Abstract
Objective
This study investigated whether sleep disturbance can predict the extent to which pain interferes with daily functioning in patients with early rheumatoid arthritis (RA).
Methods
Data were from adults with early RA (joint symptoms ≤12 months) enrolled in the Canadian Early Arthritis Cohort between 2016 and 2023. Participants underwent standardized clinical assessments and completed Patient‐Reported Outcomes Measurement Information System measures at 0, 6, 12, 18, and 24 months to assess sleep disturbance (primary predictor) and pain interference (primary outcome). Linear mixed‐effects models were used to estimate crude and adjusted (age, sex, body mass index, education, income, smoking status, comorbidities, disease activity, treatment, and depression) effects of sleep disturbance on pain interference over the 24‐month study period. The analysis was lagged so that repeat measures of sleep disturbance at 0, 6, 12, and 18 months were evaluated as predictors of pain interference 6 months later at 6, 12, 18, and 24 months’ follow‐up.
Results
The analysis included 502 patients with early RA. At baseline, the sample was 68% female and 81% White; the mean age was 56 (SD 14) years, and the mean disease duration was 5.4 (SD 2.9) months. The unadjusted and adjusted linear mixed‐effects models revealed a significant association between sleep disturbance and subsequent pain interference scores, indicating that worse sleep six months prior was associated with greater pain interference at the following six‐month evaluation.
Conclusion
These findings underscore the importance of addressing sleep disturbances as part of pain management strategies soon after RA diagnosis. Identifying and targeting problematic sleep disturbances early on may help improve long‐term pain outcomes.
INTRODUCTION
Pain and sleep disturbances have many debilitating effects on physical and mental functioning. In patients with rheumatoid arthritis (RA), pain is a common symptom and primary reason for seeking care. 1 In addition to pain, more than half of individuals with RA suffer from sleep disturbances. 1 , 2 , 3 , 4 , 5 These disturbances encompass challenges such as difficulty initiating sleep and recurrent nocturnal awakenings. The prevailing notion is that pain causes sleep disturbances, 4 , 6 , 7 , 8 , 9 but the interplay between sleep and pain is complex.
SIGNIFICANCE & INNOVATIONS.
The focus on early rheumatoid arthritis (RA) is significant because the first few years after symptom onset may represent a critical window of opportunity to alter the long‐term consequences of disease (eg, chronic pain and disability).
The choice of pain interference as the primary outcome is significant because previous studies demonstrated that among patients with RA, the effect of pain on daily function is likely higher than would be expected based on assessments of pain intensity alone.
Study results support addressing problematic sleep patterns following RA symptom onset to enhance long‐term pain management in patients with RA.
A growing consensus posits a reciprocal relationship between sleep disturbances and pain in the general population, and a few studies suggest that sleep disturbances can lead to heightened pain severity in established RA cohorts. 10 , 11 , 12 One experimental study demonstrated that one night of restricted sleep led to increases in pain severity and the number of painful joints the next day. 12 Our research group also reported cross‐sectional associations between sleep disturbances and pain sensitivity, 10 as well as longitudinal associations between sleep disturbances and pain intensity. 12 To date, no studies have investigated the longitudinal association between sleep disturbances and subsequent long‐term consequences of pain (eg, pain interference) in patients with early RA.
Pain interference is an important construct that describes the consequences of pain and how it interferes with important aspects of life (eg, social, mental, and physical functioning). Patients have identified it as pivotal to their quality of life. 13 Pain interference is distinct from pain intensity, which describes the magnitude of perceived pain experienced. Among patients with RA, median pain interference scores on a standardized common metric (Patient‐Reported Outcomes Measurement Information System [PROMIS]) were 10 points higher than median pain intensity scores. 14 These results indicate that pain has a greater impact on daily function than would be expected from assessments of pain intensity scores alone. Because pain interference incorporates both pain and function, it may be an important outcome to assess longitudinally. It is possible that sleep disturbances may drive changes in pain interference; however, this directional impact is yet to be explored.
