Abstract
Objective. Sleep disturbance and chronic fatigue are common in rheumatoid arthritis (RA) and contribute to disability, symptomatology, and healthcare use. It has long been recognized in other populations that exercise can improve sleep and diminish fatigue. The effect of exercise on sleep quality and fatigue in RA has not been evaluated.
Methods. Ours is a randomized controlled study in RA to determine the effect of an exercise program on sleep quality and fatigue. These were measured using the Pittsburgh Sleep Quality Index and the Fatigue Severity Scale. Patients were randomized to either a 12-week, home-based exercise intervention or usual care. The exercise program consisted of specific exercises to target individual deficiencies identified using the Health Assessment Questionnaire (HAQ) with cardiovascular work as per the guidelines. The intervention group was evaluated on a 3-week basis. Full evaluation was carried out at baseline and at 12 weeks.
Results. Forty patients were randomized to the intervention with 38 controls. In the exercise intervention group, there was a statistically significant improvement in HAQ (p = 0.00), pain (p = 0.05), stiffness (p = 0.05), sleep quality (p = 0.04), and fatigue (p = 0.04). In our control group, there was a statistically significant improvement demonstrated in their overall perceptions of the benefits of exercise, but none of the other variables.
Conclusion. Our study demonstrates that an exercise program resulted in significant improvement in sleep quality and fatigue. This is particularly interesting given the importance of fatigue as an outcome measure in RA and gives us yet another reason to prescribe exercise in this population.
Rheumatoid arthritis (RA) is associated with progressive functional disability and accelerated mortality1,2,3. Sleep disturbance is common in patients with RA4. Multiple causative factors include pain, depression, lack of exercise, restless legs, and corticosteroid use5,6,7,8,9. Disordered sleep and fatigue are thought to play important roles in the development of chronic pain6,7,10. Poor sleep is a common complaint in the general population11,12, but is even more common in those with rheumatic diseases5,7,8. Chronic insomnia in the general population has been shown to compromise quality of life, psychosocial well-being, and occupational and educational performance13. It results in increased morbidity11, mortality14, and healthcare use15. Pharmacological interventions aimed at improving sleep have shown short-term efficacy, but their longterm usefulness is hindered by dependency, increased mortality16, and the rapid development of tolerance17.
Exercise has long been recognized by sleep organizations as a key component in the nonpharmacological management of poor sleep18. Experimental studies have also shown that exercise improves sleep quality19,20. Patients with RA were previously cautioned regarding participation in cardiovascular (CV) exercise owing to the belief that it was deleterious to joints. For this reason, the role of exercise in improving sleep quality in RA has not been established.
Improvements in sleep quality and associated decreases in subjective fatigue are of particular importance in RA. Fatigue is considered in the Outcome Measures in Rheumatology Clinical Trials (OMERACT)21 as of 2011 and relates directly to quality of life22. A recent Cochrane review examined fatigue in RA and showed that exercise has a modest effect on fatigue; however, the specific issue of sleep quality has not been examined23.
Our study was undertaken to evaluate the effect of an exercise program on self-reported sleep quality and fatigue in RA.
MATERIALS AND METHODS
Seventy-eight patients with established RA according to the American College of Rheumatology criteria were recruited from the rheumatology outpatient clinic of a large teaching hospital. They were randomized using a preassigned protocol to receive either standard care with information regarding the benefits of exercise in RA or to take part in an exercise program. Forty individuals participated in the exercise program while the remaining 38 received standard care and were given verbal and written instruction regarding the benefits and importance of exercise in RA.
