Abstract / Summary
Background and Objective The bidirectional pain–sleep relationship is well established, but whether pain from specific anatomical regions differentially influences sleep disturbance remains unclear. This meta-analysis quantified the pain–sleep correlation in musculoskeletal disorders (MSD).Methods PubMed, Scopus, and Web of Science were searched for observational studies of adults with MSD reporting pain–sleep or disability–sleep correlations. Quality was rated with the adapted Newcastle–Ottawa scale. Unadjusted correlations were transformed to Fisher’s z and pooled using a three-level correlated-and-hierarchical-effects model with cluster-robust variance estimation, with pre-specified subgroup and sensitivity analyses. Twenty-two studies (4,133 participants) contributed 25 pain–sleep and 12 disability–sleep effect sizes.Results Pain severity and sleep disturbance showed a medium correlation (r=0.32, 95% confidence interval [CI] 0.25 to 0.39). Neck (r=0.35) and low back (r=0.36) estimates were comparable, while the temporomandibular joint (TMJ) estimate included zero (r=0.23, 95% CI -0.01 to 0.44); no regional difference was detectable. Restricting to sleep-quality instruments removed this gradient (r= 0.35–0.37). Heterogeneity was mainly within studies (I²=79.3%; 53.5% withinstudy). Functional disability was similarly correlated with sleep disturbance (r=0.38, 95% CI 0.19 to 0.55), falling to r=0.32 (0.19 to 0.44) without the most influential study; 10 of the 12 disability estimates used instruments containing a sleep item.Conclusions Pain severity and functional disability each show a medium correlation with sleep disturbance in MSD. Heterogeneity reflects differences between sleep instruments rather than populations, and the weaker TMJ association reflects the sleep instrument used. The evidence does not support conclusions about regional or pain-duration differences.