The aim of this study was to estimate to what extent self‐reported sleep disturbance may be associated with pain interference six months later in patients with early RA. We hypothesized that greater self‐reported sleep disturbance would be associated with greater subsequent pain interference. Understanding the relationship between sleep disturbances early in the disease process and long‐term consequences of pain may provide us with better opportunities for prevention and treatment (eg, cognitive behavioral therapy, light therapy).
PATIENTS AND METHODS
This study analyzed data collected at 0 (baseline), 6, 12, 18, and 24 months from adults with early RA enrolled in the Canadian Early Arthritis Cohort (CATCH) between January 2016 and March 2023. Briefly, CATCH is a multicenter observational prospective cohort study of adults diagnosed with early RA (joint symptoms ≤12 months) by a rheumatologist from academic and community clinics across Canada. Participants are eligible for enrollment if they are >18 years old, have joint symptoms for ≥6 weeks and ≤12 months, and have two or more swollen joints or one swollen metacarpophalangeal or proximal interphalangeal joint, with one of the following features: rheumatoid factor (RF) ≥20 IU, positive test for anti–citrullinated protein antibodies (ACPAs), morning stiffness ≥45 minutes, response to nonsteroidal anti‐inflammatory drug treatment, or a painful metatarsophalangeal joint squeeze test. Participants included in the present analysis had to have PROMIS sleep disturbance and pain interference scores at baseline, and they also had to contribute at least one pair of sleep disturbance and pain interference measures six months apart (ie, baseline sleep disturbance and six‐month pain interference; Figure 1). Participants were excluded or withdrawn (if identified after inclusion) for the following diagnoses: psoriatic arthritis or infectious, crystal‐induced, or connective tissue diseases. Further details of the CATCH study and protocols have been reported previously. 15 Approval from each participating site's institutional review board was obtained. All participants enrolled in the study provided written informed consent.
Figure 1.
Flowchart of participants fulfilling eligibility criteria with PROMIS data available at each time point. CATCH, Canadian Early Arthritis Cohort; PROMIS, Patient‐Reported Outcomes Measurement Information System.
Patient‐reported outcomes measures
Participants completed the PROMIS‐29 v2.0 profile measure to assess sleep disturbance (primary exposure) and pain interference (primary outcome) over the past seven days. 16 The sleep disturbance domain includes questions about perceptions of sleep quality, depth, and restoration. The pain interference domain measures the extent to which pain interferes with physical, mental, and social functioning. All PROMIS raw scores are converted to a mean T‐score of 50 with a SD of 10 based on the general US population. 17 Higher scores for sleep disturbance and pain interference represent more of the concept being measured (eg, greater disturbance and interference).
Demographic and clinical characteristics
Participants completed baseline study case report forms, which included self‐reported age, sex, income, smoking status, height and weight (used to calculate body mass index [BMI]), education, and self‐reported physician‐diagnosed health conditions (used to calculate the Rheumatic Disease Comorbidity Index [RDCI]). 18 Tender and swollen 28‐joint counts and medication use were ascertained by the rheumatology health care team. Standard laboratory tests were performed to assess ACPA, RF, and C‐reactive protein (CRP) levels and/or the erythrocyte sedimentation rate.
Statistical analysis
Descriptive statistics were used to summarize baseline sample characteristics. To examine the temporal relationship between sleep disturbance and pain interference in our longitudinal analysis, we used linear mixed‐effects models with random intercepts for participants and a compound symmetry covariance structure. Specifically, we estimated the effects of lagged measures of sleep disturbance at 0, 6, 12, and 18 months on pain interference 6 months later at 6, 12, 18, and 24 months. This modeling approach was chosen to account for the hierarchical structure of the data, repeated measurements, and covariance structure among participants. Model fit was assessed using diagnostic metrics, and assumptions of normality and homoscedasticity of residuals were checked visually using diagnostic plots. Mixed‐effects models leverage all available data, averaging across time points without requiring complete data for each participant. As a result, no participant was dropped from the analysis, and only observed data from available study visits were included in the model.
Multivariable linear mixed models were adjusted for baseline measures of age, sex, BMI, education, income, smoking status, RDCI, and time‐varying (updated) measures of swollen joint count, CRP level, steroid use, and disease‐modifying antirheumatic drug (DMARD) treatment (methotrexate and advanced therapy). We adjusted for both CRP level and swollen joint count as opposed to a composite measure of disease activity because composite measures include tender joint count and patient global assessment, which can be heavily influenced by pain. These covariates were treated as fixed effects in the models. The strength of associations was described using regression coefficients (β) with 95% confidence intervals (CIs).
