension [27]. Notably, overweight and obese individuals typically exhibit elevated plasma levels of TNF-α and IL-6 [28], which, in addition to increased mechanical stress on joints [29], may further contribute to cartilage damage and OA progression. Importantly, our study included multiple OA sites (e.g., knee, spine, hand), suggesting that chronic inflammation associated with abnormal sleep duration may exert a generalized effect on OA development.
Limitation
Several limitations of our study should be acknowledged. First, variables regarding sleep duration and OA were self-reported in CHARLS, without clinical examination or imaging confirmation, which may introduce recall bias and misclassification. Although objective measurements—such as actigraphy or the use of wearable devices—could provide more accurate and reliable data, these methods were not feasible for our large-scale cohort. Second, the analysis may be subject to residual confounding, as not all potential covariates were considered. Factors such as physical activity, sleep medication, circadian rhythm disruptions, and specific interventions were not fully accounted for, due to high rates of missing data or lack of availability, potentially influencing the observed association between sleep duration and OA. Additionally, this is a cross-sectional study, therefore failed to provide a robust causal inference.
Conclusion
Short sleep was associated with a higher incidence of OA and a U-shaped association was observed between sleep duration and OA incidence in middle-aged and older Chinese adults. BMI may act as a moderator in this relationship.
References
- 1. Pereira D, Ramos E, Branco J. Osteoarthritis. Acta Med Port. 2015;28(1):99–106.
- 2. Zhou S, Wu L, Si H, Li M, Liu Y, Shen B. Association between nighttime sleep duration and quality with knee osteoarthritis in middle-aged and older Chinese: a longitudinal cohort study. Arch Gerontol Geriatr. 2024;118:105284. pmid:38029546
- 3. Minnig MCC, Golightly YM, Nelson AE. Epidemiology of osteoarthritis: literature update 2022-2023. Curr Opin Rheumatol. 2024;36(2):108–12.
- 4. Xie J, Li Y, Zhang Y, Vgontzas AN, Basta M, Chen B, et al. Sleep duration and metabolic syndrome: an updated systematic review and meta-analysis. Sleep Med Rev. 2021;59:101451. pmid:33618187
- 5. Watson NF, Badr MS, Belenky G, Bliwise DL, Buxton OM, Buysse D, et al. Recommended amount of sleep for a healthy adult: a joint consensus statement of the American Academy of Sleep Medicine and Sleep Research Society. Sleep. 2015;38(6):843–4. pmid:26039963
- 6. Cappuccio FP, D’Elia L, Strazzullo P, Miller MA. Sleep duration and all-cause mortality: a systematic review and meta-analysis of prospective studies. Sleep. 2010;33(5):585–92. pmid:20469800
- 7. Wang T, He C. Pro-inflammatory cytokines: the link between obesity and osteoarthritis. Cytokine Growth Factor Rev. 2018;44:38–50. pmid:30340925
- 8. Jung JH, Seok H, Choi SJ, Bae J, Lee SH, Lee MH, et al. The association between osteoarthritis and sleep duration in Koreans: a nationwide cross-sectional observational study. Clin Rheumatol. 2018;37(6):1653–9. pmid:29470739
- 9. Ni J, Zhou W, Cen H, Chen G, Huang J, Yin K, et al. Evidence for causal effects of sleep disturbances on risk for osteoarthritis: a univariable and multivariable Mendelian randomization study. Osteoarthritis Cartilage. 2022;30(3):443–50. pmid:34890811
- 10. Duclos M. Osteoarthritis, obesity and type 2 diabetes: the weight of waist circumference. Ann Phys Rehabil Med. 2016;59(3):157–60. pmid:27211819
