https://orcid.org/0000-0003-0614-5561
Abstract
This systematic review and meta-analysis synthesizes the evidence on prospective bidirectional associations between sleep-related problems (SRP) and chronic musculoskeletal pain (CMP).
A literature search for cohort studies available in the PubMed, Scopus, Web of Science, PsycINFO and Cochrane Library databases as of 19 July 2022 was performed. Pooled odds ratios and effect sizes were calculated through random effects meta-analysis. Subgroup and meta-regression analyses were performed to explore differences by follow-up time, proportion of each sex and mean age. The Meta-analysis Of Observational Studies in Epidemiology guidelines were strictly followed.
Twenty studies with a total of 208 190 adults (aged 34.4–71.7 years) were included, with 17 of them being used in the meta-analysis. Individuals with SRP at baseline had a 1.79-fold higher incidence (odds ratio [OR] = 1.79; 95% CI: 1.55, 2.08; I2 = 84.7%; P < 0.001) and a 2.04-fold higher persistence (OR = 2.04; 95% CI: 1.42, 2.94; I2 = 88.5%; P < 0.005) of CMP than those without SRP. In the subgroup analysis of the association between SRP and CMP, the longer the follow-up time of the studies, the higher the heterogeneity between them. In the corresponding meta-regression, no significant effect was observed for follow-up time, sex proportion or age. Individuals with CMP at baseline had a 2.02-fold higher incidence of SRP (OR = 2.02; 95% CI: 1.62, 2.53; I2 = 90.0%; P < 0.001) than those without CMP.
This study provides robust evidence concerning the longitudinal association between SRP and incidence-persistence of CMP in adults. In addition, the available prospective studies support the existence of a bidirectional relationship between CMP and SRP.
CRD42020212360
Individuals with sleep-related problems were more likely to have incident and persistent chronic musculoskeletal pain.
Chronic musculoskeletal pain is associated with the incidence of sleep-related problems.
Chronic musculoskeletal pain and sleep-related problems are bidirectionally associated.
Introduction
Recently, as a consequence of stressful behaviours related to high work demand [1] and increasing use of smartphones [2], there have been changes in lifestyle with potentially harmful health-related implications with poorer quality of life and health problems [1]. Commonly diagnosed or self-reported problems include hypertension, diabetes [3], stress, depression, anxiety [4], chronic pain and the exponential increase in sleep-related problems (SRP) [5].
Epidemiological studies suggest an increase specifically in chronic sleep disorders, including obstructive sleep apnoea, daytime sleepiness and insomnia [6–9]. The onset and persistence of symptoms of insomnia are related to any factor that interferes with the continuity of sleep, either by awakenings caused by age-related comorbidities or by the sensitivity of the noradrenergic locus coeruleus [10]. A meta-analysis of 200 358 people from the general population of the Netherlands, the UK and the USA estimated that 23.0% of the participants aged 18–25 years could not fall asleep easily, and 24.0% of the individuals older than 65 years presented problems with waking up in the morning [11]. On the other hand, studies have noted that SRP, including insomnia and its symptoms, increases the risk of presenting chronic musculoskeletal pain (CMP) over time [12, 13]. Longitudinal research with adults has highlighted insomnia as a risk factor for chronic pain in the neck, knees, legs, feet [13], back and shoulders [14]. However, scientific evidence underscores that the relationship between sleep and chronic pain may be bidirectional, i.e. chronic [15, 16] musculoskeletal pain [17] may act as a predictor of the occurrence of sleep disturbances in the adult population.
Although there is growing evidence of a relationship between SRP and chronic pain, most studies on the subject in adult populations have been cross-sectional or in patients with general chronic pain, fibromyalgia and rheumatoid arthritis [18], thus hindering the production of longitudinal systematic reviews with the outcome of interest. Obtaining information over time about musculoskeletal symptoms and SRP reduces recall bias and provides more reliable information over time [19–21]. Additionally, it is necessary to understand the direction of the relationship that is established between the variables of interest. Although the methodological aspects listed are limitations that partially justify the present study, it is necessary to understand the complexity of this relationship to propose behavioural and sleep hygiene interventions and provide a better quality of life. Therefore, this systematic review and meta-analysis adds to the available evidence by synthesizing the results of primary cohort studies (observational prospective longitudinal studies) and calculating pooled estimators of the bidirectional relationship between SRP and CMP. Subgroup and meta-regression analyses explored whether these factors vary according to the mean age, sex proportion and follow-up time of the studies.
Methods
This systematic review with meta-analysis was performed according to the recommendations of the Cochrane Collaboration Handbook [22]. The Meta-analysis Of Observational Studies in Epidemiology (MOOSE) guidelines [23] were strictly followed. This review was registered in PROSPERO (registration number: CRD42020212360).
Search strategy and study selection
We systematically searched the PubMed, Scopus, Web of Science, PsycINFO and Cochrane Library databases from inception until 19 July 2022. The complete search strategy is detailed in Supplementary Data S1, available at Rheumatology online. The research focus was on peer-reviewed cohort studies, i.e. those analysing the prospective longitudinal relationship between SRP and CMP, and vice versa.
The criteria for inclusion of studies were as follows: (i) participants—adults (≥18 years); (ii) design—prospective observational studies with at least 1 year of follow-up and 100 or more participants to prevent low statistical power of small sample sizes; and (iii) exposure and outcome variables—presence, incidence and persistence of chronic (3 months or more) musculoskeletal pain or SRP. There were no language or time restrictions.
The exclusion criteria for the studies were as follows: (i) ineligible publication types, such as case series, editorials, non-peer reviewed preprints, meeting proceedings and letters to the editor; (ii) studies that did not work with body regions considered musculoskeletal (described in item definitions of chronic musculoskeletal pain and SRP); (iii) research with pain temporality less than 3 months; (iv) studies that did not explore the relationship between the exposure and the outcome of interest; (v) studies with only qualitative analyses, and animal experiments; and (vi) studies that focused specifically on fibromyalgia. Although musculoskeletal pain is frequent in fibromyalgia, this condition is characterized by widespread pain that may include other non-musculoskeletal body structures. Musculoskeletal pain, which was the only assessment symptom of fibromyalgia in the ACR criteria released in 1990 [24], was only one of the five criteria for identifying fibromyalgia in its 2010 update [25, 26]. Combining studies on fibromyalgia with those focused explicitly on chronic musculoskeletal pain was not considered appropriate because fibromyalgia is currently being discussed as a psychosomatic symptom or disorder [27, 28]. Although some aspects of both diseases may coincide, this factor adds an additional layer of complexity to their differentiation.