A supplemental model of the adjusted analysis was performed with the PROMIS sleep disturbance score treated as a categorical term (none < 55, mild = 55–59, moderate = 60–69, or severe disturbance ≥ 70). To address the possible influence of depression, we conducted a sensitivity analysis to examine whether the association between sleep disturbances and subsequent pain interference remained robust after accounting for symptoms of depression as a lagged covariate. Symptoms of depression over the past seven days were assessed as part of the depression domain included in the PROMIS‐29. To account for potential temporal confounding, we also performed a sensitivity analysis adjusting for time‐varying concurrent baseline pain interference, defined as pain interference assessed at the same time as the time‐varying exposure (sleep disturbance). All data analyses were performed using SAS (version 9.4; SAS Institute, Inc).
RESULTS
Characteristics of the study sample
The analysis included 502 participants who contributed a total of 1,153 study visits/time points to the unadjusted modeling and 844 study visits/time points to the adjusted modeling (Table 1). At baseline, the mean ± SD age was 56 ± 14 years, the mean ± SD disease duration was 5.4 ± 2.9 months, and the mean ± SD Clinical Disease Activity Index score was 25.8 ± 13.7; 68% were female, 81% identified as White, 73% were seropositive (RF/ACPA), and 76% were treated with methotrexate. The mean ± SD T‐score for PROMIS pain interference was 60.4 ± 8.6, and the mean ± SD T‐score for PROMIS sleep disturbance was 53.5 ± 8.8 at baseline. At baseline, 80% of the sample had T‐scores ≥55 (mild to severe) for pain interference. Forty‐four percent had T‐scores ≥55 (mild to severe) for sleep disturbance.
Table 1.
Baseline demographic and clinical characteristics of early RA sample (N = 502)*
| Characteristic | Value |
|---|---|
| Demographic | |
| Age, mean (SD), y | 56 (14) |
| Female, % | 68 |
| White, % | 81 |
| BMI ever ≥30, % a | 32 |
| Postsecondary education, % | 61 |
| Income ≤$50,000, % a | 37 |
| Current smoker, % | 15 |
| RDCI, mean (SD) | 1.4 (1.4) |
| RA disease characteristics | |
| Disease duration, mean (SD), mo | 5.4 (2.9) |
| Meet 1987 ARA classification criteria b for RA or 2010 ACR/EULAR RA classification criteria, c % | 77 |
| Clinical Disease Activity Index, mean (SD) | 25.8 (13.7) |
| Seropositivity (RF/ACPA), % a | 73 |
| CRP, median (IQR), mg/L | 6.9 (2.9–18.5) |
| TJC‐28, median (IQR) | 7 (3–12) |
| SJC‐28, median (IQR) | 6 (3–10) |
| Patient global assessment score, mean (SD) | 4.8 (2.8) |
| Assessor global assessment score, mean (SD) | 5.2 (2.5) |
| Treatment, frequency, n (%) | |
| Oral steroids | 156 (31) |
| MTX | 384 (76) |
| Non‐MTX DMARDs | 281 (56) |
| Advanced therapy | 2 (0) |
| TNFi | 2 (0) |
| JAKi | 0 (0) |
| Other MOA (all other biologics/biosimilars) | 0 (0) |
| PROMIS T‐score, mean (SD) | |
| Sleep disturbance | 53.5 (8.8) |
| Pain interference | 60.4 (8.6) |
ACPA, anti–citrullinated protein antibody; ACR, American College of Rheumatology; ARA, American Rheumatism Association; BMI, body mass index; CRP, C‐reactive protein; DMARD, disease‐modifying antirheumatic drug; EULAR, European League of Rheumatologists; IQR, interquartile range; JAKi, JAK inhibitors; MOA, mechanism of action; MTX, methotrexate; PROMIS, Patient‐Reported Outcomes Measurement Information System; RA, rheumatoid arthritis; RDCI, Rheumatic Disease Comorbidity Index; RF, rheumatoid factor; SJC, swollen joint count; TJC, tender joint count; TNFi, tumor necrosis factor inhibitors.