- 11. Theorell-Haglöw J, Berne C, Janson C, Sahlin C, Lindberg E. Associations between short sleep duration and central obesity in women. Sleep. 2010;33(5):593–8. pmid:20469801
- 12. Chaput J-P, Després J-P, Bouchard C, Tremblay A. Short sleep duration is associated with reduced leptin levels and increased adiposity: results from the Quebec family study. Obesity (Silver Spring). 2007;15(1):253–61. pmid:17228054
- 13. Ma G, Xu B, Wang Z, Duan W, Chen X, Zhu L, et al. Non-linear association of sleep duration with osteoarthritis among U.S. middle-aged and older adults. BMC Public Health. 2024;24(1):3565. pmid:39716177
- 14. Zhao Y, Hu Y, Smith JP, Strauss J, Yang G. Cohort profile: the China Health and Retirement Longitudinal Study (CHARLS). Int J Epidemiol. 2014;43(1):61–8. pmid:23243115
- 15. Zhu Z, Sheng X, Zhang J, Yao X. Association between type 2 diabetes status and osteoarthritis in adults aged ≥ 50 years. J Orthop Sci. 2022;27(2):486–91. pmid:33419624
- 16. Li X, Zhou Y, Liu J. Association between prediabetes and osteoarthritis: a meta-analysis. Horm Metab Res. 2022;54(2):104–12. pmid:35130571
- 17. Karvonen-Gutierrez CA, Sowers MR, Heeringa SG. Sex dimorphism in the association of cardiometabolic characteristics and osteophytes-defined radiographic knee osteoarthritis among obese and non-obese adults: NHANES III. Osteoarthritis Cartilage. 2012;20(7):614–21. pmid:22521953
- 18. Obesity: preventing and managing the global epidemic. Report of a WHO consultation. World Health Organ Tech Rep Ser. 2000;894:i–xii, 1–253. pmid:11234459
- 19. Cho Y, Jung B, Lee YJ, Kim M-R, Kim E-J, Sung W-S, et al. Association between sleep duration and osteoarthritis and their prevalence in Koreans: a cross-sectional study. PLoS One. 2020;15(4):e0230481. pmid:32339178
- 20. Park H-M, Kwon Y-J, Kim H-S, Lee Y-J. Relationship between sleep duration and osteoarthritis in middle-aged and older women: a nationwide population-based study. J Clin Med. 2019;8(3):356. pmid:30871193
- 21. Zhang Z, Xiao X, Ma W, Li J. Napping in older adults: a review of current literature. Curr Sleep Med Rep. 2020;6(3):129–35. pmid:33777656
- 22. Patel SR, Mehra R. The weighty issue of obesity management in sleep apnea. Chest. 2015;148(5):1127–9. pmid:26527436
- 23. Collaborators GBDO, Afshin A, Forouzanfar MH, Reitsma MB, Sur P, Estep K, et al. Health effects of overweight and obesity in 195 countries over 25 years. N Engl J Med. 2017;377(1):13–27.
- 24. Besedovsky L, Lange T, Haack M. The sleep-immune crosstalk in health and disease. Physiol Rev. 2019;99(3):1325–80. pmid:30920354
- 25. Kapoor M, Martel-Pelletier J, Lajeunesse D, Pelletier J-P, Fahmi H. Role of proinflammatory cytokines in the pathophysiology of osteoarthritis. Nat Rev Rheumatol. 2011;7(1):33–42. pmid:21119608
- 26. Cappuccio FP, D’Elia L, Strazzullo P, Miller MA. Quantity and quality of sleep and incidence of type 2 diabetes: a systematic review and meta-analysis. Diabetes Care. 2010;33(2):414–20. pmid:19910503
- 27. Lo K, Woo B, Wong M, Tam W. Subjective sleep quality, blood pressure, and hypertension: a meta-analysis. J Clin Hypertens (Greenwich). 2018;20(3):592–605. pmid:29457339
- 28. Coppack SW. Pro-inflammatory cytokines and adipose tissue. Proc Nutr Soc. 2001;60(3):349–56. pmid:11681809
- 29. Zhang H, Shao Y, Yao Z, Liu L, Zhang H, Yin J, et al. Mechanical overloading promotes chondrocyte senescence and osteoarthritis development through downregulating FBXW7. Ann Rheum Dis. 2022;81(5):676–86. pmid:35058228