The literature search and study selection were performed independently by three reviewers (M.C.S.S., F.L.G. and S.M.A.), and disagreements were resolved by consensus or with the involvement of a fourth researcher (A.E.M.).
Definitions of chronic musculoskeletal pain and sleep-related problems
In this review, CMP refers to pain symptoms reported in the body regions of the neck, shoulders, wrists/hands, elbows, arms, back, knees, legs and feet/ankles; in addition, painful symptoms had to have been persistent in at least one of these regions for 3 months or more [29]. When analysing pain as an outcome variable, it was classified as either incident (presenting pain at follow-up but not at baseline) or persistent (presenting pain at both baseline and follow-up), whereas when analysing pain as an exposure variable, it was classified as prevalent.
SRP were defined as those described by the authors as insomnia, sleep apnoea, snoring, perception of poor sleep quality, trouble falling or staying asleep, waking up earlier than usual and not falling back to sleep easily, insufficient or non-restorative sleep, and daytime dysfunction. The period considered for research after analysing the studies was the last month or quarter of data collection.
Data extraction
The following data were extracted from the original articles: (i) authors and date of publication; (ii) country; (iii) direction of association; (iv) follow-up time; (v) sample characteristics (size, percentage of women and mean age); (vi) temporality, type, body regions and measurement of pain; (vii) measurement of SRP; (viii) adjustment variables; (ix) pain outcome—presence or absence of SRP according to incidence or persistence of CMP; and (x) SRP outcome—presence or absence of CMP according to the incidence of SRP.
Risk of bias assessment
The Quality Assessment Tool for Observational Cohorts and Cross-Sectional Studies was used to assess the risk of bias of the included studies [30]. This tool evaluates 14 criteria for prospective cohort studies. Each criterion could be scored as ‘yes’ when the study achieves the criterion, ‘no’ when the study does not achieve the criterion, and ‘not reported’ when the study does not clearly report the required information [30]. Following this risk of bias tool, studies could be rated as good (i.e. at least 11 criteria were met), fair (i.e. 6–10 criteria were met), or poor (i.e. 1–5 criteria were met) [30].
The data extraction and the risk of bias assessment were performed independently by three reviewers (M.C.S.S., F.L.G. and S.M.A.), and a fourth researcher resolved the inconsistencies (A.E.M.).
Statistical analyses and data synthesis
For meta-analysis, cohort studies evaluating either the risk of CMP in participants with SRP or the risk of SRP in participants with CMP were considered. Odds ratios (ORs) and their respective 95% CIs were calculated according to the estimator used in each study (means and proportions). Pooled ORs were estimated using the DerSimonian and Laird random-effects model [31, 32]. The I2 statistic was used to assess the heterogeneity of the results [33]. We applied subgroup analyses when the percentage of total variation between studies due to heterogeneity was moderate or high (I2 > 50%) [34]. The pooled OR for the prospective associations between SRP and the risk of developing CMP was calculated according to the follow-up time of the studies, the mean age of the participants and the proportion of women [35]. For this, two groups were considered for each variable according to the median distribution: follow-up time (≤10 and >10 years), mean age (<44 and ≥44 years) and female proportion (<54.0% and ≥54.0%).
Meta-regression models were performed to analyse the impact of the continuous variables follow-up time (years), female proportion (percent) and age (years) on the relationship between SRP and the incidence of CMP. Moreover, sensitivity analyses were conducted to evaluate the robustness of the summary estimates by removing included studies one by one, as well as by excluding studies evaluated as having fair quality and a high risk of bias. Finally, Egger's analysis and funnel plots were used to assess the publication bias of the studies.
Statistical analyses were performed using the metan, metareg and metabias commands of STATA SE software, version 16 (StataCorp, College Station, TX, USA). All analyses were performed by two reviewers (A.E.M. and M.C.S.S.).
Results
Fig. 1 shows the flowchart of the study selection process. We included 20 articles from 5763 non-duplicated records after initial selection based on title and abstract and subsequent selection based on full text (Supplementary Data S2, available at Rheumatology online). Of these, three studies were excluded from the meta-analysis, one because the author did not provide the data required within 30 days after two attempts, and two because the database was not available, resulting in the exclusion of 54 505 individuals from the meta-analysis. Therefore, 17 studies totalling 153 685 individuals were finally included in the meta-analysis. Among these, 11 analysed CMP as the outcome and SRP as the exposure variable [12–14, 36–43], four analysed SRP as the outcome and CMP as the exposure [21, 44–46], and two examined the bidirectional associations between CMP and SRP [47, 48] (Table 1). According to the Quality Assessment Tool [30], prospective cohort studies scored between 8 and 13 points (40% were rated as good quality and 60% as fair quality). The four least reported in the articles were varying levels of exposure, repeated exposure assessment, outcome blinding of the assessors to the participants’ exposure status, and loss to follow-up after baseline (Supplementary Data S3, available at Rheumatology online).