Exclusion
Patients were excluded from participation if they were not independently mobile, were deemed a falls risk or had severe medical conditions that were more limiting than their arthritis. These conditions were congestive heart failure with functional limitation (New York Heart Association level III symptoms or greater), angina, active malignancy, uncontrolled thyroid disease, severe chronic obstructive pulmonary disease (gold stage III or greater), or neurologic condition limiting mobility. Patients were also excluded if they did not speak English or were unable to give informed consent. Because we are a tertiary referral center with patients attending the clinic from distant areas, we limited our study to those who could easily travel for assessments. Those who lived farther than 1 h of travel time were excluded. Ethical approval was granted by the St. James’s Hospital Ethics Committee.
Medical assessment
Medical history was reviewed to ensure suitability and current medications were documented. Smoking history was recorded. Disease-specific characteristics were assessed. Disease activity was evaluated using the Disease Activity Score with 28 joint counts24. Participants were described as having seropositive disease if either rheumatoid factor or anticyclic citrullinated peptide antibodies were present. If erosions were visible on plain film evaluation of hands or feet, they were described as having erosive disease.
Self-reported measures
For self-reported measures, patients were given the relevant questionnaires to complete while in the clinic waiting room with a doctor available to answer any questions.
Functional limitation was quantified using the Health Assessment Questionnaire (HAQ) Disability Index25,26. Both pain and stiffness were determined using visual analog scale (VAS).
Fatigue was assessed using the Fatigue Severity Scale (FSS)27 and sleep quality was measured using the Pittsburgh Sleep Quality Index (PSQI)28.
The FSS is widely used across a variety of chronic conditions. It is composed of 9 statements concerning the respondent’s fatigue. The scale measures how fatigue affects motivation, exercise, physical functioning, and carrying out of duties, and how it interferes with work, family, and social life. It was originally developed for use in multiple sclerosis and systemic lupus erythematosus (SLE), but has since been used across a broad range of rheumatic and other chronic conditions. Responses are scaled on a 7-point Likert scale from 1 = strongly disagree to 7 = strongly agree. Responses are summed and divided by the number of answers to yield a score with a range of 1–7, with higher scores indicating more fatigue. The FSS has been shown to correlate with VAS and to inversely correlate with the Rand Index of Vitality29. The FSS has a clearly outlined minimal clinically important difference. A systemic metaanalysis of the use of this tool in SLE suggested a change of 15% should be considered clinically meaningful30.
The PSQI is used to measure sleep quality and disturbances over the preceding month28,31. It is composed of 7 variables: subjective sleep quality, sleep latency, sleep duration, habitual sleep efficiency, sleep disturbances, use of sleeping medication, and daytime dysfunction. Nineteen items are included in scoring. Five additional items, which are completed by a bed partner, if applicable, are not included in the total score because they may not be available. The PSQI has been used in multiple disease areas and as an outcome in clinical trials. These include, among others, cancer32,33, restless legs34, gastroesophageal reflux35, and obstructive sleep apnea36. It has also been used to evaluate sleep quality in RA37. The PSQI has been used to define inclusion criteria for poor sleep quality where a score of over 5 is considered to indicate troublesome sleep. The range for this tool is 0–21, with a higher score indicating worse sleep quality. When scoring this tool, each of the 7 sleep variables are scored separately and then summed to give an overall score. The component scores correlate well with sleep latency, duration, and quality on subsequent polysomnography testing. The PSQI has demonstrated sensitivity to change.
Subjects’ opinions regarding exercise and the barriers they encounter that could limit their exercise were evaluated using the Exercise Benefits and Barriers Scale. This was developed by Sechrist, et al38 to determine perceptions of individuals concerning the benefits of and barriers to participating in exercise. It is based on 43 questions. There are 29 benefits items in 5 categories. These are physical performance, preventive health, psychological outlook, social interaction, and life enhancement. There are 14 barriers items in 3 categories: physical exertion, time expenditure, and exercise environment. Individuals rate their agreement to each perceived benefit and barrier item on a Likert scale consisting of 4 answer options from strongly disagree to strongly agree. The possible range of scores on the questionnaire as a whole is 43–172, and higher scores indicate a more positive perception of exercise. For the barriers scale, the range is 14–56, higher scores indicating a greater perception of barriers to exercise.
