Prevalence and associated factors of poor sleep quality, restless legs syndrome, insomnia, and sleep apnea in hemodialysis patients: a systematic review and meta-analysis

The primary objective of this systematic review is to synthesize the factors associated with poor sleep quality (PSQ), restless legs syndrome (RLS), insomnia, and sleep apnea (SA) in patients undergoing hemodialysis (HD), and to subsequently estimate their prevalence. Seven databases were systematically searched for observational studies published up to May 2026. Study quality was assessed using the Agency for Healthcare Research and Quality (AHRQ) checklist and the Newcastle-Ottawa Scale (NOS). Random-effects meta-analyses were conducted to pool multivariable-adjusted associations, with prevalence summarized as a secondary outcome when available. The certainty of evidence was evaluated using the GRADE approach. Eighty-six studies involving 21,119 patients were included. Pooled prevalence was 60.6% for PSQ (95% CI 56.2%–64.9%; 95% PI 29.2%–87.7%), 23.1% for RLS (95% CI 18.3%–28.4%; 95% PI 4.1%–51.8%), 42.7% for insomnia (95% CI 39.8%–45.7%; 95% PI 34.7%–51.0%), and 47.8% for SA (95% CI 33.1%–62.8%; 95% PI 1.3%–97.6%). Older age, female sex, lower education, diabetes, and depression were associated with higher odds of PSQ, whereas higher serum albumin was inversely associated. Anxiety showed a borderline, non-significant association with PSQ. For continuous PSQI scores, older age was positively associated with greater sleep-disturbance severity, whereas hemoglobin was inversely associated. For RLS, higher serum phosphorus, longer dialysis vintage, higher iPTH, and higher exercise frequency showed borderline but non-significant associations. Age was not significantly associated with insomnia, and age, sex, and hypertension were not significantly associated with SA. Several demographic, clinical, psychological, and biochemical factors were associated with PSQ, while several factors showed borderline but non-significant associations with RLS. No significant associations were identified for insomnia or SA. The certainty of evidence was very low, and the secondary prevalence estimates indicated a substantial burden of sleep-related conditions. PROSPERO CRD420251241385.