Flow diagram of the literature search and study selection
Characteristics of the included studies
| Author (year) | Association direction | Country | Follow-up time, mean, years | Women, n (%) | Mean age, years | Pain duration (months) | Measurement of CMP | Types of pain | Pain region | Measurement of insomnia | Adjustment variables |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Agmon and Armon (2014) [5] | Insomnia—BP | Israel | 3.7 | 2131 (34.0) | 46.2 | ≥ 3 | Medical record and interview | Incident | Back pain | AIS-5 | Gender, age, education, smoking, physical activity, self-rated health, body mass index and high-sensitivity C-reactive protein |
| Aili et al. (2018) [12] | Sleep problems—CWPa | Sweden | 18 | 791 (52.0) | 49.0 | > 3 | Question | Incident | 18 predefined regions, except head and abdomen | USI | Age, gender, socioeconomic, mental health and number of pain sites |
| Campanini et al. (2022) [47] | Bidirectional sleep quality LBP | Brazil | 2 | 530 (66.0) | 42.3 | ≥ 6 | Question | Prevalent and incident | Low back | PSQI | Age, sex, body mass index, physical activity, smoking, coffee intake, alcohol, self-rated health, depression and anxiety |
| García-Esquinas et al. (2019) [38] | Sleep quality—pain | Spain | 2.8 | 851 (43.1) | 71.7 | ≥ 6 | Survey on chronic pain in Europe | Incident | Neck, back, bones, joints, legs, arms and other sites | Question | Age and sex |
| Generaal et al. (2017) [41] | Insomnia—CMMP | Netherlands | 6 | 1860 (66.5) | 42.1 | ≥ 6 | CPG | Incident | Arms, hands, legs, feet, back and neck | IRS | Age, sex, education, body mass index, smoking, alcohol, physical activity, number of chronic diseases, pain intensity, sleep medication, pain medication, anxiety and depression |
| Glette et al. (2020) [43] | Sleep difficulties—CP | Norway | 4 | 1905 (61.2) | 56.6 | > 6 | Question | Persistent | Jaw/teeth, shoulder/arm, wrists/hands, elbows, calves, hips, thighs, knees, feet/ankles, chest, stomach, pelvis/genitalia, neck, low back, upper back and head | Question | Age and sex |
| Ho et al. (2022) [48] | Bidirectional Insomnia LBP | Norway | 11 | 11606 (73.1) | 55.6 | ≥ 3 | Question | Prevalent and Incident | Low back and lower limb | Questions | Age, sex, education, leisure time physical activity, body mass index, work, alcohol and smoking |
| Kääriä et al. (2012) [37] | Sleep problems-NP and LBP | Finland | 6 | 5277 (80.0) | 49.6 | ≥ 3 | Question | Incident | Neck and low back | JSQ | Age |
| Lindell and Grimby-Ekman (2022) [49] | Non-restorative sleep-CP | Sweden | 1 | 1567 (41.5) | 23.0 | > 3 | Question | Incident | Back, neck, or upper extremities | Sleep and wakefulness form | Gender and education |
| Mork et al. (2014) [14] | Sleep problems-CMP | Norway | 11 | 26896 (50.2) | 43.8 | ≥ 3 | SNQ | Incident | Neck/shoulders and low back | Question | Age, body mass index, physical exercise, psychological well-being, smoking and occupation |
| Mundal et al. (2014) [39] | Sleep problems-CWPa | Norway | 11 | 19192 (53.8) | 44.5 | ≥ 3 | SNQ | Incident | Neck, shoulder, elbow, hand/wrist, upper back, lower back, hip, knee, and ankle/foot | Questions | Age, sex, marital status, education, exercise, chronic pain and chronic disease at baseline |
| Mundal et al. (2016) [50] | Sleep problems-CMP | Norway | 11 | 50807 (53.0) | NI | ≥ 3 | SNQ | Incident | Neck, shoulder, elbow, hand/wrist, upper back, lower back, hip, knee, and ankle/foot | Questions | Age, sex, marital status, education, physical activity and chronic disease |
| Nitter et al. (2012) [40] | Sleep problems-CP | Norway | 17 | 1338 (100.0) | 34.4 | ≥ 3 | Questions | Incident | Muscles, joints and back | Questions | Age |
| Ødegård et al. (2013) [46] | CMSC-Insomnia | Norway | 11 | 19271 (54.7) | 45.8 | ≥ 3 | ARC | Prevalent | Neck, shoulders, elbows, wrist/hands, upper back, lower back, hips, knees, and/or ankles/feet | Questions | Age, gender, anxiety, depression and gastrointestinal complaints |
| Sit et al. (2021) [44] | CMP-Insomnia | China | 1.5 | 498 (73.1) | 69.1 | > 3 | ICD and BPI | Prevalent | Joints, limbs, back, and/or neck | ISI | Age, gender, body mass index and number of comorbid diseases |
| Skarpsno et al. (2018) [45] | CMP-Insomnia | Norway | 11 | 21847 (54.5) | 46.5 | ≥ 3 | Question | Prevalent | Neck, shoulders, elbows, wrists/hands, upper back, low back, hips, knees and ankles/feet | Questions | Age, leisure-time physical activity, body mass index, education, smoking, shift work, alcohol, depression and anxiety |
| Skarpsno et al. (2018) [21] | CMP-Insomnia | Norway | 11 | 16161 (52.9) | 42.8 | ≥ 3 | Question | Prevalent | Not described | DSM-V | Age, sex, education, physical work demands, body mass index, smoking and leisure-time physical activity |
| Skarpsno et al. (2020) [42] | Sleeplessness-LBP | Norway | 11 | 6200 (59.8) | 49.6 | ≥ 3 | SNQ | Persistent | Low back | Questions | Age, body mass index, leisure-time physical activity, education and smoking |
| Skarpsno et al. (2021) [36] | Sleep quality-CMP | Norway | 24 | 6033 (52.9) | 54.7 | ≥ 3 | SNQ | Incident | Neck, shoulders, upper back, elbows, low back, hips, wrists/hand, knees, and ankles/feet | Questions | Age, sex, education, body mass index, weight, leisure time physical activity and smoking |
| Uhlig et al. (2018) [13] | Insomnia-CMSC | Norway | 11 | 13429 (54.9) | 43.9 | ≥ 3 | ARC | Incident | Neck, shoulders, upper back, elbows, lower back, hands/wrists, hips, knees, ankles/foot | DSM-V | Age, gender, education, smoking, physical activity, anxiety, depression and body mass index |
| Author (year) | Association direction | Country | Follow-up time, mean, years | Women, n (%) | Mean age, years | Pain duration (months) | Measurement of CMP | Types of pain | Pain region | Measurement of insomnia | Adjustment variables |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Agmon and Armon (2014) [5] | Insomnia—BP | Israel | 3.7 | 2131 (34.0) | 46.2 | ≥ 3 | Medical record and interview | Incident | Back pain | AIS-5 | Gender, age, education, smoking, physical activity, self-rated health, body mass index and high-sensitivity C-reactive protein |