The intervention program
A 12-week home exercise program was prescribed for the intervention group. Participants were assessed by a doctor (LD) and senior physiotherapist (FW) at baseline and then every 3 weeks for the duration of the program. Functional limitation was assessed by HAQ. A full physical examination was carried out by a physiotherapist to identify deficiencies or functional limitations in muscular fitness, range of motion, and coordination of affected joints. Specific exercises were prescribed to target the individual deficiencies identified. Resistance exercise was prescribed 3 times per week, and dosage was prescribed and increased according to the American College of Sports Medicine (ACSM) guidelines for healthy individuals, based on their advice for sedentary persons beginning a resistance program39 (with adaptation where required to avoid aggravation of symptoms where no specific guidelines are available40; Table 1). Range of motion exercise was prescribed in functional patterns to be done on a daily basis.
Methods of exercise prescription used in this study.
In addition to these strengthening and stretching exercises, a walking program was devised according to ACSM guidelines on physical activity39. Based on their CV expenditure prior to our study and their functional capability as measured by a 6-min walk test, a program was devised that included incremental targets for daily walks based on step count and rate of perceived exertion. Patients were given daily step count targets and advised on the level of exertion for which they should be aiming. They were instructed that they should be moderately short of breath on exertion, i.e., unable to comfortably hold a conversation while walking.
The control group
The control group was composed of patients with RA who received advice only on the benefits of exercise in RA. They were assessed at baseline and at 12 weeks.
Statistical analysis
SPSS version 18 was used for analysis. The primary endpoint was sleep improvement, although a lack of consensus on what could be considered a minimum clinically important difference in sleep quality meant that we specifically powered the work to detect a clinically important difference in HAQ. A sample size of 38 patients in each group was calculated to detect a HAQ difference of 0.25 between them with a 90% power and significance level of 0.05. To allow for potential noncompleters, we aimed to recruit 40 individuals per group. Randomization was performed using Excel random numbers. All means are expressed with their SD provided normally distributed and with their interquartile range in the case of non-normal data. Data were assessed for normality using the Kolmogorov-Smirnov test. Before and after scores, provided they were normally distributed, were assessed using a paired t test. The magnitude of changes seen in both the intervention and control group were then compared, and α was set at 0.05.
RESULTS
Four hundred potential participants with RA were assessed for eligibility. Of the 400, 220 lived within an easily accessible distance. From these 220 potential participants, 136 were suitable for the program based on our exclusion criteria. One hundred thirty-six patients were approached and given information regarding the exercise program. Eighty individuals were interested in taking part. All were given at least 24 h “opt out time” as per our protocol. They were then randomized to either intervention or control group (Figure 1). Forty-two participants were randomized for intervention; however, 2 of them had abnormalities at their initial assessment, which meant they were unable to take part. One individual described exertional chest pain and another was found to have a new diagnosis of uncontrolled hyperthyroidism.
Diagram outlining patient recruitment. RA: rheumatoid arthritis.
All participants were assessed at baseline and following the 12-week exercise program. Interim evaluations were carried out at 3 weeks, 6 weeks, and 9 weeks. All participants who completed the program attended at least 2 of the 3 interim assessments. Three individuals (7.5%) missed 1 assessment.
The demographics and disease-specific characteristics of the group are outlined in Table 2.
Comparison of characteristics of intervention group to RA controls at baseline.
Sleep quality as measured by the PSQI was 5.6 in the control group and 7.2 in the intervention group. This was not statistically significant. The level of perceptual barriers to exercise was comparable in both groups at 31.2 and 28.6. Their overall opinions of the benefits of exercise were also similar at 126.3 in the control group and 125.8 in those who took part in the exercise intervention. Levels of fatigue were similar in both groups at 30.5 and 29.5 in the control and intervention groups.