Percentage of nonmissing.
Arnett et al. 21
Aletaha et al. 22
Unadjusted effects of sleep disturbance on pain interference
Participants who reported higher sleep disturbance subsequently reported greater pain interference at the following six‐month evaluation (β 0.76, 95% CI 0.49–1.02). Specifically, for every 5‐unit increase in sleep disturbance T‐score, there was a corresponding 0.76‐unit increase in pain disturbance T‐score six months later.
Adjusted effects of sleep disturbance on pain interference
Higher sleep disturbance was associated with greater pain interference at the subsequent six‐month evaluation, even after adjusting for age, sex, BMI, education, income, smoking status, RDCI, swollen joint count, CRP level, steroid use, and DMARD therapy (adjusted β 0.76, 95% CI 0.44–1.09; Table 2). Specifically, for every 5‐unit increase in sleep disturbance T‐score, there was a 0.76‐unit increase in pain disturbance T‐score six months later. In the supplemental model using categorical sleep disturbance, the results remained consistent, indicating that higher levels of sleep disturbance (moderate to severe) were associated with greater pain interference (Supplementary Table 1). In the sensitivity analysis adjusting the multivariable model by time‐varying symptoms of depression, the direction and significance of the effect of sleep disturbance on subsequent pain interference remained unchanged (Supplementary Table 2). Similarly, in the sensitivity analysis adjusting for time‐varying concurrent baseline pain interference, results remained consistent in direction and statistically significant (Supplementary Table 3).
Table 2.
Linear mixed‐effects regression models estimating association of PROMIS sleep disturbance with pain interference over 2 y of follow‐up in patients with early RA (N = 502)*
| Variables | Unadjusted model a | Adjusted multivariable model b , c | ||
|---|---|---|---|---|
| Regression coefficient | 95% CI | Regression coefficient | 95% CI | |
| Intercept | 45.32 | 42.20 to 48.43 | 39.26 | 34.83 to 43.69 |
| Time (mo) | −0.03 | −0.09 to 0.02 | −0.004 | −0.09 to 0.08 |
| Baseline time invariant variables | ||||
| Age (y) d | – | – | −0.01 | −0.07 to 0.05 |
| Female sex | – | – | 3.27 | 1.64 to 4.90 |
| Obese BMI (≥30) | – | – | 2.00 | 0.11 to 3.89 |
| Postsecondary education | – | – | −0.45 | −2.11 to 1.20 |
| Income >$50,000 | – | – | 0.74 | −1.14 to 2.61 |
| Smoking, current vs past or never | – | – | 3.25 | 1.13 to 5.38 |
| Comorbidity score (0–9) | – | – | 1.24 | 0.69 to 1.78 |
| Time‐varying variables (lagged by 6 mo) | ||||
| Sleep disturbance T score e | 0.76 | 0.49 to 1.02 | 0.76 | 0.44 to 1.09 |
| SJC‐28 | – | – | 0.00 | −0.12 to 0.12 |
| CRP (mg/L) | – | – | 0.04 | 0.00 to 0.08 |
| RA treatment | ||||
| Oral steroids | – | – | 1.25 | −0.28 to 2.77 |
| MTX | – | – | −0.71 | −2.15 to 0.72 |
| Advanced therapy | – | – | −0.73 | −2.85 to 1.38 |
Bold values indicate significant associations. BMI, body mass index; CI, confidence interval; CRP, C‐reactive protein; MTX, methotrexate; PROMIS, Patient‐Reported Outcomes Measurement Information System; RA, rheumatoid arthritis; SJC, swollen joint count.
1,153 visits included in unadjusted model.
Model adjusted for age (centered at the mean), sex, BMI, education, income, current smoking status, comorbidity index, as well as SJC‐28, CRP, MTX use, oral steroids use, and advanced therapy use, which were lagged from the previous visit.
844 visits included in adjusted model.
Per 10‐y increase.
Per 5‐unit increase in PROMIS sleep disturbance T‐score.