| Aili et al. (2018) [12] | Sleep problems—CWPa | Sweden | 18 | 791 (52.0) | 49.0 | > 3 | Question | Incident | 18 predefined regions, except head and abdomen | USI | Age, gender, socioeconomic, mental health and number of pain sites |
| Campanini et al. (2022) [47] | Bidirectional sleep quality LBP | Brazil | 2 | 530 (66.0) | 42.3 | ≥ 6 | Question | Prevalent and incident | Low back | PSQI | Age, sex, body mass index, physical activity, smoking, coffee intake, alcohol, self-rated health, depression and anxiety |
| García-Esquinas et al. (2019) [38] | Sleep quality—pain | Spain | 2.8 | 851 (43.1) | 71.7 | ≥ 6 | Survey on chronic pain in Europe | Incident | Neck, back, bones, joints, legs, arms and other sites | Question | Age and sex |
| Generaal et al. (2017) [41] | Insomnia—CMMP | Netherlands | 6 | 1860 (66.5) | 42.1 | ≥ 6 | CPG | Incident | Arms, hands, legs, feet, back and neck | IRS | Age, sex, education, body mass index, smoking, alcohol, physical activity, number of chronic diseases, pain intensity, sleep medication, pain medication, anxiety and depression |
| Glette et al. (2020) [43] | Sleep difficulties—CP | Norway | 4 | 1905 (61.2) | 56.6 | > 6 | Question | Persistent | Jaw/teeth, shoulder/arm, wrists/hands, elbows, calves, hips, thighs, knees, feet/ankles, chest, stomach, pelvis/genitalia, neck, low back, upper back and head | Question | Age and sex |
| Ho et al. (2022) [48] | Bidirectional Insomnia LBP | Norway | 11 | 11606 (73.1) | 55.6 | ≥ 3 | Question | Prevalent and Incident | Low back and lower limb | Questions | Age, sex, education, leisure time physical activity, body mass index, work, alcohol and smoking |
| Kääriä et al. (2012) [37] | Sleep problems-NP and LBP | Finland | 6 | 5277 (80.0) | 49.6 | ≥ 3 | Question | Incident | Neck and low back | JSQ | Age |
| Lindell and Grimby-Ekman (2022) [49] | Non-restorative sleep-CP | Sweden | 1 | 1567 (41.5) | 23.0 | > 3 | Question | Incident | Back, neck, or upper extremities | Sleep and wakefulness form | Gender and education |
| Mork et al. (2014) [14] | Sleep problems-CMP | Norway | 11 | 26896 (50.2) | 43.8 | ≥ 3 | SNQ | Incident | Neck/shoulders and low back | Question | Age, body mass index, physical exercise, psychological well-being, smoking and occupation |
| Mundal et al. (2014) [39] | Sleep problems-CWPa | Norway | 11 | 19192 (53.8) | 44.5 | ≥ 3 | SNQ | Incident | Neck, shoulder, elbow, hand/wrist, upper back, lower back, hip, knee, and ankle/foot | Questions | Age, sex, marital status, education, exercise, chronic pain and chronic disease at baseline |
| Mundal et al. (2016) [50] | Sleep problems-CMP | Norway | 11 | 50807 (53.0) | NI | ≥ 3 | SNQ | Incident | Neck, shoulder, elbow, hand/wrist, upper back, lower back, hip, knee, and ankle/foot | Questions | Age, sex, marital status, education, physical activity and chronic disease |
| Nitter et al. (2012) [40] | Sleep problems-CP | Norway | 17 | 1338 (100.0) | 34.4 | ≥ 3 | Questions | Incident | Muscles, joints and back | Questions | Age |
| Ødegård et al. (2013) [46] | CMSC-Insomnia | Norway | 11 | 19271 (54.7) | 45.8 | ≥ 3 | ARC | Prevalent | Neck, shoulders, elbows, wrist/hands, upper back, lower back, hips, knees, and/or ankles/feet | Questions | Age, gender, anxiety, depression and gastrointestinal complaints |
| Sit et al. (2021) [44] | CMP-Insomnia | China | 1.5 | 498 (73.1) | 69.1 | > 3 | ICD and BPI | Prevalent | Joints, limbs, back, and/or neck | ISI | Age, gender, body mass index and number of comorbid diseases |
| Skarpsno et al. (2018) [45] | CMP-Insomnia | Norway | 11 | 21847 (54.5) | 46.5 | ≥ 3 | Question | Prevalent | Neck, shoulders, elbows, wrists/hands, upper back, low back, hips, knees and ankles/feet | Questions | Age, leisure-time physical activity, body mass index, education, smoking, shift work, alcohol, depression and anxiety |
| Skarpsno et al. (2018) [21] | CMP-Insomnia | Norway | 11 | 16161 (52.9) | 42.8 | ≥ 3 | Question | Prevalent | Not described | DSM-V | Age, sex, education, physical work demands, body mass index, smoking and leisure-time physical activity |
| Skarpsno et al. (2020) [42] | Sleeplessness-LBP | Norway | 11 | 6200 (59.8) | 49.6 | ≥ 3 | SNQ | Persistent | Low back | Questions | Age, body mass index, leisure-time physical activity, education and smoking |
| Skarpsno et al. (2021) [36] | Sleep quality-CMP | Norway | 24 | 6033 (52.9) | 54.7 | ≥ 3 | SNQ | Incident | Neck, shoulders, upper back, elbows, low back, hips, wrists/hand, knees, and ankles/feet | Questions | Age, sex, education, body mass index, weight, leisure time physical activity and smoking |
| Uhlig et al. (2018) [13] | Insomnia-CMSC | Norway | 11 | 13429 (54.9) | 43.9 | ≥ 3 | ARC | Incident | Neck, shoulders, upper back, elbows, lower back, hands/wrists, hips, knees, ankles/foot | DSM-V | Age, gender, education, smoking, physical activity, anxiety, depression and body mass index |
Only musculoskeletal regions of the body were considered in this systematic review. AIS: Athens Insomnia Scale; ARC: American College of Rheumatology; BP: back pain; BPI: Brief Pain Inventory; CMMP: chronic multisite musculoskeletal pain; CMP: chronic musculoskeletal pain; CMSC: chronic musculoskeletal complaints; CP: chronic pain; CPG: chronic pain grade; CWP: chronic widespread pain; DSM-V: Diagnostic and Statistical Manual of Mental Disorders, 5th edition; ICD: International Classification of Disease; IRS: Initiative Insomnia Rating Scale; ISI: Insomnia Severity Index; JSQ: Jenkins Sleep Questionnaire; LBP: low back pain; NI: no information; NP: neck pain; PSQI: Pittsburgh Sleep Quality Index; SNQ: Standardized Nordic Questionnaire; USI: Uppsala Sleep Inventory.
