Poor sleep and high rheumatoid arthritis risk: Evidence from large UK Biobank cohort

Yi-Qun Yang; Han-Wen Cao; Xing-Hao Yu; Lin Bo; Fei-Yan Deng; Shu-Feng Lei

doi:10.1371/journal.pone.0318728

Abstract

Objective

To evaluate the relationship between sleep behaviors and rheumatoid arthritis (RA) risk.

Methods

First, based on large-scale data from the UK Biobank, we analyzed the associations between multiple sleep factors and RA risk and calculated a poor sleep score (PSS) to assess their combined effect. Then we constructed genetic risk scores (GRS) based on a large genome-wide association study and assessed the interaction or combined effect of sleep and genetic factors on RA risk. Finally, we conducted a case-control study to explore the effect of RA on sleep.

Results

Sleep duration, getting up, napping during the day, insomnia, and daytime dozing were associated with the risk of RA, but no associations were observed for chronotype and snoring with RA. Participants in the high genetic risk and high PSS group had a 363.1% higher risk of developing RA compared to those with low genetic risk and low PSS. We also found that women were more likely than men to be affected by the combined effects of PSS and GRS. In the case-control study, there were statistically significant associations of RA with getting up, PSS grade and PSS.

Conclusions

Unhealthy sleep patterns were associated with increasing risk of RA among participants with low, intermediate, or high genetic risk.

Citation: Yang Y-Q, Cao H-W, Yu X-H, Bo L, Deng F-Y, Lei S-F (2025) Poor sleep and high rheumatoid arthritis risk: Evidence from large UK Biobank cohort. PLoS One 20(4): e0318728. https://doi.org/10.1371/journal.pone.0318728

Editor: Qian Wu, The First Affiliated Hospital of Soochow University, CHINA

Received: October 16, 2024; Accepted: January 22, 2025; Published: April 23, 2025

Copyright: © 2025 Yang 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: All relevant data are within the paper and its Supporting Information files. GWAS Summary data on sleep phenotypes and rheumatoid arthritis can be obtained from the corresponding articles, and we list the DOI numbers of the involved articles: sleep duration (https://doi.org/10.1038/s41467-019-08917-4); difficulty getting up (https://doi.org/10.1038/s41588-018-0333-3); chronotype (https://doi.org/10.1038/s41467-018-08259-7); nap during day (https://doi.org/10.1038/s41467-020-20585-3); insomnia (https://doi.org/10.1038/s41588-022-01124-w); snoring (https://doi.org/10.1038/s41467-020-14625-1); daytime dozing (https://doi.org/10.1038/s41467-019-11456-7); rheumatoid arthritis (https://doi.org/10.1038/nature12873). The UK Biobank genotype and phenotype data is available on application to the UK Biobank project (ID: 76875) at http://www.ukbiobank.ac.uk. Code is available at https://github.com/Cyanyyq/sleep_RA.

Funding: The study was supported by Natural Science Foundation of China (82373587, 82173529, 82173598, and 82103922) and a Project of the Priority Academic Program Development of Jiangsu Higher Education Institutions. The funder participated in the study design, data collection and analysis, decision to publish and manuscript writing aspects.

Competing interests: No authors have competing interests.

Introduction

Rheumatoid arthritis (RA) is a chronic systemic inflammatory disease characterized by synovitis. If left untreated, it can lead to joint deformity and loss of function. The pathogenesis of RA is still under investigation, but it may involve a combination of autoimmune factors, genetics, smoking, and other environmental influences [1]. Individuals with RA often experience pain, obesity, and poor sleep, which significantly reduce their quality of life compared to those without the condition [2,3].

Sleep is essential for maintaining human health [4,5]. While the relationship between RA and sleep has been extensively studied, it remains unclear whether poor sleep is a cause or consequence of RA. In most cases, sleep disturbances in RA patients are linked to pain. A Swedish case-control study found that sleep problems worsen as the duration of the disease is extended [6]. A cross-sectional, multicenter study involving 305 Belgian patients reported that poor sleep quality is associated with inadequate RA control, potentially due to pain [5]. Moreover, medications prescribed for RA treatment may further impact sleep quality [5,7,8]. Recent research has highlighted a strong connection between sleep and the immune system, which plays a crucial role in complex diseases [9]. During sleep, the immune system releases proteins called cytokines, some of which promote sleep. In the presence of infection or inflammation, the demand for certain cytokines increases [10]. Conversely, sleep deprivation may reduce the production of these protective cytokines [9], leaving the body with fewer antibodies and cells to combat infection [11].

The relationship between poor sleep and various complex diseases has been a focus of recent investigation. Studies have shown that women with extreme sleep durations are more likely to develop metabolic syndrome and exhibit higher metabolic syndrome severity scores. In contrast, women who consistently sleep for seven hours per night have the lowest severity scores for metabolic syndrome [12]. A meta-analysis revealed a U-shaped relationship between sleep duration and the risk of osteoporosis, with the lowest risk observed among individuals who slept for eight hours per night [13]. Additionally, a UK cohort study demonstrated that subjects who slept fewer than four hours or more than ten hours per night experienced significantly faster cognitive decline compared to those who slept seven hours per night [14]. Either too much or too little sleep increased all-cause mortality and cardiovascular events, with 7 hours of sleep per night having the lowest risk [15]. Several studies have also suggested that poor sleep might contribute to the development of RA [16–18]. However, most of these studies are limited by cross-sectional designs and small sample sizes, and have only established a correlation between sleep and RA. There is a lack of research on whether sleep characteristics modulate the relationship between genetic factors and RA. Importantly, the interaction between sleep characteristics and genetic factors in determining RA risk remains largely unknown.

A composite measure of sleep health, known as the sleep score, was recently proposed. A prospective study examined the combined association of five sleep behaviors—sleep duration, chronotype, insomnia, snoring, and excessive daytime sleepiness—in the form of a healthy sleep score, with the risk of cardiovascular disease (CVD) [19].

Building on this concept, we prospectively assessed the effects of seven sleep behaviors on the risk of developing RA. To better understand the combined effect of poor sleep patterns and high genetic risk on RA, we followed [19] and constructed a poor sleep score (PSS) based on key sleep behaviors. Additionally, we examined the mediation effect of inflammation as a potential pathway from PSS to RA incidence. Furthermore, we explored the combined influence of sleep characteristics and genetic susceptibility on RA risk, and investigated potential gene-environment interactions. Finally, we conducted a case-control study using baseline data to investigate the impact of RA on sleep. The findings of our research are illustrated in Fig 1.

Download:

Fig 1. Overflow of our research.

https://doi.org/10.1371/journal.pone.0318728.g001

Materials and methods

Study design

We first examined the associations between multiple sleep behaviors and RA risk in a prospective study, calculating the PSS for each participant to evaluate its effect on RA risk. The participants were then stratified based on their calculated sleep scores and Genetic Risk Scores (GRS) to investigate the interaction and combined effects of sleep and genetic factors on RA risk. Additionally, a case-control study was conducted to assess the impact of RA onset on sleep quality in later life. Finally we used linkage disequilibrium scores regression (LDSC) [20] to assess the genetic association between sleep behaviors and RA.

UK Biobank data sets

The UK Biobank is a large biomedical database initiated in 2006, containing comprehensive genetic and health data from 500,000 UK participants aged 40 to 70 (http://www.ukbiobank.ac.uk/). The study has collected, and continues to collect, detailed genotypic and phenotypic information, including questionnaires, physical measurements, multimodal imaging, and data for a wide range of health-related longitudinal follow-up investigations [21]. Participants from 22 assessment centers with each eligible participant completing a written informed consent form were included at baseline. This database offers extensive health-related information, including biological measurements, lifestyle indicators, and biomarkers in blood and urine. It provides researchers with a unique opportunity to explore the combined effects of genes and the environment on complex diseases, offering a novel approach to understanding disease development. The present analyses were conducted under UK Biobank application number 76875. Detailed information is as follows:

Ethics approval statement.

Ethical approval was granted by the Northwest Centre for Research Ethics Committee, and informed consent was obtained from all participants.

Definition of RA cases.

Rheumatoid arthritis (RA) was defined according to the International Classification of Diseases, Tenth Revision (ICD-10-CM), as recorded in the UK Biobank (coded as “M05” and “M06”), with seropositive RA defined by “M05.” The outcome event was determined by the onset of RA from baseline to the date of diagnosis, death, loss to follow-up, or January 1, 2022, whichever occurred first. Self-reported non-cancer illness (coded as “1524”) was used to identify RA cases at baseline, which were included in case-control studies and excluded from prospective studies.

Definition of sleeping condition.