In both groups, the most commonly reported statement relating to fatigue and quality of life was “Fatigue is among my three most disabling symptoms”. This was reported as being greater than 5 (on a scale of 0–7) by 75%. The second most commonly reported statement was “I am easily fatigued”. This was reported as greater than 5/7 by 70%. A score of greater than 36 is thought to represent debilitating fatigue. This was present in 27 participants (35%).
Overall, 38 individuals (49%) had a score of greater than 5 indicating problematic sleep. Table 3 outlines the specific sleep problems encountered. With regard to the individual components of sleep where disturbances were identified, 70 (90%) reported poor sleep quality subjectively, 56 (72%) had problems with sleep latency, 56 (72%) had disturbed duration of sleep, 29 (37%) had problems with habitual sleep efficacy, 71 (91%) reported disturbed sleep, 23 (29%) required sleeping tablets, and 65 (83%) reported daytime dysfunction as a result of sleep disturbance. The intervention and control groups demonstrated similar proportions of patients with difficulty in the individual components of sleep.
Pittsburgh Sleep Quality Index, components, and no. participants.
Table 4 demonstrates the differences seen after 12 weeks in both the intervention and control groups. In those who took part in the exercise intervention, there was a statistically significant improvement in HAQ, pain, stiffness, and perceptions regarding the benefits of exercise, sleep quality, and fatigue. In our control group, there was a statistically significant improvement demonstrated in their overall perceptions of the benefits of exercise, but there were no benefits observed for any of the other variables.
Difference in self–reported outcomes in the intervention and control groups at baseline and following 12 weeks. Pain is the level of pain measured on the visual analog scale (VAS). Stiffness is the level of stiffness measured on the VAS. “Barriers” is the perceptual barriers to exercise as measured by the Exercise Benefits and Barriers Scale (EBBS). EBBS indicates the overall perception of exercise. FSS is the level of fatigue as measured by the Fatigue Severity Scale.
In Table 5, the changes seen in both the control and intervention groups are analyzed. On comparison of the control group to the intervention group, there were significantly greater improvements in pain, stiffness, subjective sleep quality, and fatigue in the intervention group. There was a nonsignificant difference in the subject’s feelings regarding their barriers to exercise and their opinions regarding its benefits.
Comparison of intervention group to control group, and changes seen in clinical measures. Pain is the level of pain measured on the visual analog scale (VAS). Stiffness is the level of stiffness measured on the VAS. “Barriers” is the perceptual barriers to exercise as measured by the Exercise Benefits and Barriers Scale (EBBS). EBBS indicates the overall perception of exercise. FSS is the level of fatigue as measured by the Fatigue Severity Scale.
DISCUSSION
Our study evaluated the effect of a targeted exercise intervention on sleep quality and fatigue in RA. It showed that an exercise program focusing on functional disability with CV targets had a significant and clinically important effect on fatigue and sleep quality. This intervention also yielded significant improvements in pain, stiffness, and functional disability.
Fatigue is an important outcome measure and is experienced in up to 90% of patients with RA41,42. This high prevalence is likely multidimensional, reflecting disease severity, inflammatory burden, functional limitation, pain, stiffness, and depression22,41,43,44,45,46. Fatigue relates directly to quality of life22 and is considered in the outcome measures in OMERACT21 quality of care domains. It is one of the variables used in assessing response to treatment in clinical trials. A metaanalysis evaluating the effect of biologic therapies on fatigue demonstrated a modest effect at best47. This makes any intervention that influences fatigue particularly important.
Fatigue and poor sleep are known to associate with chronic pain. Studies based in pain clinics commonly link poor sleep with higher levels of pain in those who have sleep disturbances48. This has been shown across a wide variety of chronic pain conditions. In fibromyalgia, poor sleep is almost universal and those who sleep badly have a higher burden of symptoms49. Patients with RA commonly have disturbed sleep5,7,37, which is likely multifactorial and influenced by pain, corticosteroid use, depression, lack of exercise, and restless legs. This contributes to fatigue, symptomatology, healthcare use, and poor quality of life.