Characteristics of the included studies
| Author (year) | Association direction | Country | Follow-up time, mean, years | Women, n (%) | Mean age, years | Pain duration (months) | Measurement of CMP | Types of pain | Pain region | Measurement of insomnia | Adjustment variables |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Agmon and Armon (2014) [5] | Insomnia—BP | Israel | 3.7 | 2131 (34.0) | 46.2 | ≥ 3 | Medical record and interview | Incident | Back pain | AIS-5 | Gender, age, education, smoking, physical activity, self-rated health, body mass index and high-sensitivity C-reactive protein |
| Aili et al. (2018) [12] | Sleep problems—CWPa | Sweden | 18 | 791 (52.0) | 49.0 | > 3 | Question | Incident | 18 predefined regions, except head and abdomen | USI | Age, gender, socioeconomic, mental health and number of pain sites |
| Campanini et al. (2022) [47] | Bidirectional sleep quality LBP | Brazil | 2 | 530 (66.0) | 42.3 | ≥ 6 | Question | Prevalent and incident | Low back | PSQI | Age, sex, body mass index, physical activity, smoking, coffee intake, alcohol, self-rated health, depression and anxiety |
| García-Esquinas et al. (2019) [38] | Sleep quality—pain | Spain | 2.8 | 851 (43.1) | 71.7 | ≥ 6 | Survey on chronic pain in Europe | Incident | Neck, back, bones, joints, legs, arms and other sites | Question | Age and sex |
| Generaal et al. (2017) [41] | Insomnia—CMMP | Netherlands | 6 | 1860 (66.5) | 42.1 | ≥ 6 | CPG | Incident | Arms, hands, legs, feet, back and neck | IRS | Age, sex, education, body mass index, smoking, alcohol, physical activity, number of chronic diseases, pain intensity, sleep medication, pain medication, anxiety and depression |
| Glette et al. (2020) [43] | Sleep difficulties—CP | Norway | 4 | 1905 (61.2) | 56.6 | > 6 | Question | Persistent | Jaw/teeth, shoulder/arm, wrists/hands, elbows, calves, hips, thighs, knees, feet/ankles, chest, stomach, pelvis/genitalia, neck, low back, upper back and head | Question | Age and sex |
| Ho et al. (2022) [48] | Bidirectional Insomnia LBP | Norway | 11 | 11606 (73.1) | 55.6 | ≥ 3 | Question | Prevalent and Incident | Low back and lower limb | Questions | Age, sex, education, leisure time physical activity, body mass index, work, alcohol and smoking |
| Kääriä et al. (2012) [37] | Sleep problems-NP and LBP | Finland | 6 | 5277 (80.0) | 49.6 | ≥ 3 | Question | Incident | Neck and low back | JSQ | Age |
| Lindell and Grimby-Ekman (2022) [49] | Non-restorative sleep-CP | Sweden | 1 | 1567 (41.5) | 23.0 | > 3 | Question | Incident | Back, neck, or upper extremities | Sleep and wakefulness form | Gender and education |
| Mork et al. (2014) [14] | Sleep problems-CMP | Norway | 11 | 26896 (50.2) | 43.8 | ≥ 3 | SNQ | Incident | Neck/shoulders and low back | Question | Age, body mass index, physical exercise, psychological well-being, smoking and occupation |
| Mundal et al. (2014) [39] | Sleep problems-CWPa | Norway | 11 | 19192 (53.8) | 44.5 | ≥ 3 | SNQ | Incident | Neck, shoulder, elbow, hand/wrist, upper back, lower back, hip, knee, and ankle/foot | Questions | Age, sex, marital status, education, exercise, chronic pain and chronic disease at baseline |
| Mundal et al. (2016) [50] | Sleep problems-CMP | Norway | 11 | 50807 (53.0) | NI | ≥ 3 | SNQ | Incident | Neck, shoulder, elbow, hand/wrist, upper back, lower back, hip, knee, and ankle/foot | Questions | Age, sex, marital status, education, physical activity and chronic disease |
| Nitter et al. (2012) [40] | Sleep problems-CP | Norway | 17 | 1338 (100.0) | 34.4 | ≥ 3 | Questions | Incident | Muscles, joints and back | Questions | Age |
| Ødegård et al. (2013) [46] | CMSC-Insomnia | Norway | 11 | 19271 (54.7) | 45.8 | ≥ 3 | ARC | Prevalent | Neck, shoulders, elbows, wrist/hands, upper back, lower back, hips, knees, and/or ankles/feet | Questions | Age, gender, anxiety, depression and gastrointestinal complaints |
| Sit et al. (2021) [44] | CMP-Insomnia | China | 1.5 | 498 (73.1) | 69.1 | > 3 | ICD and BPI | Prevalent | Joints, limbs, back, and/or neck | ISI | Age, gender, body mass index and number of comorbid diseases |
| Skarpsno et al. (2018) [45] | CMP-Insomnia | Norway | 11 | 21847 (54.5) | 46.5 | ≥ 3 | Question | Prevalent | Neck, shoulders, elbows, wrists/hands, upper back, low back, hips, knees and ankles/feet | Questions | Age, leisure-time physical activity, body mass index, education, smoking, shift work, alcohol, depression and anxiety |
| Skarpsno et al. (2018) [21] | CMP-Insomnia | Norway | 11 | 16161 (52.9) | 42.8 | ≥ 3 | Question | Prevalent | Not described | DSM-V | Age, sex, education, physical work demands, body mass index, smoking and leisure-time physical activity |
| Skarpsno et al. (2020) [42] | Sleeplessness-LBP | Norway | 11 | 6200 (59.8) | 49.6 | ≥ 3 | SNQ | Persistent | Low back | Questions | Age, body mass index, leisure-time physical activity, education and smoking |
| Skarpsno et al. (2021) [36] | Sleep quality-CMP | Norway | 24 | 6033 (52.9) | 54.7 | ≥ 3 | SNQ | Incident | Neck, shoulders, upper back, elbows, low back, hips, wrists/hand, knees, and ankles/feet | Questions | Age, sex, education, body mass index, weight, leisure time physical activity and smoking |