The data on sleeping habits of the participants was obtained through detailed self-reported questionnaires collected by the UK Biobank. A total of seven sleeping phenotypes were included in our study: sleep duration (Field ID: 1160), difficulty getting up (Field ID: 1170), chronotype (Field ID: 1180), nap during the day (Field ID: 1190), sleeplessness/insomnia (Field ID: 1200), snoring (Field ID: 1210), daytime dozing/sleeping (Field ID: 1220). The definitions of these sleep traits are provided in S1 Table. Poor sleep factors were defined as follows: late chronotype (‘evening’ or ‘evening than morning’); sleep less than 7h or more than 8h per day; reported ‘sometimes’ or ‘usually’ insomnia symptoms; self-reported snoring; and frequent daytime sleepiness (‘usually’). The PSS was generated, incorporating five sleep factors (chronotype, duration, insomnia, snoring, and excessive daytime sleepiness) to better assess an individual’s sleep status and to accurately classify the population. The PSS ranges from 0 to 5, with higher scores indicating a poorer sleep pattern. Sleep patterns were categorized as ‘healthy sleep pattern’ (PSS ≤ 1), ‘intermediate sleep pattern’ (2 ≤ PSS ≤ 3), and ‘poor sleep pattern’ (PSS ≥ 4) based on the constructed PSS. Additionally, a weighted PSS (PSS_CV) was created using a 10-fold cross-validation procedure as a sensitivity analysis. The weighted PSS was derived by combining the five sleep factors.

Summary-level data and construction of genetic risk score (GRS)

Genetic data on sleep behaviors (sleep duration [22], difficulty getting up [23], chronotype [24], nap during day [25], insomnia [26], snoring [27], daytime dozing [28]) were utilized in the largest relevant genome-wide study of its kind to date.

In the genetic analysis, we randomly divided the total sample into a training set (20%) and a test set (80%). The training set was used to identify the best-performing GRS, and all subsequent analyses were conducted using the test set. The GRS was calculated based on genotypes of selected single nucleotide polymorphisms (SNP) from UK Biobank under different thresholds (P from 5 × 10⁻⁸ to 5 × 10⁻⁴), where LD r² was set to 0.001 within 1Mb physical distance. The r² was derived from the 1000 Genomes Projects for individuals of European ancestry. We applied strict quality control measures to both genotypes and individuals. For genotypes, variants were retained if the minor allele frequency (MAF) was greater than 0.02, the imputation information score was greater than 0.5, the missing genotype rate was less than 5%, and the Hardy-Weinberg equilibrium P-value was greater than 1 × 10 ⁻ ⁷ [21]. The GRS for each individual is calculated by summing the product of genotypes (0, 1, 2) of risk loci and their corresponding weights, which were from the largest GWAS analysis for RA of European populations [29]. Additionally, we normalized the GRS to a mean of 0 and a standard deviation of 1. We restricted the analysis to approximately 330,000 individuals with successful genotyping, of white British ancestry, and without identified relatives. The best-performing GRS was defined as the score with the strongest association with RA incidence. Finally, genetic risk groups were categorized based on the selected GRS: low genetic risk (bottom quartile), intermediate genetic risk (2nd and 3rd quartiles), and high genetic risk (top quartile). Detailed information on the SNPs used for the GRS is provided in our previous work [30,31].

Statistical analysis

The basic characteristics of the 379,314 participants were summarized using the mean (standard deviation) for continuous variables and frequencies (percentages) for categorical variables. A Cox regression model was employed to assess the effect of various sleep phenotypes on the risk of RA over time, calculating hazard ratios (HRs) and their corresponding 95% confidence intervals (CIs). Three models were constructed to examine the relationship between sleep behavior and RA risk in this observational study: Model 1, adjusted for genotyping batch (UKB vs. BiLEVE), assessment center, sex, Townsend Deprivation Index (TDI), age, and kinship; Model 2, adjusted for batch, assessment center, sex, TDI, age, kinship, body mass index (BMI), smoking status, and alcohol consumption; Model 3, adjusted for batch, assessment center, sex, TDI, age, kinship, BMI, smoking status, alcohol consumption, diet, physical activity (PA), diastolic blood pressure (DBP), systolic blood pressure (SBP), high-density lipoprotein (HDL) cholesterol, low-density lipoprotein (LDL) cholesterol, triglycerides, and cancer. The TDI is a regionally-based proxy for socioeconomic status. A restricted cubic spline curve and a linearity test were used to assess the non-linear relationship between continuous variables, such as sleep duration and the PSS, and the risk of RA. Incidence rates in different subgroups were evaluated in a prospective study, and Harrell’s c-index was used to evaluate model performance. To assess the potential mediated role of chronic inflammation between sleep and RA, we used an inflammation-based score, the modified Glasgow Prognostic Score (mGPS). Participants with both elevated serum C-reactive protein (CRP) (>1.0 mg/dL) and hypoalbuminemia (<3.5 g/dL) were assigned an mGPS of 2; those with one abnormal value were assigned an mGPS of 1, and those with neither abnormality were assigned an mGPS of 0 [32]. Covariates used in GRS analysis included age, sex, TDI, genotyping batch, and top 10 genetic principal components (PCs). Additionally, we plotted the cumulative incidences of RA based on genetic and sleep risk in this cohort.

In case-control research, propensity score matching (PSM) was used to select participants for the control group, adjusting for genotyping batch, assessment center, sex, TDI, age, race, BMI, smoking status, alcohol consumption, diet, and physical activity. Linear regression was applied to examine the effect of RA onset on sleep duration, while ordinary logistic regression assessed the impact of RA onset on morning wakefulness, chronotype (morning/evening person), daytime napping, insomnia, daytime drowsiness, and PSS. Multivariable logistic regression was then used to evaluate the effect of RA onset on snoring.

Additionally, we plotted the cumulative incidences of RA according to genetic and sleep risk. Statistical significance was defined by a two-tailed P-value < 0.05. To account for multiple comparisons, Bonferroni correction was applied, with a significance threshold of P < 0.006 (0.05/8 sleep behaviors, including PSS). All statistical analyses were conducted using R (version 3.6.1) and PLINK software (version 1.90 b3.38) [33].

Results

Baseline characteristics

This study incorporated 379,314 subjects (comprising 171,207 men and 208,107 women), among whom 4,610 were diagnosed with RA. Table 1 shows the baseline characteristics of the study participants in the different PSS groups. Of all participants, 7.09%, 30.40%, 38.45%, 19.75% and 4.30% had a PSS of 0, 1, 2, 3 and 4–5, respectively (S2 Table). As PSS increased, the proportion of regular physical activity and a healthy diet decreased. In contrast, BMI was positively associated with PSS, while the risks of chronic kidney disease, type 2 diabetes, cancer, and RA increased with an increase in PSS.

Download:

Table 1. Baseline characteristics of the study subjects in the UK Biobank data.

https://doi.org/10.1371/journal.pone.0318728.t001

Association between sleep behaviors and RA risk

Poor sleep conditions and RA risk.

In the prospective study, U-shaped relationships between sleep duration and RA risk were observed (P_{non-linearity} < 0.001) by using RCS method in the primary analysis (Model 1), with 7h of sleep duration having the lowest risk of RA (Fig 2A). Similar U-shaped associations were observed in sex-stratified groups (P_{non-linearity} < 0.001) (Fig 2B and 2C). Participants with short sleep duration (less than 7 hours) and long sleep duration (more than 8 hours) had a higher risk of RA than those who slept 7 to 8 hours per night after multiple test adjustments (HR_{sleep duration < 7} = 1.389, 95%CI = 1.303 ~ 1.480; HR_{sleep duration > 8} = 1.482, 95%CI = 1.352 ~ 1.624) (Table 2). After correcting for more covariates (e.g., BMI, physical activity, and diabetes), the association remained statistically significant in Models 2 and 3. An association between insomnia and RA risk was observed in Model 1. Compared to never/rarely insomnia, the HR estimate for sometimes insomnia was 1.168 (95%CI = 1.077 ~ 1.267) and for frequent insomnia was 1.628 (95%CI = 1.498 ~ 1.769). After correcting for more covariates, the association is barely weakened. In the fully adjusted model (model 3), usually insomnia was associated with incident RA with a 54.4% higher risk (HR: 1.544, 95% CI: 1.404 ~ 1.698). When compared to people who never or seldom encountered daytime dozing, those who usually experienced had a higher risk of developing RA, with the HRs (95% CI) estimated to be 1.683 (1.466 ~ 1.932) in model 1, which is significant after multiple test adjustments (P = 1.58 × 10⁻¹³). Associations were still significant after controlling for more covariates in Models 2 and 3. Besides, we found no evidence of an increased risk of RA due to chronotype and snoring.

Download:

Table 2. The adjusted hazard ratio for sleep with the risk of RA in the UK Biobank data.

https://doi.org/10.1371/journal.pone.0318728.t002

Download:

Fig 2. Restricted cubic spline models for the relationship between sleep duration and RA risk.

(A) All participates; (B) In men; (C) In women. Note: Associations were adjusted for age, sex, genotyped batch, assessment center, kinship, TDI. Abbreviation: RA, rheumatoid arthritis; TDI, Townsend deprivation index.

https://doi.org/10.1371/journal.pone.0318728.g002

Daytime napping and difficulty of getting up and RA risk.

Similarly, naps during day and difficulty in getting up might also contribute to the development of RA. In Model 3, compared to those who never/rarely nap during day, participants who sometimes and usually nap had a 7.8% (HR: 1.078, 95%CI: 1.007 ~ 1.154) and 35.8% (HR: 1.358, 95%CI: 1.197 ~ 1.540) higher risk of developing RA. Interestingly, we also found that difficulty in getting up was also positively associated with the risk of RA, where subjects who got up not at all easy had around 2 times higher risk of RA than those who got up very easily in all models.

Sleep behaviors and seropositive RA risk.