DISCUSSION
This study estimated adjusted associations between sleep disturbance and subsequent pain interference six months later among a large sample of patients with early RA receiving routine care in rheumatology practices across Canada. We found modest longitudinal associations between sleep disturbance and subsequent pain interference in both unadjusted and adjusted analyses. Our findings indicated that in patients with early RA, more disturbed sleep was associated with greater pain interference six months later. This association persisted even after accounting for potential confounders, such as steroid use, DMARD therapy, and other demographic and clinical factors.
To our knowledge, this is the first study to examine the association of sleep disturbances on subsequent pain interference in patients with newly diagnosed (early) RA over 24 months. Although the overall effects observed were modest, the relationship between sleep disturbances and subsequent pain interference remained significant. Our findings align with prior reports in established RA cohorts, highlighting a significant association between sleep disturbances and various pain measures. 4 , 6 , 11 , 12 Collectively, these results suggest that sleep disturbances could play a role in both the onset and persistence of pain in RA. In addition, there is also prior evidence supporting the inverse relationship between sleep and pain in patients with early RA. A large population‐based study investigated predictors of self‐reported sleep measures among Swedish patients with RA with a disease duration of 1 to 12 years 9 and showed that problems with sleep increased with disease duration. Pain attributed to RA, assessed by a numeric rating scale, and functional impairment were the strongest predictors of reduced sleep quality. Although our study differs in terms of analysis, assessments, and follow‐up periods, a reciprocal relationship likely exists between sleep problems and pain outcomes, whereby each influences the other in addition to influences from other factors over time.
A particularly noteworthy contribution of our study is the unique focus on the directional impact of sleep disturbances on pain interference, providing a novel and clinically meaningful perspective. Using a lagged repeated‐measures design in an early RA cohort, we were able to capture within‐person changes over time, offering a more robust understanding of how sleep disturbances may influence pain interference. This approach advances the understanding of the sleep‐to‐pain pathway and highlights the importance of considering temporal relationships in this context. Importantly, our outcome of interest (pain interference) is distinct from pain intensity, which has been the most used pain assessment among previous studies. Pain interference encompasses the extent to which pain disrupts daily activities and quality of life, reflecting a broader and more functional dimension of the pain experience. Although the pathways linking sleep disturbances and pain interference likely involve pain intensity, other factors may also contribute, warranting further exploration in future research.
One potential factor linking these constructs is depression. In cross‐sectional investigations, depression has been significantly associated with sleep disturbances in RA cohorts. 3 , 6 , 7 Our sensitivity analysis accounting for time‐varying symptoms of depression demonstrated similar findings as the main model, though there was a slight decrease in the β coefficient for sleep disturbance. This observation suggests that depressive symptoms may partially explain the relationship between sleep disturbances and subsequent pain interference. In other words, depressive symptoms may contribute to, but do not fully account for, the impact of sleep disturbances on pain interference.
Dysregulated central pain processing may be another mechanism linking sleep disturbance with pain interference. In a cross‐sectional analysis of 58 women with RA and 54 matched controls, sleep disturbances partially mediated the relationship between RA and abnormalities in descending pain inhibition. 10 In other words, patients with RA may have abnormalities in descending pain inhibition, at least in part, because they are not sleeping well. In a longitudinal analysis in a different cohort, sleep disturbance predicted higher pain intensity. This relationship was mediated by enhanced pain sensitivity and ascending pain facilitation. 11 Together, these findings suggest that underlying abnormalities in pain processing may be another potential link between sleep disturbances and pain interference.
Inflammation may also play an important role in the association between sleep disturbance and pain. Studies have demonstrated that poor sleep could lead to elevated levels of proinflammatory cytokines, which may exacerbate pain. 19 , 20 This inflammatory response could serve as a biologic mediator between sleep disturbance and heightened pain interference. This potential pathway warrants further exploration of inflammatory biomarkers in future research. More longitudinal studies in patients with early RA not only investigating the underlying causes of sleep disturbances but also examining how confounding processes may mediate the relationship between sleep disturbance and pain outcomes are needed to provide a more comprehensive understanding of the sleep–pain connection.
