Abstract
BACKGROUND: Sleep health is a critical determinant of older adults' physical, cognitive, and emotional well-being, yet pharmacological treatments for sleep disturbances carry substantial risks. OBJECTIVE: This umbrella review aims to synthesize evidence on the effectiveness of non-pharmacological interventions (NPIs) for improving sleep in older adults, to inform clinical decision-making and future guidelines. METHODS: This review adhered to PRIOR and PRISMA guidelines and was registered in PROSPERO (CRD42024565849). Systematic searches were conducted across six databases from inception to July 6, 2024. Eligible studies were systematic reviews with meta-analyses of randomized controlled trials (RCTs) targeting adults aged ≥60 years. Two reviewers independently screened studies, extracted data, and assessed methodological quality using AMSTAR 2. Certainty of evidence was rated with GRADE, and review overlap was quantified using the Corrected Covered Area (CCA). Narrative synthesis was conducted due to high heterogeneity. RESULTS: Nineteen systematic reviews comprising 160 RCTs were included. Interventions covered six categories: cognitive behavioral therapy (CBT), mindfulness, exercise (e.g., Tai Chi), music, manual therapies (e.g., massage, acupuncture), and joint approaches. CBT significantly improved sleep onset latency (-9.29 min), wake after sleep onset (-22 min), and sleep efficiency (+7.9%). Exercise, particularly Tai Chi, reduced PSQI global scores by -1.05. Music and manual therapies also showed benefits, though with inconsistent effect sizes. Most reviews were of low methodological quality, and the certainty of evidence ranged from low to very low. CCA was 3.68%, indicating slight overlap. CONCLUSION: CBT and exercise-based interventions are promising for improving sleep in older adults. However, the certainty of evidence remains limited. Future high-quality RCTs are needed, and the evidence map highlights priority areas for research in geriatric sleep health. SYSTEMATIC REVIEW REGISTRATION: https://www.crd.york.ac.uk/PROSPERO/, identifier CRD42024565849.