We performed a sensitivity analysis to evaluate associations between sleeping conditions and seropositive RA risk. As shown in S3 Table, two sleeping problems (i.e., long sleep duration and insomnia) were also associated with an increased risk of seropositive RA after multiple test adjustments. A long duration of sleep (sleeping ≥ 8h) was associated with an increased risk of seropositive RA. Usually insomnia have a higher risk of RA compared with never/rarely insomnia. Besides, difficulty getting up and daytime napping is also associated with an increased risk of seropositive RA.

Association between poor sleep score (PSS) and RA risk

When a PSS was used to evaluate the combined effects of five sleep behaviors (i.e., sleep duration, insomnia, chronotype, snoring, and daytime dozing) on the risk of RA, we observed linear relationships between PSS and RA risk overall by using RCS method (Fig 3A), as well as sex-stratified groups (S1 Fig). As shown in S4 Table, participants with high PSS have a 70.2% (HR = 1.722, 95% CI: 1.487 ~ 1.949) increased risk of RA compared to those with low scores. Similar results were observed in the different PSS scores (S5 Table) and PSS calculated by cross-validation (PSS_CV) (S2 Fig). The sensitivity analysis yielded consistent results that people with high PSS_CV scores were 60.6% (HR = 1.606, 95%CI: 1.468 ~ 1.757) more likely to develop RA than those with low PSS_CV scores. The mediation analyses showed that 2.47% of the effect of PSS on the risk of RA was mediated by mGPS (P = 0.002) (S3 Fig). PSS can increase mGPS levels and further increase RA risk (β_{PSS → mGRS} = 0.020, β_{mGRS → RA risk} = 0.197). We also observed an independent effect of PSS on RA risk after correction for mGPS (β_{Direct effect} = 0.154). Linear relationships were also observed in associations between PSS and seropositive RA risk in overall and women (P _{non-linearity} > 0.05) but not in men (P_{non-linearity} = 0.038) (S3 Fig). As shown in S3 Table, the association between intermediate PSS and seropositive RA no longer existed. Similar results were observed in the different PSS scores (S4 Table) and PSS_CV (S3 Table). The cumulative incidence of RA in different PSS groups was shown in Fig 3B.

Download:

Fig 3. Relationship between poor sleep score (PSS) and RA risk.

(A) Restricted cubic spline models for the relationship between PSS and RA risk; (B) Standardized rates of RA events in low (bottom quartile), intermediate (quartiles 2 to 3), and high (top quartile) PSS groups in the UKB cohort; (C) Standardized rates of RA events in participants with different genetic risk score groups and different PSS groups in the UKB cohort. Note: Associations in A were adjusted for age, sex, genotyped batch, assessment center, kinship, and TDI. Associations in B and C were adjusted for genotype batch, assessment centre, sex, TDI, age, kinship, BMI, smoking status, alcohol, diet, PA, DBP, SBP, HDL cholesterol, LDL direct, Cholesterol, Triglycerides, and Cancer. Abbreviations: HR, hazard ratio; TDI, Townsend deprivation index; PA, physical activity; DBP, diastolic blood pressure; SBP, systolic blood pressure; HDL, high-density lipoprotein; LDL, low-density lipoprotein; PSS, poor sleep score.

https://doi.org/10.1371/journal.pone.0318728.g003

The combined effect of sleeping traits and genetic factors on RA risk

After adjustment, we chose the GRS with P < 5 × 10⁻⁷ to represent the genetic component, which had the best performance in terms of the significance of associations (S6 Table). As shown in S7 Table, among all participants, with worsening insomnia, snoring, and daytime dozing, the risk of RA was significantly increased in the moderate GRS group and high GRS group, and participants with high GRS and the worst sleep quality had the highest risk of RA (HR_insomnia = 3.652, 95%CI = 2.752 ~ 4.846, HR_snoring = 2.306, 95%CI = 1.918 ~ 2.773, HR_{daytime dozing} = 3.805, 95%CI = 2.672 ~ 5.416). For sleep duration and chronotype, with a certain GRS, we also found a U-shaped relationship between sleep duration and RA risk, with the lowest risk observed for 7–8 hours of sleep per night (in the high GRS group, HR_{sleep duration7 ~ 8} = 2.109, 95%CI = 1.789 ~ 2.485, HR_{sleep duration < 7} = 3.158, 95%CI = 2.616 ~ 3.811, HR_{sleep duration > 8} = 2.888, 95%CI = 2.218 ~ 3.761), an evening person were more likely to suffer from RA than morning person (in the high GRS group, HR_{definitely morning} = 2.045, 95%CI = 1.583 ~ 2.642, HR_{morning more} = 2.051, 95%CI = 1.606 ~ 2.619, HR_{evening more} = 2.268, 95%CI = 1.764 ~ 2.916, HR_{definitely evening} = 2.410, 95%CI = 1.741 ~ 3.336). Furthermore, given the same sleep conditions, we found that the high GRS group had a higher risk of RA than the medium GRS group, whereas no such results were observed in the low GRS group (except for sleep duration). In the sex-stratified analysis, the combined effect of sleep factors and GRS was greater in women than in men.

We next assessed the combined association of PSS and GRS on the risk of developing RA, detailed results were presented in Fig 4. According to our findings, the combined effect of GRS and PSS enhanced the chance of RA incident. We also discovered that subjects with high PSS and high GRS had the highest risk of developing RA (HR = 4.062, 95%CI: 2.792 ~ 5.910). Results stratified by gender showed that women were more likely than men to be affected by the combined effects of PSS and GRS. In the high PSS and high GRS group, the HRs (95% CI) were estimated to be 3.880 (2.230 ~ 6.750) for men, and 4.072 (2.440 ~ 6.797) for women. Similar results were also observed with the PSS calculated by using cross-validation method (S4 Fig). Fig 3C shows the cumulative incidence of RA events in participants with different GRS groups and different PSS groups. C-statistic was calculated to be 0.657 in model without GRS and 0.672 in model with GRS (2.2% improvement).

Download:

Fig 4. RA risks in the subgroups stratified by genetic risk and poor sleep score (PSS) (versus participants with the lowest PSS in the lowest genetic risk group) in the UKB cohort.

Note: Associations were adjusted for age, sex, TDI, genotyping batch, and top 10 genetic PCs. Abbreviations: PSS, poor sleep score; CI, confidence interval; HR, hazard ratio; PC, principal components; GRS, genetic risk score.

https://doi.org/10.1371/journal.pone.0318728.g004

Effect of RA onset on sleep condition

As shown in S8 Table, in the case-control study, there were statistically significant associations between RA onset and poor sleep conditions (i.e., PSS) (β = 0.191, 95%CI = 0.095 ~ 0.288). Besides, ordinal regression analysis indicated that RA onset could increase the risk of difficulty getting up (OR = 1.293, 95%CI = 1.156 ~ 1.446,). However, no significant associations were observed between RA onset and other sleep behaviors, including insomnia, sleep duration, and daytime drowsiness, in later life.

Genetic correlations between RA and sleep phenotypes

In the analysis of genetic correlations using summary data from GWAS, we observed positive genetic associations between RA and insomnia (rg = 0.13; P = 0.0002) and chronotype (rg = 0.1; P = 0.00008), as well as negative genetic associations with sleep duration. Similar results were observed in seropositive RA (S5 Fig provides detailed information).

Discussion

This study, utilizing large-scale data from the UK Biobank, this study systematically explored the relationship between seven sleep behaviors and RA risk. The prospective study indicated that abnormal sleep duration, insomnia, and daytime dozing could increase the risk of RA. A U-shaped relationship between sleep duration and RA risk, with the lowest risk of RA when the daily sleep duration was 7–8 hours. Additionally, a PSS was developed to evaluate the combined effects of five poor sleep behaviors, revealing that higher PSS values were linked to an increased RA risk. In particular, individuals in the highest PSS group were found to have a 70.2% greater likelihood of developing RA. We also conducted a mediation analysis, which revealed that the mGPS acted as a mediator between sleep behaviors and RA risk. Furthermore, we observed the combined effects of genetic factors and PSS on RA risk, with sex-stratified analysis showing a stronger effect in women than in men. The case-control study further confirmed that RA risk is associated with long sleep duration, insomnia, and difficulty waking up in the morning. Lastly, significant genetic correlations were identified between sleep phenotypes and various types of RA.

Our findings provide robust evidence supporting the association between poor sleep and an elevated risk of RA, particularly with abnormal sleep duration, insomnia, and daytime drowsiness. A study conducted in China similarly found that poor sleep quality and short sleep duration increase RA risk [34], consistent with our results. However, we identified a U-shaped relationship between sleep duration and RA risk, with the lowest risk found in individuals who slept 7–8 hours per day. This pattern is consistent with studies on other diseases and sleep duration. For instance, research by Abbas Smiley et al. showed that women with extreme sleep durations had an increased likelihood of developing metabolic syndrome, with the lowest severity scores observed among those sleeping 7 hours per night [12]. Similarly, a meta-analysis reported a U-shaped relationship between sleep duration and osteoporosis risk, with the lowest risk for 8 hours of sleep per night [13]. A cohort study of 65,754 patients in Taiwan found a link between sleep disorders and a higher risk of RA [16], and a study by a conclusion supported by Taiwan’s National Health Insurance Scheme [17]. A UK cohort study revealed that faster cognitive decline in individuals who slept fewer than 4 hours or more than 10 hours per night, compared to those who slept 7 hours [14]. Jiawei Yin et al. also demonstrated that both insufficient and excessive sleep increase all-cause mortality and cardiovascular events, with 7 hours of sleep per night associated with the lowest risk [15]. Notably, we also observed significant associations between poor sleep patterns and RA risk in seropositive RA, especially long sleep duration. Furthermore, a cohort study from the Taiwan National Health Insurance Research Database suggested that obstructive sleep apnea, a sleep disorder linked to snoring, may increase RA risk, although treating OSA could mitigate this risk [35]. However, we did not find a significant association between snoring and RA risk, which may be due to the self-reported nature of snoring data and the lack of detailed information such as snoring frequency and duration.