Several noteworthy implications emerge from our study. First, our findings support the importance of monitoring and addressing sleep disturbances in the comprehensive management of RA. Early intervention might be particularly crucial to mitigate the development of persistent sleep problems and adverse pain outcomes. Additionally, various self‐report instruments are available for assessing pain in RA, and our choice to focus on pain interference is motivated by its reflection of pain consequences on daily functioning, encapsulating both pain and overall function. Given the variable responses to pain among patients, with some adapting to limitations and others avoiding activities exacerbating pain, measures such as PROMIS pain interference may offer a valuable supplement in research and clinical care. Future studies should consider incorporating patient‐reported sleep disturbances and pain interference because these measures may provide essential insights into tracking improvements in RA management.
It is important to acknowledge that although the effect size for the relationship between sleep disturbance and pain interference was statistically significant, the magnitude of effect was modest. However, our results were consistent across several sensitivity analyses, suggesting that these relationships are real. Although this effect may not have a direct clinical impact, we still believe it is a valuable contribution to understanding the nuanced relationship between sleep disturbance and pain interference. These findings provide support for designing and implementing additional studies to further probe these relationships. There may be subgroups of patients in whom these relationships are stronger. It is also possible that our patient‐reported measure of sleep disturbance was not nuanced enough to capture specific types of sleep disturbance that may have a greater impact on pain interference (eg, sleep duration or fragmentation, sleep efficiency, circadian rhythm disorders). Further research is needed to identify (1) specific subgroups for whom a sleep‐targeted intervention may be particularly beneficial and (2) specific types of intervention (eg, sleep restriction, cognitive behavioral therapy, light therapy) that may be particularly effective for minimizing pain interference.
A major strength of our study includes the real‐world sample of patients with early RA. Additionally, the longitudinal design with lagged repeated measurements allowed us to explore the relationship between sleep disturbance and pain interference in the early stages of the disease over the first 24 months following diagnosis. There are a few limitations to our findings. First, we used patient‐reported assessments for sleep and pain interference to address our hypothesis. Reliance on patient‐reported data introduces potential bias, as certain participants may consistently report more symptoms or experiences, which could influence observed associations. Results may vary across different forms of assessments, such as objectively measured sleep parameters (eg, actigraphy, polysomnography). However, there are strengths to using patient‐reported measurements. These measures reflect the patient experience and are feasible to implement, requiring minimal effort to administer and score. Additionally, this study was not set up to examine impact closer in time. Examining the relationship between sleep disturbances and pain interference from day to day or over shorter time frames (eg, three months) may yield different findings. Similarly, we were unable to assess the impact of duration of sleep disturbances, such that persistent or temporary sleep problems may impact pain outcomes differently. Lastly, although we accounted for several important confounders, it is unlikely we were able to eliminate all potential sources of biases from unmeasured confounding. Further investigations are warranted to examine how other factors may contribute to or potentially mediate the relationship between sleep disturbances and pain interference.
In conclusion, our study reveals a consistent and significant association between heightened sleep disturbance and the subsequent escalation of pain interference over time. These results highlight the critical role of addressing sleep disruptions as an integral component of pain management strategies, particularly in the early stages following RA diagnosis. Identification and early intervention in problematic sleep patterns may contribute to enhanced long‐term pain outcomes.
AUTHOR CONTRIBUTIONS
All authors contributed to at least one of the following manuscript preparation roles: conceptualization AND/OR methodology, software, investigation, formal analysis, data curation, visualization, and validation AND drafting or reviewing/editing the final draft. As corresponding author, Dr Aydemir confirms that all authors have provided the final approval of the version to be published and takes responsibility for the affirmations regarding article submission (eg, not under consideration by another journal), the integrity of the data presented, and the statements regarding compliance with institutional review board/Declaration of Helsinki requirements.
ROLE OF THE STUDY SPONSOR
Pfizer Canada, AbbVie, Hoffman‐La Roche, Sandoz Biopharmaceuticals Canada, Fresenius Kabi Canada, Viatris Canada, Jamp Pharma, Celltrion Healthcare Canada, Amgen Canada, Janssen Canada, UCB Canada, Bristol‐Myers Squibb Canada, Medexus Pharmaceuticals, Sanofi Genzyme, Eli Lilly Canada, Merck Canada, Gilead Sciences Canada, and Organon Canada played no part in planning or conducting the study. Publication of this article was not contingent upon the approval of these funding sources.