| Uhlig et al. (2018) [13] | Insomnia-CMSC | Norway | 11 | 13429 (54.9) | 43.9 | ≥ 3 | ARC | Incident | Neck, shoulders, upper back, elbows, lower back, hands/wrists, hips, knees, ankles/foot | DSM-V | Age, gender, education, smoking, physical activity, anxiety, depression and body mass index |
| Author (year) | Association direction | Country | Follow-up time, mean, years | Women, n (%) | Mean age, years | Pain duration (months) | Measurement of CMP | Types of pain | Pain region | Measurement of insomnia | Adjustment variables |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Agmon and Armon (2014) [5] | Insomnia—BP | Israel | 3.7 | 2131 (34.0) | 46.2 | ≥ 3 | Medical record and interview | Incident | Back pain | AIS-5 | Gender, age, education, smoking, physical activity, self-rated health, body mass index and high-sensitivity C-reactive protein |
| Aili et al. (2018) [12] | Sleep problems—CWPa | Sweden | 18 | 791 (52.0) | 49.0 | > 3 | Question | Incident | 18 predefined regions, except head and abdomen | USI | Age, gender, socioeconomic, mental health and number of pain sites |
| Campanini et al. (2022) [47] | Bidirectional sleep quality LBP | Brazil | 2 | 530 (66.0) | 42.3 | ≥ 6 | Question | Prevalent and incident | Low back | PSQI | Age, sex, body mass index, physical activity, smoking, coffee intake, alcohol, self-rated health, depression and anxiety |
| García-Esquinas et al. (2019) [38] | Sleep quality—pain | Spain | 2.8 | 851 (43.1) | 71.7 | ≥ 6 | Survey on chronic pain in Europe | Incident | Neck, back, bones, joints, legs, arms and other sites | Question | Age and sex |
| Generaal et al. (2017) [41] | Insomnia—CMMP | Netherlands | 6 | 1860 (66.5) | 42.1 | ≥ 6 | CPG | Incident | Arms, hands, legs, feet, back and neck | IRS | Age, sex, education, body mass index, smoking, alcohol, physical activity, number of chronic diseases, pain intensity, sleep medication, pain medication, anxiety and depression |
| Glette et al. (2020) [43] | Sleep difficulties—CP | Norway | 4 | 1905 (61.2) | 56.6 | > 6 | Question | Persistent | Jaw/teeth, shoulder/arm, wrists/hands, elbows, calves, hips, thighs, knees, feet/ankles, chest, stomach, pelvis/genitalia, neck, low back, upper back and head | Question | Age and sex |
| Ho et al. (2022) [48] | Bidirectional Insomnia LBP | Norway | 11 | 11606 (73.1) | 55.6 | ≥ 3 | Question | Prevalent and Incident | Low back and lower limb | Questions | Age, sex, education, leisure time physical activity, body mass index, work, alcohol and smoking |
| Kääriä et al. (2012) [37] | Sleep problems-NP and LBP | Finland | 6 | 5277 (80.0) | 49.6 | ≥ 3 | Question | Incident | Neck and low back | JSQ | Age |
| Lindell and Grimby-Ekman (2022) [49] | Non-restorative sleep-CP | Sweden | 1 | 1567 (41.5) | 23.0 | > 3 | Question | Incident | Back, neck, or upper extremities | Sleep and wakefulness form | Gender and education |
| Mork et al. (2014) [14] | Sleep problems-CMP | Norway | 11 | 26896 (50.2) | 43.8 | ≥ 3 | SNQ | Incident | Neck/shoulders and low back | Question | Age, body mass index, physical exercise, psychological well-being, smoking and occupation |
| Mundal et al. (2014) [39] | Sleep problems-CWPa | Norway | 11 | 19192 (53.8) | 44.5 | ≥ 3 | SNQ | Incident | Neck, shoulder, elbow, hand/wrist, upper back, lower back, hip, knee, and ankle/foot | Questions | Age, sex, marital status, education, exercise, chronic pain and chronic disease at baseline |
| Mundal et al. (2016) [50] | Sleep problems-CMP | Norway | 11 | 50807 (53.0) | NI | ≥ 3 | SNQ | Incident | Neck, shoulder, elbow, hand/wrist, upper back, lower back, hip, knee, and ankle/foot | Questions | Age, sex, marital status, education, physical activity and chronic disease |
| Nitter et al. (2012) [40] | Sleep problems-CP | Norway | 17 | 1338 (100.0) | 34.4 | ≥ 3 | Questions | Incident | Muscles, joints and back | Questions | Age |
| Ødegård et al. (2013) [46] | CMSC-Insomnia | Norway | 11 | 19271 (54.7) | 45.8 | ≥ 3 | ARC | Prevalent | Neck, shoulders, elbows, wrist/hands, upper back, lower back, hips, knees, and/or ankles/feet | Questions | Age, gender, anxiety, depression and gastrointestinal complaints |
| Sit et al. (2021) [44] | CMP-Insomnia | China | 1.5 | 498 (73.1) | 69.1 | > 3 | ICD and BPI | Prevalent | Joints, limbs, back, and/or neck | ISI | Age, gender, body mass index and number of comorbid diseases |
| Skarpsno et al. (2018) [45] | CMP-Insomnia | Norway | 11 | 21847 (54.5) | 46.5 | ≥ 3 | Question | Prevalent | Neck, shoulders, elbows, wrists/hands, upper back, low back, hips, knees and ankles/feet | Questions | Age, leisure-time physical activity, body mass index, education, smoking, shift work, alcohol, depression and anxiety |
| Skarpsno et al. (2018) [21] | CMP-Insomnia | Norway | 11 | 16161 (52.9) | 42.8 | ≥ 3 | Question | Prevalent | Not described | DSM-V | Age, sex, education, physical work demands, body mass index, smoking and leisure-time physical activity |
| Skarpsno et al. (2020) [42] | Sleeplessness-LBP | Norway | 11 | 6200 (59.8) | 49.6 | ≥ 3 | SNQ | Persistent | Low back | Questions | Age, body mass index, leisure-time physical activity, education and smoking |
| Skarpsno et al. (2021) [36] | Sleep quality-CMP | Norway | 24 | 6033 (52.9) | 54.7 | ≥ 3 | SNQ | Incident | Neck, shoulders, upper back, elbows, low back, hips, wrists/hand, knees, and ankles/feet | Questions | Age, sex, education, body mass index, weight, leisure time physical activity and smoking |