The PSS constructed in our study is essential for assessing individuals’ sleep patterns. Sleep behaviors are interrelated and can influence one another, with the body holistically regulating sleep. Changes in one sleep factor often led to compensatory changes in other [36,37]. For example, insomnia could result in shorter sleep duration and increased daytime drowsiness. The PSS developed here provides a comprehensive measure of sleep behavior, capturing the combined impact of five poor sleep factors on RA risk. This score offers a useful framework for assessing overall sleep quality, allowing for more accurate identification of high-risk individuals. Moreover, the simplicity of the scoring algorithm enhances the clarity and applicability of epidemiological findings, increasing their information value [36].Recent studies investigating the relationship between RA and sleep, such as that by Liu et al. [38], have provided valuable insights. Our study contributes to this field by introducing several innovative approaches. First, we included a sensitivity analysis focused on seropositive RA, enriching our understanding of the impact on different RA subtypes. Second, we performed a mediation analysis, which demonstrated that the mGPS acts as a mediator between sleep behaviors and RA risk. Third, we conducted a PSM case-control study to minimize confounding, thus strengthening the reliability of our results. Finally, we employed LDSC to explore genetic correlations between sleep behaviors and RA, uncovering potential shared genetic mechanisms. These methodological innovations help further elucidate the complex interactions between sleep patterns, genetic predisposition, and RA.

Several mechanisms may link sleep disturbances to RA risk, including chronic inflammation. We found that the mGPS, an inflammation-based score, mediated the relationship between sleep and RA risk, although the effect was modest. Sleep and immune system disorders are linked in both directions. Sleep disturbances and immune system disorders are bidirectional: a disrupted immune system can impair sleep, while poor sleep, in turn, can affect immune function [39]. Previous studies have suggested that insomnia and other non-apnea sleep disorders may increase the risk of autoimmune diseases, including RA, systemic lupus erythematosus, and ankylosing spondylitis [17]. Chronic low-grade inflammation associated with poor sleep is thought to contribute to RA risk [40]. Immunological studies indicate that sleep deprivation can elevate the number of circulating T cells and B cells, while disrupting the rhythm of CD4+ regulatory T cells (Tregs) [41,42]. Activated T cells can stimulate osteoclasts, leading to joint destruction. Additionally, the interaction between T cells and B cells can perpetuate the degradation of the synovial membrane [43]. Furthermore, poor sleep can upregulate inflammatory markers such as tumor necrosis factor-alpha (TNF-α), interleukin (IL)-1, IL-6, IL-8, and IL-17 [44,45]. The finding that women generally exhibit higher levels of inflammatory markers aligns with our observation that women with poor sleep quality may be at greater risk for RA [46]. Dysregulation of IL-6 and TNF-α is known to disrupt the balance between cartilage and bone matrix degradation in joints [47]. Notably, an animal study confirmed that IL-17 levels remained elevated in sleep-deprived rats even after 7 days of recovery [48]. IL-17 not only triggers an inflammatory response but also enhances RANKL sensitivity by modulating RANK expression, which increases osteoclast activity and bone resorption [49].

Strength and limitations

This study has several strengths. First, we assessed the relationship between sleep and various types of RA by integrating prospective cohort and PSM case-control studies. Second, we developed a sleep score to evaluate the association between overall sleep patterns and RA risk, incorporating multiple sleep behaviors. Third, we explored the combined effects of sleep patterns and genetic risk on RA. Additionally, we identified significant genetic correlations between sleep phenotypes and different RA subtypes. However, our study also has limitations. First, it was conducted among European Caucasians, limiting the generalizability of the findings to other racial groups. Second, sleep patterns were assessed only at baseline, and the use of a single baseline measure to create the sleep score does not account for changes in sleep behaviors over time. Future studies should incorporate multiple time points to better understand how changes in sleep behavior affect RA risk. Third, while we adjusted for known confounders, there may still be unmeasured or unknown confounding factors. Finally, the sleep data were self-reported, introducing potential measurement errors.

Supporting information

S1 Fig. Restricted cubic spline models for the relationship between PSS and RA risk in men (A) and women (B).

Note: Associations were adjusted for age, sex, genotyped batch, assessment center, kinship, and TDI. Abbreviation: TDI, Townsend deprivation index.

https://doi.org/10.1371/journal.pone.0318728.s001

(TIF)

S2 Fig. The contribution of mGPS for the association between PSS and RA risk.

Note: * p < 0.05, **p < 0.01, ***p < 0.001. Associations were adjusted for age, sex, genotyped batch, assessment center, kinship, and TDI. Abbreviation: RA, rheumatoid arthritis; PSS, poor sleep score; mGPS, modified Glasgow Prognostic Score; TDI, Townsend deprivation index.

https://doi.org/10.1371/journal.pone.0318728.s002

(TIF)

S3 Fig. Restricted cubic spline models for the relationship between PSS and seropositive RA risk.

(A) All participants; (B) In men; (C) In women. Note: Associations were adjusted for age, sex, genotyped batch, assessment center, kinship, and TDI. Abbreviation: TDI, Townsend deprivation index.

https://doi.org/10.1371/journal.pone.0318728.s003

(TIF)

S4 Fig. RA risks in the subgroups stratified by genetic risk and PSSCV (versus participants with lowest PSS in the lowest genetic risk group) in the UKB cohort.

Note: Associations were adjusted for age, sex, TDI, genotyping batch, and top 10 genetic PCs. Abbreviations: PSS, poor sleep score; CI, confidence interval; HR, hazard ratio; PC, principal components; GRS, genetic risk score; CV, cross-validation.

https://doi.org/10.1371/journal.pone.0318728.s004

(TIF)

S5 Fig. Genetic correlation of sleep phenotypes with different types of RA (P < 0.05).

Abbreviation: RA; rheumatoid arthritis.

https://doi.org/10.1371/journal.pone.0318728.s005

(TIF)

S1 Table. Description and definition of sleeping behaviors in UK Biobank data.

https://doi.org/10.1371/journal.pone.0318728.s006

(PDF)

S2 Table. Baseline characteristics of the study subjects stratified by different PSS scores in the UK Biobank data.

Abbreviation: SD, standard deviation; RA, rheumatoid arthritis; BMI, body mass index; PSS, poor sleep score.

https://doi.org/10.1371/journal.pone.0318728.s007

(PDF)

S3 Table. The adjusted hazard ratio for sleep behaviors on the risk of seropositive RA in the UK Biobank data.

Note: Model 1: Associations were adjusted for genotype batch, assessment centre, sex, TDI, age, kinship; Model 2: Associations were adjusted for genotype batch, assessment centre, sex, TDI, age, kinship, BMI, smoking status, alcohol; Model 3: Associations were adjusted for genotype batch, assessment centre, sex, TDI, age, kinship, BMI, smoking status, alcohol, diet, PA, DBP, SBP, HDL cholesterol, LDL direct, Cholesterol, Triglycerides, Cancer. Abbreviations: CI, confidence interval; HR, hazard ratio; TDI, Townsend deprivation index; BMI, body mass index; PA, physical activity; DBP, diastolic blood pressure; SBP, systolic blood pressure; HDL, high-density lipoprotein; LDL, low-density lipoprotein; PSS, poor sleep score; CV, cross-validation.

https://doi.org/10.1371/journal.pone.0318728.s008

(PDF)

S4 Table. The adjusted hazard ratio for RA and seropositive RA risk stratified by three PSS categories in the UK Biobank data

. Note: Model 1: Associations were adjusted for genotype batch, assessment centre, sex, TDI, age, kinship; Model 2: Associations were adjusted for genotype batch, assessment centre, sex, TDI, age, kinship, BMI, smoking status, alcohol; Model 3: Associations were adjusted for genotype batch, assessment centre, sex, TDI, age, kinship, BMI, smoking status, alcohol, diet, PA, DBP, SBP, HDL cholesterol, LDL direct, Cholesterol, Triglycerides, Cancer. Abbreviations: CI, confidence interval; HR, hazard ratio; TDI, Townsend deprivation index; BMI, body mass index; PA, physical activity; DBP, diastolic blood pressure; SBP, systolic blood pressure; HDL, high-density lipoprotein; LDL, low-density lipoprotein; PSS, poor sleep score; CV, cross-validation.

https://doi.org/10.1371/journal.pone.0318728.s009

(PDF)

S5 Table. The adjusted hazard ratio for RA and seropositive RA risk stratified by five PSS categories in the UK Biobank data.