Supporting information
Disclosure form.
Supplementary Table 1: Lagged Linear Mixed Effects Regression Models Estimating Association of categorical PROMIS Sleep Disturbance with Pain Interference over 2 years of Follow Up (N = 502).
Supplementary Table 2: Multivariable Linear Mixed Effects Regression Model Estimating Adjusted Effects of Sleep Disturbance with Pain Interference, Including Time‐Varying PROMIS Depression t‐scores as a Covariate (N= 502).
Supplementary Table 3: Linear Mixed Effects Regression Models Estimating Association of PROMIS Sleep Disturbance with Pain Interference Over 2 years of Follow Up in Patients with Early RA (N = 502). Sensitivity analysis adjusting for baseline and lagged (time variant) PROMIS t‐score for pain interference as a covariate.
ACKNOWLEDGMENT
The authors would like to thank Dr Edward Keystone for his contributions to the CATCH study.
The Canadian Early Arthritis Cohort study was independently designed and implemented by the investigators. It has been financially supported through unrestricted research grants from Pfizer Canada, AbbVie Corporation, Hoffman La Roche Limited, Sandoz Biopharmaceuticals Canada, Fresenius Kabi Canada, Viatris Canada, Jamp Pharma (BIOJAMP), Celltrion Healthcare Canada, Amgen Canada, Janssen Canada, UCB Canada, Bristol‐Myers Squibb Canada, Medexus Pharmaceuticals, Sanofi Genzyme, Eli Lilly Canada, Merck Canada, Gilead Sciences Canada, and Organon Canada. Dr Aydemir's work was supported by the National Institute of Arthritis and Musculoskeletal and Skin Diseases (NIAMS), NIH (grant T32‐AR‐007611) and National Center for Advancing Translational Sciences, NIH (grant K12‐TR‐005104). Dr Muhammad, Ms. Song, and Drs Dunlop and Chang's work was supported by NIAMS, NIH (grant P30‐AR‐072579). Dr Lee's work was supported by NIAMS, NIH (grants R01‐AR‐064850 and K24‐AR‐080840).
Additional supplementary information cited in this article can be found online in the Supporting Information section (https://acrjournals.onlinelibrary.wiley.com/doi/10.1002/acr.25568).
Author disclosures are available at https://onlinelibrary.wiley.com/doi/10.1002/acr.25568.
REFERENCES
- 1. Heiberg T, Finset A, Uhlig T, et al. Seven year changes in health status and priorities for improvement of health in patients with rheumatoid arthritis. Ann Rheum Dis 2005;64(2):191–195. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Katz P, Pedro S, Michaud K. Sleep disorders among individuals with rheumatoid arthritis. Arthritis Care Res (Hoboken) 2023;75(6):1250–1260. [DOI] [PubMed] [Google Scholar]
- 3. Hughes M, Chalk A, Sharma P, et al. A cross‐sectional study of sleep and depression in a rheumatoid arthritis population. Clin Rheumatol 2021;40(4):1299–1305. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Grabovac I, Haider S, Berner C, et al. Sleep quality in patients with rheumatoid arthritis and associations with pain, disability, disease duration, and activity. J Clin Med 2018;7(10):336. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Drewes AM, Jennum P, Andreasen A, et al. Self‐reported sleep disturbances and daytime complaints in women with fibromyalgia and rheumatoid arthritis. J Musculoskeletal Pain 1994;2(4):15–31. [Google Scholar]
- 6. Nicassio PM, Ormseth SR, Kay M, et al. The contribution of pain and depression to self‐reported sleep disturbance in patients with rheumatoid arthritis. Pain 2012;153(1):107–112. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Taylor‐Gjevre RM, Gjevre JA, Nair B, et al. Components of sleep quality and sleep fragmentation in rheumatoid arthritis and osteoarthritis. Musculoskeletal Care 2011;9(3):152–159. [DOI] [PubMed] [Google Scholar]
- 8. Westhovens R, Van der Elst K, Matthys A, et al. Sleep problems in patients with rheumatoid arthritis. J Rheumatol 2014;41(1):31–40. [DOI] [PubMed] [Google Scholar]