| Uhlig et al. (2018) [13] | Insomnia-CMSC | Norway | 11 | 13429 (54.9) | 43.9 | ≥ 3 | ARC | Incident | Neck, shoulders, upper back, elbows, lower back, hands/wrists, hips, knees, ankles/foot | DSM-V | Age, gender, education, smoking, physical activity, anxiety, depression and body mass index |
Only musculoskeletal regions of the body were considered in this systematic review. AIS: Athens Insomnia Scale; ARC: American College of Rheumatology; BP: back pain; BPI: Brief Pain Inventory; CMMP: chronic multisite musculoskeletal pain; CMP: chronic musculoskeletal pain; CMSC: chronic musculoskeletal complaints; CP: chronic pain; CPG: chronic pain grade; CWP: chronic widespread pain; DSM-V: Diagnostic and Statistical Manual of Mental Disorders, 5th edition; ICD: International Classification of Disease; IRS: Initiative Insomnia Rating Scale; ISI: Insomnia Severity Index; JSQ: Jenkins Sleep Questionnaire; LBP: low back pain; NI: no information; NP: neck pain; PSQI: Pittsburgh Sleep Quality Index; SNQ: Standardized Nordic Questionnaire; USI: Uppsala Sleep Inventory.
This systematic review covered 208 190 participants with a mean age range between 34.4 [40] and 71.7 [38] years. The follow-up time of the studies ranged from two [47] to 24 [36] years and were conducted in eight different countries. Only five [5, 38, 41, 44, 47] of the 20 studies included in the review were not carried out in Nordic countries (i.e. Finland, Norway and Sweden). Among the analysed articles, 80% considered chronic pain as pain that persists for more than 3 months, and the regions frequently reported as musculoskeletal were the neck, shoulder, arms, back, knee, legs and feet. The SRP variable was subjectively assessed in all the included studies. Adjustment variables comprised sociodemographic, health and lifestyle factors (Table 1).
As shown in Fig. 2, participants with SRP at baseline had a higher risk of developing CMP (OR = 1.79; 95% CI: 1.55, 2.08; I2 = 84.7%) and having persistent CMP (OR = 2.04; 95% CI: 1.42, 2.94) (Supplementary Fig. S1, available at Rheumatology online) than individuals without SRP. Furthermore, bidirectionally, participants with CMP at baseline had a higher risk of developing SRP (OR = 2.02; 95% CI: 1.62, 2.53; I2 = 90.0%) than individuals without CMP (Fig. 3).
Meta-analysis of incident CMP in subjects with sleep-related problems compared with those without sleep-related problems. CMP: chronic musculoskeletal pain
Meta-analysis of incident sleep-related problems in subjects with CMP compared with those without CMP. CMP: chronic musculoskeletal pain
The pooled estimates for the prospective subgroup analyses between SRP and the risk of developing CMP were similar to the main pooled OR (Fig. 4). Additionally, there was no statistical significance in any meta-regression analysis considering the variables follow-up time (P = 0.809), sex ratio (P = 0.572) and age (P = 0.132), as observed in Supplementary Fig. S2, available at Rheumatology online.
Meta-analysis of incident CMP according to subgroups of subjects with sleep-related problems. CMP: chronic musculoskeletal pain
Bidirectionally, sensitivity analyses showed that the pooled OR was not modified when removing each included study one by one (Supplementary Tables S1 and S2, available at Rheumatology online). Likewise, no relevant change in the pooled OR was observed after removing studies of fair quality and high risk of bias (Supplementary Figs S3 and S4, available at Rheumatology online). Finally, according to Egger’s test and funnel plot asymmetry, the results suggest that there is no significant publication bias regarding the incidence of pain (P = 0.199) (Supplementary Fig. S5, available at Rheumatology online) and SRP (P = 0.776) (Supplementary Fig. S6, available at Rheumatology online) in the meta-analysis of interest.
Discussion
Our systematic review and meta-analysis results support the notion that exposure to SRP increases the incidence of CMP in adults. Longitudinal epidemiological studies were used to understand the chronology between SRP and musculoskeletal painful symptoms, highlighting the predictive role that exposure to SRP has on the incidence of CMP. Although most of the reviewed studies support this specific relationship direction, i.e. SRP leading to the incidence of CMP, additional analyses indicate that reverse causality exists, in which CMP plays an important role in the incidence of SRP.
The data reveal that exposure to SRP increased the odds of developing CMP by 79%. This result agrees with the findings from a longitudinal study that indicated that insomnia symptoms increased the odds of chronic back pain by 40.0% in adults [5]. Consistently, a study that considered indicators of better sleep quality as exposures revealed that rapid sleep onset, absence of early awakening, and restorative sleep were associated with pain resolution [51]. Scientific evidence indicates that insomnia is related to the increase in basal inflammation measured by C-reactive protein, and this process is associated with chronic pain symptoms, specifically in the back region [52] and musculoskeletal grouping [20].
