Sleep problems in rheumatoid arthritis over 12 years from diagnosis: results from the Swedish EIRA study

Abstract

Objective: Most studies of rheumatoid arthritis (RA) and sleep have focused on established RA. We here investigate sleep quality and sleep duration in patients with newly diagnosed RA and during 1-12 years after diagnosis.

Methods: Data were collected on sleep 1-12 years after diagnosis from patients diagnosed 1998-2018 in the Swedish study Epidemiological Investigation of RA. Six sleep domains (sleep problems, non-restorative sleep, insomnia, insufficient sleep, sleep quality perceived as poor and sleep considered a health problem); a global sleep score and time spent in bed were estimated. Using logistic regression, ORs were calculated for each sleep outcome by disease duration. We explored whether pain (low (Visual Analogue Scale=0-20 mm, reference), intermediate=21-70, high=71-100) or functional impairment (Health Assessment Questionnaire>1.0) was associated with problems.

Results: We had sleep data on 4131 observations (n=3265 individuals). Problems with ≥1 sleep domain (global sleep score) was reported in 1578 observations (38%) and increased with disease duration (OR 1.04, 95% CI 1.02 to 1.07). Median time in bed was 8 hours (Q1-Q3: 7.5-9.0). High-grade pain increased the likelihood of sleep problems ~3-9 fold, and increased functional impairment ~4-8 fold.

Conclusion: In this cohort of newly diagnosed patients with RA with access to the current treatment from diagnosis, we did not find any major problems with sleep, and existing sleep problems related mainly to pain and reduced function. Treatment of sleep problems in RA should be guided towards treating the underlying problem causing the sleep disturbance.

Keywords: epidemiology; patient reported outcome measures; rheumatoid arthritis.

Figure 1

Sleep problems by domain and disease duration. Proportion of patients experiencing problems with each sleep domain stratified by disease duration using 4131 observations on sleep from 3265 individuals newly diagnosed with RA during 1996–2018 in Sweden and with a disease duration of 1–12 years at the time of data collection. Error bars represent the 95% binomial CIs calculated with the normal approximative method. RA, rheumatoid arthritis.

Figure 2

Sleep duration by disease duration. Sleep duration by disease duration using 4131 observations from 3265 individuals newly diagnosed with RA during 1996–2018 in Sweden. The boxplot represents the median, upper quartile, lower quartile, maximum and minimum values of sleep duration by disease duration. RA, rheumatoid arthritis.

Figure 3

Sleep problems by pain level. Proportion of patients experiencing problems with each sleep domain stratified by pain level using 4131 observations for 3265 individuals newly diagnosed with RA during 1996–2018 in Sweden and with a disease duration of 1–12 years at the time of data collection. Error bars represent the 95% binomial CIs calculated with the normal approximative method. RA, rheumatoid arthritis.

Figure 4

Sleep problems by disability. Proportion of patients experiencing problems with each sleep domain stratified by HAQ-level using 4131 observations for 3265 individuals newly diagnosed with RA during 1996–2018 in Sweden and with a disease duration of 1–12 years at the time of data collection. Error bars represent the 95% binomial CIs calculated with the normal approximative method. HAQ, Health Assessment Questionnaire; RA, rheumatoid arthritis.