Note: Model 1: Associations were adjusted for genotype batch, assessment centre, sex, TDI, age, kinship; Model 2: Associations were adjusted for genotype batch, assessment centre, sex, TDI, age, kinship, BMI, smoking status, alcohol; Model 3: Associations were adjusted for genotype batch, assessment centre, sex, TDI, age, kinship, BMI, smoking status, alcohol, diet, PA, DBP, SBP, HDL cholesterol, LDL direct, Cholesterol, Triglycerides, Cancer. Abbreviations: CI, confidence interval; HR, hazard ratio; TDI, Townsend deprivation index; BMI, body mass index; PA, physical activity; DBP, diastolic blood pressure; SBP, systolic blood pressure; HDL, high-density lipoprotein; LDL, low-density lipoprotein; PSS, poor sleep score.

https://doi.org/10.1371/journal.pone.0318728.s010

(PDF)

S6 Table. The adjusted hazard ratio for the effect of GRSs in different cutoffs on RA risk in both training set and test set.

Note: Associations were adjusted for age, sex, TDI, genotyping chip (UKB vs BiLEVE), and top 10 genetic PCs. Abbreviations: CI, confidence interval; HR, hazard ratio; GRS, genetic risk score; TDI, Townsend deprivation index; PC, principal components.

https://doi.org/10.1371/journal.pone.0318728.s011

(PDF)

S7 Table. RA risks in the subgroups stratified by genetic risk and different sleep behaviors (versus participants with healthy sleep behaviors in the lowest genetic risk group) in the UK Biobank.

Note: Associations were adjusted for age, sex, TDI, genotyping batch, and top 10 genetic PCs. Abbreviations: CI, confidence interval; HR, hazard ratio; PC, principal components; GRS, genetic risk score.

https://doi.org/10.1371/journal.pone.0318728.s012

(PDF)

S8 Table. The effect of RA onset on sleep condition in the UK Biobank data.

Note: # represents the utilization of linear regression due to the continuous Abbreviations: CI, confidence interval; PSS, poor sleep score; CV, cross-validation.

https://doi.org/10.1371/journal.pone.0318728.s013

(PDF)

Acknowledgments

We thank previous studies for making the summary association statistics data publicly available and we are grateful to all the investigators and participants in UK Biobank for the public individual-level data sets and their contribution to those studies. UK Biobank approval was given for this project (ID: 76875) and can be downloaded from (https://www.ukbiobank.ac.uk/ or https://biota.osc.ox.ac.uk/).