Regarding the association between SRP and pain, studies show that there is substantial overlap in dopaminergic abnormalities for both variables [53, 54], and such abnormalities might be able to act as a mediator of this association. In addition, individuals with insomnia symptoms tend to perceive life as stressful, which generates tension of prolonged duration [55] that can affect the musculoskeletal system and ultimately lead to pain [56, 57]. The stress generated by this symptomatology is also related to the activation of the sympathetic nervous system, which increases the risk of inflammatory processes [58, 59] and triggers an excessive release of cytokines and mediators [60, 61], strengthening the evidence that excitation of the sympathetic nervous system is related to SRP [62]. Moreover, this sympathetic activity favours the secretion of noradrenaline, which is responsible for increasing muscle tone and the risk of pain and injuries of muscle origin [57].
The present study also expands the body of evidence on the bidirectional relationship between CMP and SRP. The odds of developing SRP were 102% higher in adults with CMP than in those without painful symptoms. Corroborating a review that addresses the concomitance of both variables, highlighting this bidirectional and reciprocal relationship [15], individuals with chronic pain are predisposed to worsening SRP [16], explained by the central sensitization mechanism, which occurs by means of neuroinflammation [63].
To understand the bidirectional relationship between SRP and CMP, it is crucial to consider the potential mediating role of mental health on these associations. Individuals experiencing depression or anxiety, for instance, may be more prone to perceive pain as more intense or bothersome, even though there is no increase in the actual physical sensation of pain. Specifically, depressive symptoms [41] and other mood indicators [64] were found to mediate the relationship between sleep disturbances and chronic musculoskeletal pain. This is based on the suggested pathway by which poor sleep worsens mood, which, as a result of the activation of the central sensitization process and physiological arousal over time [65], induces a decrease in pain threshold and thus increases pain perception. On the other hand, a systematic review found that anxiety plays an intermediate role between impaired sleep and worsening pain symptoms [66]. In general, the relationship between sleep, mental health and pain is complex and likely bidirectional, with each factor influencing and being influenced by the others. Further investigation is required to thoroughly examine the mediating role of mental health in the SRP–CMP relationship to support the development of interventions aimed at improving sleep and reducing musculoskeletal pain in individuals experiencing these problems.
Some methodological considerations should be taken into account for the correct interpretation of these results. First, the information for SRP was self-reported in all the analysed studies. Although polysomnographic methods are the gold standard of objective measures to evaluate SRP [67], questionnaires are commonly used in epidemiological studies because of their practicality [12, 13, 42]. Furthermore, the sleep-related issues covered in this review were obtained through the use of single questions or questionnaires that provided subjective indicators of sleep quality. Additionally, several of the included cohort studies had their baseline before 2010, that is, before the DSM-V updated the diagnostic criteria for sleep disorders [68]. Therefore, although some of these indicators are often used to characterize insomnia, our findings do not allow us to conclude that insomnia, or any other sleep disorder in particular, is associated with CMP. Third, pain chronicity was not defined uniformly in the analysed studies. Part of the divergence in the standardization of the cut-off point is due to the update of the definition of chronic pain in 2019 by the International Association for the Study of Pain [29], when most of the articles included in this study had already been published. However, this research analysed a specific condition that lasted 3 months or more and in regions involving bones, joints, muscles or surrounding structures to capture the experience lived by a large part of the adult population [69–71]. Fourth, four of the 11 studies included in the meta-analysis of SRP and incidence of CMP are part of the same cohort [13, 14, 36, 39], which may over-represent the sample. The same problem was observed in the analysis of CMP as an exposure variable, as three articles were derived from the same cohort [21, 45, 46]. However, similar results were observed even after sensitivity analysis. Finally, high heterogeneity was observed in the meta-analysis of the incidence of CMP in subjects exposed to SRP; however, we explored most of the results in subgroup analyses to understand the heterogeneity found, partially justified by the follow-up time.
In conclusion, this study provides robust evidence for the predictive role that exposure to SRP has on the incidence of CMP in adults. Furthermore, while some studies indicate a two-way relationship between SRP and CMP [47, 52, 72], additional long-term prospective studies are necessary to corroborate that CMP can lead to SRP in adults. The current findings underscore the importance of a comprehensive, multidisciplinary and sustained approach for preventing and managing both CMP and SRP. In terms of public and occupational health, this implies, for instance, strengthening health education on ergonomics and sleep hygiene, especially for those with musculoskeletal pain or sleep issues and those at higher risk due to factors such as occupation, body composition or mental health status. In clinical settings, health professionals managing one of these conditions (CMP or SRP) should be aware of the increased risk of developing the other condition, as well as the heightened impact on physical and mental health when both are present, compared with either alone. Long-term clinical trials particularly designed for this purpose are needed, therefore, to evaluate the potential effectiveness of these interventions in the prevention and management of pain, sleep-related problems and mental wellbeing. In addition, considering that substantial unexplained between-study heterogeneity was detected in this review, it is recommended that future observational and experimental studies provide more detailed data on the study population (e.g. occupation, level of physical activity and presence of comorbidities, especially mental health disorders such as depression and anxiety), CMP (e.g. location, duration, intensity and use of pain medication) and SRPs (e.g. type of sleep complaint, frequency, and impact on daytime function and quality of life).
Supplementary data
Supplementary data are available at Rheumatology online.
Data availability
The data underlying this article are available in the article and in its online supplementary material.
Funding
M.S. receives a scholarship from the Coordination for the Improvement of Higher Level Personnel, Ministry of Education, Brazil (88887.480548/2020–00). B.B.-P. receives a scholarship from the University of Castilla-La Mancha, Spain, co-financed by the European Social Fund (2020-PREDUCLM-16746). A.E.M. is supported by a ‘Beatriz Galindo’ contract (BEAGAL18/00093) from the Spanish Ministry of Education, Culture and Sport.
Disclosure statement: The authors have declared no conflicts of interest.
Comments