In keeping with our findings, a study evaluating sleep quality in RA demonstrated PSQI scores over 5 in 50% of the participants10. A score of 5 or greater has high diagnostic specificity for detecting clinical sleep impairment as defined by the diagnostic criteria for insomnia in the general population. The findings are also in keeping with those seen in ankylosing spondylitis (AS). A study demonstrated 35.4% of patients with AS had poor sleep as measured by the PSQI with a mean total score of 6.6250.
The current guidelines for physical activity from the ACSM and the World Health Organization are at least 150 min per week of moderate-intensity aerobic exercise. These recommendations also include advice on muscular strength, endurance, flexibility, and balance. They appear to be sufficient to improve subjective sleep quality in other populations20,51,52. Generally, in those who have dysfunctional sleep, exercise is beneficial. This has been shown in many studies. A study by Irwin, et al, which examined the effects of tai chi on older adults, found improvements among poor sleepers engaging in this form of exercise53. King, et al examined the effect of CV exercise on sleep quality in older, inactive adults with moderate sleep complaints. They found significant improvements in sleep quality as measured by the PSQI52. Further study by King, et al examined the effect of a year-long exercise intervention on sleep quality in sedentary adult caregivers. They found significant improvements in the PSQI with inverse correlations with self-perceived stress51. These findings are in keeping with our work, which demonstrates a significant improvement in sleep quality and fatigue as a result of a physical activity intervention.
Sleep professionals have long advocated exercise. There are many proposed mechanisms by which exercise improves sleep. Exercise has well-described antidepressant effects and it reduces anxiety, which contributes to hyperarousal and is an important factor in insomnia. Exercise is also thought to improve circadian rhythms and thermoregulation, both of which affected sleep quality. Santos, et al suggested in a recent review that exercise affects sleep through modest elevations in the levels of proinflammatory cytokines, interleukin (IL)-1, IL-6, and tumor necrosis factor-α54. Modest concentrations are thought to promote sleep while higher volumes are associated with increased nighttime wakefulness54. This is in keeping with the findings of poor sleep in our patients with inflammatory arthritis and is consistent with other research suggesting that acute ultra-endurance activity causes greater elevations in inflammatory cytokines and can increase nighttime wakefulness55. Further, exercise has been shown in recent metaanalyses to decrease the erythrocyte sedimentation rate in RA56 and C-reactive protein in heterogeneous populations57. This supports the theory that physical activity can have antiinflammatory properties.
In our study, sleep quality was poor in 50% (n = 20) of the intervention and 47.4% (n = 18) of the control group. This improved significantly as a result of the exercise intervention. Consistent with this, fatigue also declined. These changes remained significant when changes in the control group were taken into account and is, to the best of our knowledge, the first time that the effect of exercise on sleep quality has been evaluated in RA.
The focus of our study was patient-based and for this reason, self-reported outcomes are relied upon and our patients did not undergo polysomnography testing. Although our numbers are small in both groups, they were adequate to demonstrate the difference that was present in the intervention arm of our study. We did not control for baseline differences of pain as this was not a primary outcome and is a fluctuant variable. This may have biased our results, although the groups were in all other ways comparable. Challenges for the future include the implementation of more longterm strategies for increasing physical activity in our patient population and reevaluating those who took part in our 12-week program to see whether it has had a lasting effect.
Fatigue, poor sleep, functional limitation, and chronic pain are common in RA. A 12-week exercise program can yield significant and clinically important improvements in fatigue and sleep quality. Exercise has long been encouraged in both health and chronic disease for its numerous health benefits. To our knowledge, this is the first time the effect of exercise on sleep and fatigue in RA has been explored. It gives us further reasons to encourage and specifically prescribe exercise in RA.
- Accepted for publication June 4, 2014.