References

1. Smolen JS, Aletaha D, McInnes IB. Rheumatoid arthritis. Lancet. 2016;388(10055):2023–38. pmid:27156434
- View Article
- PubMed/NCBI
- Google Scholar
2. Clynes MA, Jameson K, Prieto-Alhambra D, Harvey NC, Cooper C, Dennison EM. Impact of rheumatoid arthritis and its management on falls, fracture and bone mineral density in UK Biobank. Front Endocrinol (Lausanne). 2019;10:817. pmid:31849841
- View Article
- PubMed/NCBI
- Google Scholar
3. Glanville KP, Coleman JRI, O’Reilly PF, Galloway J, Lewis CM. Investigating pleiotropy between depression and autoimmune diseases using the UK Biobank. Biol Psychiatry Glob Open Sci. 2021;1(1):48–58. pmid:34278373
- View Article
- PubMed/NCBI
- Google Scholar
4. Besedovsky L, Lange T, Born J. Sleep and immune function. Pflugers Arch. 2012;463(1):121–37. pmid:22071480
- View Article
- PubMed/NCBI
- Google Scholar
5. Westhovens R, Van der Elst K, Matthys A, Tran M, Gilloteau I. Sleep problems in patients with rheumatoid arthritis. J Rheumatol. 2014;41(1):31–40. pmid:24293569
- View Article
- PubMed/NCBI
- Google Scholar
6. Lyne L, Åkerstedt T, Alfredsson L, Lehtonen T, Saevarsdottir S, Klareskog L, et al. Sleep problems in rheumatoid arthritis over 12 years from diagnosis: results from the Swedish EIRA study. RMD Open. 2022;8(1):e001800. pmid:34987091
- View Article
- PubMed/NCBI
- Google Scholar
7. Goes ACJ, Reis LAB, Silva MBG, Kahlow BS, Skare TL. Rheumatoid arthritis and sleep quality. Rev Bras Reumatol Engl Ed. 2017;57(4):294–8. pmid:28743355
- View Article
- PubMed/NCBI
- Google Scholar
8. Sariyildiz MA, Batmaz I, Bozkurt M, Bez Y, Cetincakmak MG, Yazmalar L, et al. Sleep quality in rheumatoid arthritis: relationship between the disease severity, depression, functional status and the quality of life. J Clin Med Res. 2014;6(1):44–52. pmid:24400031
- View Article
- PubMed/NCBI
- Google Scholar
9. Garbarino S, Lanteri P, Bragazzi NL, Magnavita N, Scoditti E. Role of sleep deprivation in immune-related disease risk and outcomes. Commun Biol. 2021;4(1):1304. pmid:34795404
- View Article
- PubMed/NCBI
- Google Scholar
10. Dolsen EA, Crosswell AD, Prather AA. Links between stress, sleep, and inflammation: are there sex differences? Curr Psychiatry Rep. 2019;21(2):8. pmid:30729328
- View Article
- PubMed/NCBI
- Google Scholar
11. Asif N, Iqbal R, Nazir CF. Human immune system during sleep. Am J Clin Exp Immunol. 2017;6(6):92–6. pmid:29348984
- View Article
- PubMed/NCBI
- Google Scholar
12. Smiley A, King D, Bidulescu A. The association between sleep duration and metabolic syndrome: the NHANES 2013/2014. Nutrients. 2019;11(11):2582. pmid:31717770
- View Article
- PubMed/NCBI
- Google Scholar
13. Wang D, Ruan W, Peng Y, Li W. Sleep duration and the risk of osteoporosis among middle-aged and elderly adults: a dose-response meta-analysis. Osteoporos Int. 2018;29(8):1689–95. pmid:29574520
- View Article
- PubMed/NCBI
- Google Scholar
14. Ma Y, Liang L, Zheng F, Shi L, Zhong B, Xie W. Association between sleep duration and cognitive decline. JAMA Netw Open. 2020;3(9):e2013573. pmid:32955572
- View Article
- PubMed/NCBI
- Google Scholar
15. Yin J, Jin X, Shan Z, Li S, Huang H, Li P, et al. Relationship of sleep duration with all-cause mortality and cardiovascular events: a systematic review and dose-response meta-analysis of prospective cohort studies. J Am Heart Assoc. 2017;6(9):e005947. pmid:28889101
- View Article
- PubMed/NCBI
- Google Scholar
16. Chung W-S, Lin C-L. Sleep disorders associated with risk of rheumatoid arthritis. Sleep Breath. 2018;22(4):1083–91. pmid:29428977
- View Article
- PubMed/NCBI
- Google Scholar
17. Hsiao Y-H, Chen Y-T, Tseng C-M, Wu L-A, Lin W-C, Su VY-F, et al. Sleep disorders and increased risk of autoimmune diseases in individuals without sleep apnea. Sleep. 2015;38(4):581–6. pmid:25669189
- View Article
- PubMed/NCBI
- Google Scholar
18. Katz P. Causes and consequences of fatigue in rheumatoid arthritis. Curr Opin Rheumatol. 2017;29(3):269–76. pmid:28207494
- View Article
- PubMed/NCBI
- Google Scholar
19. Fan M, Sun D, Zhou T, Heianza Y, Lv J, Li L, et al. Sleep patterns, genetic susceptibility, and incident cardiovascular disease: a prospective study of 385 292 UK biobank participants. Eur Heart J. 2020;41(11):1182–9. pmid:31848595
- View Article
- PubMed/NCBI
- Google Scholar
20. Bulik-Sullivan BK, Loh P-R, Finucane HK, Ripke S, Yang J, Patterson N, et al.; Schizophrenia Working Group of the Psychiatric Genomics Consortium. LD Score regression distinguishes confounding from polygenicity in genome-wide association studies. Nat Genet. 2015;47(3):291–5. pmid:25642630
- View Article
- PubMed/NCBI
- Google Scholar
21. Bycroft C, Freeman C, Petkova D, Band G, Elliott LT, Sharp K, et al. The UK Biobank resource with deep phenotyping and genomic data. Nature. 2018;562(7726):203–9. pmid:30305743
- View Article
- PubMed/NCBI
- Google Scholar
22. Dashti HS, Jones SE, Wood AR, Lane JM, van Hees VT, Wang H, et al. Genome-wide association study identifies genetic loci for self-reported habitual sleep duration supported by accelerometer-derived estimates. Nat Commun. 2019;10(1):1100. pmid:30846698
- View Article
- PubMed/NCBI
- Google Scholar
23. Jansen PR, Watanabe K, Stringer S, Skene N, Bryois J, Hammerschlag AR, et al. Genome-wide analysis of insomnia in 1,331,010 individuals identifies new risk loci and functional pathways. Nat Genet. 2019;51(3):394–403. pmid:30804565
- View Article
- PubMed/NCBI
- Google Scholar
24. Jones SE, Lane JM, Wood AR, van Hees VT, Tyrrell J, Beaumont RN, et al. Genome-wide association analyses of chronotype in 697,828 individuals provides insights into circadian rhythms. Nat Commun. 2019;10(1):343. pmid:30696823
- View Article
- PubMed/NCBI
- Google Scholar
25. Dashti HS, Daghlas I, Lane JM, Huang Y, Udler MS, Wang H, et al. Genetic determinants of daytime napping and effects on cardiometabolic health. Nat Commun. 2021;12(1):900. pmid:33568662
- View Article
- PubMed/NCBI
- Google Scholar
26. Watanabe K, Jansen PR, Savage JE, Nandakumar P, Wang X, Hinds DA, et al.; 23andMe Research Team. Genome-wide meta-analysis of insomnia prioritizes genes associated with metabolic and psychiatric pathways. Nat Genet. 2022;54(8):1125–32. pmid:35835914
- View Article
- PubMed/NCBI
- Google Scholar
27. Campos AI, García-Marín LM, Byrne EM, Martin NG, Cuéllar-Partida G, Rentería ME. Insights into the aetiology of snoring from observational and genetic investigations in the UK Biobank. Nat Commun. 2020;11(1):817. pmid:32060260
- View Article
- PubMed/NCBI
- Google Scholar
28. Wang H, Lane JM, Jones SE, Dashti HS, Ollila HM, Wood AR, et al. Genome-wide association analysis of self-reported daytime sleepiness identifies 42 loci that suggest biological subtypes. Nat Commun. 2019;10(1):3503. pmid:31409809
- View Article
- PubMed/NCBI
- Google Scholar
29. Okada Y, Wu D, Trynka G, Raj T, Terao C, Ikari K, et al. Genetics of rheumatoid arthritis contributes to biology and drug discovery. Nature. 2014;506(7488):376–81. pmid:24390342
- View Article
- PubMed/NCBI
- Google Scholar
30. Yu X-H, Bo L, Cao R-R, Yang Y-Q, He P, Lei S-F, et al. Systematic evaluation of rheumatoid arthritis risk by integrating lifestyle factors and genetic risk scores. Front Immunol. 2022;13:901223. pmid:35874719
- View Article
- PubMed/NCBI
- Google Scholar
31. Yang Y-Q, Yu X-H, Bo L, Lei S-F, Deng F-Y. Genetic risk for osteoporosis and the benefit of adherence to healthy lifestyles. Int J Public Health. 2022;67:1605114. pmid:36176357
- View Article
- PubMed/NCBI
- Google Scholar
32. McMillan DC. The systemic inflammation-based Glasgow Prognostic Score: a decade of experience in patients with cancer. Cancer Treat Rev. 2013;39(5):534–40. pmid:22995477
- View Article
- PubMed/NCBI
- Google Scholar
33. Chambers J. Software for data analysis: programming with R. Springer Science & Business Media; 2008.
34. Wu W, Yang J, Gu Y, Chen X, Tan X. Dose-response relationship between sleep and rheumatoid arthritis. Am J Health Behav. 2020;44(1):40–53. pmid:31783931
- View Article
- PubMed/NCBI
- Google Scholar
35. Chen W-S, Chang Y-S, Chang C-C, Chang D-M, Chen Y-H, Tsai C-Y, et al. Management and risk reduction of rheumatoid arthritis in individuals with obstructive sleep apnea: a nationwide population-based study in Taiwan. Sleep. 2016;39(10):1883–90. pmid:27397567
- View Article
- PubMed/NCBI
- Google Scholar
36. Fan M, Sun D, Zhou T, Heianza Y, Lv J, Li L, et al. Sleep patterns, genetic susceptibility, and incident cardiovascular disease: a prospective study of 385 292 UK biobank participants. Eur Heart J. 2020;41(11):1182–9. pmid:31848595
- View Article
- PubMed/NCBI
- Google Scholar
37. Li X, Xue Q, Wang M, Zhou T, Ma H, Heianza Y, et al. Adherence to a healthy sleep pattern and incident heart failure: a prospective study of 408 802 UK Biobank participants. Circulation. 2021;143(1):97–9. pmid:33190528
- View Article
- PubMed/NCBI
- Google Scholar
38. Liu B, Qian Y, Lin H, Zhao S, Ying J, Chen W, et al. Sleep pattern, genetic risk, and the risk of incident rheumatoid arthritis: A cohort study. Sleep Health. 2024;10(6):635–42. pmid:39306487
- View Article
- PubMed/NCBI
- Google Scholar
39. Aldabal L, Bahammam AS. Metabolic, endocrine, and immune consequences of sleep deprivation. Open Respir Med J. 2011;5:31–43. pmid:21754974
- View Article
- PubMed/NCBI
- Google Scholar
40. Besedovsky L, Lange T, Haack M. The sleep-immune crosstalk in health and disease. Physiol Rev. 2019;99(3):1325–80. pmid:30920354
- View Article
- PubMed/NCBI
- Google Scholar
41. Lasselin J, Rehman J-U, Åkerstedt T, Lekander M, Axelsson J. Effect of long-term sleep restriction and subsequent recovery sleep on the diurnal rhythms of white blood cell subpopulations. Brain Behav Immun. 2015;47:93–9. pmid:25451611
- View Article
- PubMed/NCBI
- Google Scholar
42. Born J, Lange T, Hansen K, Mölle M, Fehm HL. Effects of sleep and circadian rhythm on human circulating immune cells. J Immunol. 1997;158(9):4454–64. pmid:9127011
- View Article
- PubMed/NCBI
- Google Scholar
43. Cope AP, Schulze-Koops H, Aringer M. The central role of T cells in rheumatoid arthritis. Clin Exp Rheumatol. 2007;25(5 Suppl 46):S4-11. pmid:17977483
- View Article
- PubMed/NCBI
- Google Scholar
44. Ananthakrishnan AN, Long MD, Martin CF, Sandler RS, Kappelman MD. Sleep disturbance and risk of active disease in patients with Crohn’s disease and ulcerative colitis. Clin Gastroenterol Hepatol. 2013;11(8):965–71. pmid:23376797
- View Article
- PubMed/NCBI
- Google Scholar
45. van Leeuwen WMA, Lehto M, Karisola P, Lindholm H, Luukkonen R, Sallinen M, et al. Sleep restriction increases the risk of developing cardiovascular diseases by augmenting proinflammatory responses through IL-17 and CRP. PLoS ONE. 2009;4(2):e4589. pmid:19240794
- View Article
- PubMed/NCBI
- Google Scholar
46. Suarez EC. Self-reported symptoms of sleep disturbance and inflammation, coagulation, insulin resistance and psychosocial distress: evidence for gender disparity. Brain Behav Immun. 2008;22(6):960–8. pmid:18328671
- View Article
- PubMed/NCBI
- Google Scholar
47. Scherer HU, Häupl T, Burmester GR. The etiology of rheumatoid arthritis. J Autoimmun. 2020;110:102400. pmid:31980337
- View Article
- PubMed/NCBI
- Google Scholar
48. Yehuda S, Sredni B, Carasso RL, Kenigsbuch-Sredni D. REM sleep deprivation in rats results in inflammation and interleukin-17 elevation. J Interferon Cytokine Res. 2009;29(7):393–8. pmid:19450150
- View Article
- PubMed/NCBI
- Google Scholar
49. Adamopoulos IE, Chao C-C, Geissler R, Laface D, Blumenschein W, Iwakura Y, et al. Interleukin-17A upregulates receptor activator of NF-kappaB on osteoclast precursors. Arthritis Res Ther. 2010;12(1):R29. pmid:20167120
- View Article
- PubMed/NCBI
- Google Scholar

[ref1] 1. Smolen JS, Aletaha D, McInnes IB. Rheumatoid arthritis. Lancet. 2016;388(10055):2023–38. pmid:27156434
View Article
PubMed/NCBI
Google Scholar

[2] View Article

[3] PubMed/NCBI

[4] Google Scholar

[ref2] 2. Clynes MA, Jameson K, Prieto-Alhambra D, Harvey NC, Cooper C, Dennison EM. Impact of rheumatoid arthritis and its management on falls, fracture and bone mineral density in UK Biobank. Front Endocrinol (Lausanne). 2019;10:817. pmid:31849841
View Article
PubMed/NCBI
Google Scholar

[6] View Article

[7] PubMed/NCBI

[8] Google Scholar

[ref3] 3. Glanville KP, Coleman JRI, O’Reilly PF, Galloway J, Lewis CM. Investigating pleiotropy between depression and autoimmune diseases using the UK Biobank. Biol Psychiatry Glob Open Sci. 2021;1(1):48–58. pmid:34278373
View Article
PubMed/NCBI
Google Scholar

[10] View Article

[11] PubMed/NCBI

[12] Google Scholar

[ref4] 4. Besedovsky L, Lange T, Born J. Sleep and immune function. Pflugers Arch. 2012;463(1):121–37. pmid:22071480
View Article
PubMed/NCBI
Google Scholar

[14] View Article

[15] PubMed/NCBI

[16] Google Scholar

[ref5] 5. Westhovens R, Van der Elst K, Matthys A, Tran M, Gilloteau I. Sleep problems in patients with rheumatoid arthritis. J Rheumatol. 2014;41(1):31–40. pmid:24293569
View Article
PubMed/NCBI
Google Scholar

[18] View Article

[19] PubMed/NCBI

[20] Google Scholar

[ref6] 6. Lyne L, Åkerstedt T, Alfredsson L, Lehtonen T, Saevarsdottir S, Klareskog L, et al. Sleep problems in rheumatoid arthritis over 12 years from diagnosis: results from the Swedish EIRA study. RMD Open. 2022;8(1):e001800. pmid:34987091
View Article
PubMed/NCBI
Google Scholar