- 9. Lyne L, Åkerstedt T, Alfredsson 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. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Lee YC, Lu B, Edwards RR, et al. The role of sleep problems in central pain processing in rheumatoid arthritis. Arthritis Rheum 2013;65(1):59–68. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Song J, Muhammad LN, Neogi T, et al. Pain sensitization as a potential mediator of the relationship between sleep disturbance and subsequent pain in rheumatoid arthritis. Arthritis Care Res (Hoboken) 2023;75(4):778–784. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Irwin MR, Olmstead R, Carrillo C, et al. Sleep loss exacerbates fatigue, depression, and pain in rheumatoid arthritis. Sleep 2012;35(4):537–543. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Bacalao EJ, Greene GJ, Beaumont JL, et al. Standardizing and personalizing the treat to target (T2T) approach for rheumatoid arthritis using the Patient‐Reported Outcomes Measurement Information System (PROMIS): baseline findings on patient‐centered treatment priorities. Clin Rheumatol 2017;36(8):1729–1736. [DOI] [PubMed] [Google Scholar]
- 14. Bartlett SJ, Orbai AM, Duncan T, et al. Reliability and validity of selected PROMIS measures in people with rheumatoid arthritis. PLoS One 2015;10(9):e0138543. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Bykerk VP, Jamal S, Boire G, et al. The Canadian Early Arthritis Cohort (CATCH): patients with new‐onset synovitis meeting the 2010 ACR/EULAR classification criteria but not the 1987 ACR classification criteria present with less severe disease activity. J Rheumatol 2012;39(11):2071–2080. [DOI] [PubMed] [Google Scholar]
- 16. Katz P, Pedro S, Michaud K. Performance of the Patient‐Reported Outcomes Measurement Information System 29‐item profile in rheumatoid arthritis, osteoarthritis, fibromyalgia, and systemic lupus erythematosus. Arthritis Care Res (Hoboken) 2017;69(9):1312–1321. [DOI] [PubMed] [Google Scholar]
- 17. PROMIS reference populations . Accessed January 7, 2024. https://www.healthmeasures.net/score‐and‐interpret/interpret‐scores/promis/reference‐populations
- 18. England BR, Sayles H, Mikuls TR, et al. Validation of the Rheumatic Disease Comorbidity Index. Arthritis Care Res (Hoboken) 2015;67(6):865–872. [DOI] [PubMed] [Google Scholar]
- 19. Irwin MR, Straub RH, Smith MT. Heat of the night: sleep disturbance activates inflammatory mechanisms and induces pain in rheumatoid arthritis. Nat Rev Rheumatol 2023;19(9):545–559. [DOI] [PubMed] [Google Scholar]
- 20. Irwin MR, Olmstead R, Bjurstrom MF, et al. Sleep disruption and activation of cellular inflammation mediate heightened pain sensitivity: a randomized clinical trial. Pain 2023;164(5):1128–1137. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Arnett FC, Edworthy SM, Bloch DA, et al. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum 1988;31(3):315–324. [DOI] [PubMed] [Google Scholar]
- 22. Aletaha D, Neogi T, Silman AJ, et al. 2010 Rheumatoid arthritis classification criteria: an American College of Rheumatology/European League Against Rheumatism collaborative initiative. Arthritis and Rheum 2010;62(9):2569–2581. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Disclosure form.
Supplementary Table 1: Lagged Linear Mixed Effects Regression Models Estimating Association of categorical PROMIS Sleep Disturbance with Pain Interference over 2 years of Follow Up (N = 502).
Supplementary Table 2: Multivariable Linear Mixed Effects Regression Model Estimating Adjusted Effects of Sleep Disturbance with Pain Interference, Including Time‐Varying PROMIS Depression t‐scores as a Covariate (N= 502).
Supplementary Table 3: Linear Mixed Effects Regression Models Estimating Association of PROMIS Sleep Disturbance with Pain Interference Over 2 years of Follow Up in Patients with Early RA (N = 502). Sensitivity analysis adjusting for baseline and lagged (time variant) PROMIS t‐score for pain interference as a covariate.