[22] View Article

[23] PubMed/NCBI

[24] Google Scholar

[ref7] 7. Goes ACJ, Reis LAB, Silva MBG, Kahlow BS, Skare TL. Rheumatoid arthritis and sleep quality. Rev Bras Reumatol Engl Ed. 2017;57(4):294–8. pmid:28743355
View Article
PubMed/NCBI
Google Scholar

[26] View Article

[27] PubMed/NCBI

[28] Google Scholar

[ref8] 8. Sariyildiz MA, Batmaz I, Bozkurt M, Bez Y, Cetincakmak MG, Yazmalar L, et al. Sleep quality in rheumatoid arthritis: relationship between the disease severity, depression, functional status and the quality of life. J Clin Med Res. 2014;6(1):44–52. pmid:24400031
View Article
PubMed/NCBI
Google Scholar

[30] View Article

[31] PubMed/NCBI

[32] Google Scholar

[ref9] 9. Garbarino S, Lanteri P, Bragazzi NL, Magnavita N, Scoditti E. Role of sleep deprivation in immune-related disease risk and outcomes. Commun Biol. 2021;4(1):1304. pmid:34795404
View Article
PubMed/NCBI
Google Scholar

[34] View Article

[35] PubMed/NCBI

[36] Google Scholar

[ref10] 10. Dolsen EA, Crosswell AD, Prather AA. Links between stress, sleep, and inflammation: are there sex differences? Curr Psychiatry Rep. 2019;21(2):8. pmid:30729328
View Article
PubMed/NCBI
Google Scholar

[38] View Article

[39] PubMed/NCBI

[40] Google Scholar

[ref11] 11. Asif N, Iqbal R, Nazir CF. Human immune system during sleep. Am J Clin Exp Immunol. 2017;6(6):92–6. pmid:29348984
View Article
PubMed/NCBI
Google Scholar

[42] View Article

[43] PubMed/NCBI

[44] Google Scholar

[ref12] 12. Smiley A, King D, Bidulescu A. The association between sleep duration and metabolic syndrome: the NHANES 2013/2014. Nutrients. 2019;11(11):2582. pmid:31717770
View Article
PubMed/NCBI
Google Scholar

[46] View Article

[47] PubMed/NCBI

[48] Google Scholar

[ref13] 13. Wang D, Ruan W, Peng Y, Li W. Sleep duration and the risk of osteoporosis among middle-aged and elderly adults: a dose-response meta-analysis. Osteoporos Int. 2018;29(8):1689–95. pmid:29574520
View Article
PubMed/NCBI
Google Scholar

[50] View Article

[51] PubMed/NCBI

[52] Google Scholar

[ref14] 14. Ma Y, Liang L, Zheng F, Shi L, Zhong B, Xie W. Association between sleep duration and cognitive decline. JAMA Netw Open. 2020;3(9):e2013573. pmid:32955572
View Article
PubMed/NCBI
Google Scholar

[54] View Article

[55] PubMed/NCBI

[56] Google Scholar

[ref15] 15. Yin J, Jin X, Shan Z, Li S, Huang H, Li P, et al. Relationship of sleep duration with all-cause mortality and cardiovascular events: a systematic review and dose-response meta-analysis of prospective cohort studies. J Am Heart Assoc. 2017;6(9):e005947. pmid:28889101
View Article
PubMed/NCBI
Google Scholar

[58] View Article

[59] PubMed/NCBI

[60] Google Scholar

[ref16] 16. Chung W-S, Lin C-L. Sleep disorders associated with risk of rheumatoid arthritis. Sleep Breath. 2018;22(4):1083–91. pmid:29428977
View Article
PubMed/NCBI
Google Scholar

[62] View Article

[63] PubMed/NCBI

[64] Google Scholar

[ref17] 17. Hsiao Y-H, Chen Y-T, Tseng C-M, Wu L-A, Lin W-C, Su VY-F, et al. Sleep disorders and increased risk of autoimmune diseases in individuals without sleep apnea. Sleep. 2015;38(4):581–6. pmid:25669189
View Article
PubMed/NCBI
Google Scholar

[66] View Article

[67] PubMed/NCBI

[68] Google Scholar

[ref18] 18. Katz P. Causes and consequences of fatigue in rheumatoid arthritis. Curr Opin Rheumatol. 2017;29(3):269–76. pmid:28207494
View Article
PubMed/NCBI
Google Scholar

[70] View Article

[71] PubMed/NCBI

[72] Google Scholar

[ref19] 19. Fan M, Sun D, Zhou T, Heianza Y, Lv J, Li L, et al. Sleep patterns, genetic susceptibility, and incident cardiovascular disease: a prospective study of 385 292 UK biobank participants. Eur Heart J. 2020;41(11):1182–9. pmid:31848595
View Article
PubMed/NCBI
Google Scholar

[74] View Article

[75] PubMed/NCBI

[76] Google Scholar

[ref20] 20. Bulik-Sullivan BK, Loh P-R, Finucane HK, Ripke S, Yang J, Patterson N, et al.; Schizophrenia Working Group of the Psychiatric Genomics Consortium. LD Score regression distinguishes confounding from polygenicity in genome-wide association studies. Nat Genet. 2015;47(3):291–5. pmid:25642630
View Article
PubMed/NCBI
Google Scholar

[78] View Article

[79] PubMed/NCBI

[80] Google Scholar

[ref21] 21. Bycroft C, Freeman C, Petkova D, Band G, Elliott LT, Sharp K, et al. The UK Biobank resource with deep phenotyping and genomic data. Nature. 2018;562(7726):203–9. pmid:30305743
View Article
PubMed/NCBI
Google Scholar

[82] View Article

[83] PubMed/NCBI

[84] Google Scholar

[ref22] 22. Dashti HS, Jones SE, Wood AR, Lane JM, van Hees VT, Wang H, et al. Genome-wide association study identifies genetic loci for self-reported habitual sleep duration supported by accelerometer-derived estimates. Nat Commun. 2019;10(1):1100. pmid:30846698
View Article
PubMed/NCBI
Google Scholar

[86] View Article

[87] PubMed/NCBI

[88] Google Scholar

[ref23] 23. Jansen PR, Watanabe K, Stringer S, Skene N, Bryois J, Hammerschlag AR, et al. Genome-wide analysis of insomnia in 1,331,010 individuals identifies new risk loci and functional pathways. Nat Genet. 2019;51(3):394–403. pmid:30804565
View Article
PubMed/NCBI
Google Scholar

[90] View Article

[91] PubMed/NCBI

[92] Google Scholar

[ref24] 24. Jones SE, Lane JM, Wood AR, van Hees VT, Tyrrell J, Beaumont RN, et al. Genome-wide association analyses of chronotype in 697,828 individuals provides insights into circadian rhythms. Nat Commun. 2019;10(1):343. pmid:30696823
View Article
PubMed/NCBI
Google Scholar

[94] View Article

[95] PubMed/NCBI

[96] Google Scholar

[ref25] 25. Dashti HS, Daghlas I, Lane JM, Huang Y, Udler MS, Wang H, et al. Genetic determinants of daytime napping and effects on cardiometabolic health. Nat Commun. 2021;12(1):900. pmid:33568662
View Article
PubMed/NCBI
Google Scholar

[98] View Article

[99] PubMed/NCBI

[100] Google Scholar

[ref26] 26. Watanabe K, Jansen PR, Savage JE, Nandakumar P, Wang X, Hinds DA, et al.; 23andMe Research Team. Genome-wide meta-analysis of insomnia prioritizes genes associated with metabolic and psychiatric pathways. Nat Genet. 2022;54(8):1125–32. pmid:35835914
View Article
PubMed/NCBI
Google Scholar

[102] View Article

[103] PubMed/NCBI

[104] Google Scholar

[ref27] 27. Campos AI, García-Marín LM, Byrne EM, Martin NG, Cuéllar-Partida G, Rentería ME. Insights into the aetiology of snoring from observational and genetic investigations in the UK Biobank. Nat Commun. 2020;11(1):817. pmid:32060260
View Article
PubMed/NCBI
Google Scholar

[106] View Article

[107] PubMed/NCBI

[108] Google Scholar

[ref28] 28. Wang H, Lane JM, Jones SE, Dashti HS, Ollila HM, Wood AR, et al. Genome-wide association analysis of self-reported daytime sleepiness identifies 42 loci that suggest biological subtypes. Nat Commun. 2019;10(1):3503. pmid:31409809
View Article
PubMed/NCBI
Google Scholar

[110] View Article

[111] PubMed/NCBI

[112] Google Scholar

[ref29] 29. Okada Y, Wu D, Trynka G, Raj T, Terao C, Ikari K, et al. Genetics of rheumatoid arthritis contributes to biology and drug discovery. Nature. 2014;506(7488):376–81. pmid:24390342
View Article
PubMed/NCBI
Google Scholar

[114] View Article

[115] PubMed/NCBI

[116] Google Scholar

[ref30] 30. Yu X-H, Bo L, Cao R-R, Yang Y-Q, He P, Lei S-F, et al. Systematic evaluation of rheumatoid arthritis risk by integrating lifestyle factors and genetic risk scores. Front Immunol. 2022;13:901223. pmid:35874719
View Article
PubMed/NCBI
Google Scholar

[118] View Article

[119] PubMed/NCBI

[120] Google Scholar

[ref31] 31. Yang Y-Q, Yu X-H, Bo L, Lei S-F, Deng F-Y. Genetic risk for osteoporosis and the benefit of adherence to healthy lifestyles. Int J Public Health. 2022;67:1605114. pmid:36176357
View Article
PubMed/NCBI
Google Scholar

[122] View Article

[123] PubMed/NCBI

[124] Google Scholar

[ref32] 32. McMillan DC. The systemic inflammation-based Glasgow Prognostic Score: a decade of experience in patients with cancer. Cancer Treat Rev. 2013;39(5):534–40. pmid:22995477
View Article
PubMed/NCBI
Google Scholar

[126] View Article

[127] PubMed/NCBI

[128] Google Scholar

[ref33] 33. Chambers J. Software for data analysis: programming with R. Springer Science & Business Media; 2008.

[ref34] 34. Wu W, Yang J, Gu Y, Chen X, Tan X. Dose-response relationship between sleep and rheumatoid arthritis. Am J Health Behav. 2020;44(1):40–53. pmid:31783931
View Article
PubMed/NCBI
Google Scholar

[131] View Article

[132] PubMed/NCBI

[133] Google Scholar

[ref35] 35. Chen W-S, Chang Y-S, Chang C-C, Chang D-M, Chen Y-H, Tsai C-Y, et al. Management and risk reduction of rheumatoid arthritis in individuals with obstructive sleep apnea: a nationwide population-based study in Taiwan. Sleep. 2016;39(10):1883–90. pmid:27397567
View Article
PubMed/NCBI
Google Scholar

[135] View Article

[136] PubMed/NCBI

[137] Google Scholar

[ref36] 36. Fan M, Sun D, Zhou T, Heianza Y, Lv J, Li L, et al. Sleep patterns, genetic susceptibility, and incident cardiovascular disease: a prospective study of 385 292 UK biobank participants. Eur Heart J. 2020;41(11):1182–9. pmid:31848595
View Article
PubMed/NCBI
Google Scholar

[139] View Article

[140] PubMed/NCBI

[141] Google Scholar

[ref37] 37. Li X, Xue Q, Wang M, Zhou T, Ma H, Heianza Y, et al. Adherence to a healthy sleep pattern and incident heart failure: a prospective study of 408 802 UK Biobank participants. Circulation. 2021;143(1):97–9. pmid:33190528
View Article
PubMed/NCBI
Google Scholar

[143] View Article

[144] PubMed/NCBI

[145] Google Scholar

[ref38] 38. Liu B, Qian Y, Lin H, Zhao S, Ying J, Chen W, et al. Sleep pattern, genetic risk, and the risk of incident rheumatoid arthritis: A cohort study. Sleep Health. 2024;10(6):635–42. pmid:39306487
View Article
PubMed/NCBI
Google Scholar

[147] View Article

[148] PubMed/NCBI

[149] Google Scholar

[ref39] 39. Aldabal L, Bahammam AS. Metabolic, endocrine, and immune consequences of sleep deprivation. Open Respir Med J. 2011;5:31–43. pmid:21754974
View Article
PubMed/NCBI
Google Scholar

[151] View Article

[152] PubMed/NCBI

[153] Google Scholar

[ref40] 40. Besedovsky L, Lange T, Haack M. The sleep-immune crosstalk in health and disease. Physiol Rev. 2019;99(3):1325–80. pmid:30920354
View Article
PubMed/NCBI
Google Scholar

[155] View Article

[156] PubMed/NCBI

[157] Google Scholar

[ref41] 41. Lasselin J, Rehman J-U, Åkerstedt T, Lekander M, Axelsson J. Effect of long-term sleep restriction and subsequent recovery sleep on the diurnal rhythms of white blood cell subpopulations. Brain Behav Immun. 2015;47:93–9. pmid:25451611
View Article
PubMed/NCBI
Google Scholar

[159] View Article

[160] PubMed/NCBI

[161] Google Scholar

[ref42] 42. Born J, Lange T, Hansen K, Mölle M, Fehm HL. Effects of sleep and circadian rhythm on human circulating immune cells. J Immunol. 1997;158(9):4454–64. pmid:9127011
View Article
PubMed/NCBI
Google Scholar

[163] View Article

[164] PubMed/NCBI

[165] Google Scholar

[ref43] 43. Cope AP, Schulze-Koops H, Aringer M. The central role of T cells in rheumatoid arthritis. Clin Exp Rheumatol. 2007;25(5 Suppl 46):S4-11. pmid:17977483
View Article
PubMed/NCBI
Google Scholar

[167] View Article

[168] PubMed/NCBI

[169] Google Scholar

[ref44] 44. Ananthakrishnan AN, Long MD, Martin CF, Sandler RS, Kappelman MD. Sleep disturbance and risk of active disease in patients with Crohn’s disease and ulcerative colitis. Clin Gastroenterol Hepatol. 2013;11(8):965–71. pmid:23376797
View Article
PubMed/NCBI
Google Scholar

[171] View Article

[172] PubMed/NCBI

[173] Google Scholar

[ref45] 45. van Leeuwen WMA, Lehto M, Karisola P, Lindholm H, Luukkonen R, Sallinen M, et al. Sleep restriction increases the risk of developing cardiovascular diseases by augmenting proinflammatory responses through IL-17 and CRP. PLoS ONE. 2009;4(2):e4589. pmid:19240794
View Article
PubMed/NCBI
Google Scholar

[175] View Article

[176] PubMed/NCBI

[177] Google Scholar

[ref46] 46. Suarez EC. Self-reported symptoms of sleep disturbance and inflammation, coagulation, insulin resistance and psychosocial distress: evidence for gender disparity. Brain Behav Immun. 2008;22(6):960–8. pmid:18328671
View Article
PubMed/NCBI
Google Scholar

[179] View Article

[180] PubMed/NCBI

[181] Google Scholar

[ref47] 47. Scherer HU, Häupl T, Burmester GR. The etiology of rheumatoid arthritis. J Autoimmun. 2020;110:102400. pmid:31980337
View Article
PubMed/NCBI
Google Scholar

[183] View Article

[184] PubMed/NCBI

[185] Google Scholar

[ref48] 48. Yehuda S, Sredni B, Carasso RL, Kenigsbuch-Sredni D. REM sleep deprivation in rats results in inflammation and interleukin-17 elevation. J Interferon Cytokine Res. 2009;29(7):393–8. pmid:19450150
View Article
PubMed/NCBI
Google Scholar

[187] View Article

[188] PubMed/NCBI

[189] Google Scholar

[ref49] 49. Adamopoulos IE, Chao C-C, Geissler R, Laface D, Blumenschein W, Iwakura Y, et al. Interleukin-17A upregulates receptor activator of NF-kappaB on osteoclast precursors. Arthritis Res Ther. 2010;12(1):R29. pmid:20167120
View Article
PubMed/NCBI
Google Scholar

[191] View Article

[192] PubMed/NCBI

[193] Google Scholar

Figures

Abstract

Objective

Methods

Results

Conclusions

Introduction

Materials and methods

Study design

UK Biobank data sets

Ethics approval statement.

Definition of RA cases.

Definition of sleeping condition.

Summary-level data and construction of genetic risk score (GRS)

Statistical analysis

Results

Baseline characteristics

Association between sleep behaviors and RA risk

Poor sleep conditions and RA risk.

Daytime napping and difficulty of getting up and RA risk.

Sleep behaviors and seropositive RA risk.

Association between poor sleep score (PSS) and RA risk

The combined effect of sleeping traits and genetic factors on RA risk

Effect of RA onset on sleep condition

Genetic correlations between RA and sleep phenotypes

Discussion

Strength and limitations

Supporting information

S1 Fig. Restricted cubic spline models for the relationship between PSS and RA risk in men (A) and women (B).

S2 Fig. The contribution of mGPS for the association between PSS and RA risk.

S3 Fig. Restricted cubic spline models for the relationship between PSS and seropositive RA risk.

S4 Fig. RA risks in the subgroups stratified by genetic risk and PSSCV (versus participants with lowest PSS in the lowest genetic risk group) in the UKB cohort.

S5 Fig. Genetic correlation of sleep phenotypes with different types of RA (P < 0.05).

S1 Table. Description and definition of sleeping behaviors in UK Biobank data.

S2 Table. Baseline characteristics of the study subjects stratified by different PSS scores in the UK Biobank data.

S3 Table. The adjusted hazard ratio for sleep behaviors on the risk of seropositive RA in the UK Biobank data.

S4 Table. The adjusted hazard ratio for RA and seropositive RA risk stratified by three PSS categories in the UK Biobank data

S5 Table. The adjusted hazard ratio for RA and seropositive RA risk stratified by five PSS categories in the UK Biobank data.

S6 Table. The adjusted hazard ratio for the effect of GRSs in different cutoffs on RA risk in both training set and test set.

S7 Table. RA risks in the subgroups stratified by genetic risk and different sleep behaviors (versus participants with healthy sleep behaviors in the lowest genetic risk group) in the UK Biobank.

S8 Table. The effect of RA onset on sleep condition in the UK Biobank data.

Acknowledgments

References