Effects of exercise interventions on subjective sleep quality in older adults: a systematic review and meta-analysis of studies using the Pittsburgh sleep quality index

OBJECTIVE: In accordance with the 2024 Physical Activity Guidelines, this study aimed to evaluate the effects of exercise interventions on subjective sleep quality in older adults and to explore the potential dose-response relationship. METHODS: A systematic search was conducted in Web of Science, PubMed, Cochrane Library, Scopus, and Embase for randomized controlled trials (RCTs) published up to May 1, 2025. Meta-analysis was performed using R, with standardized mean differences (SMD) and 95% confidence intervals (95% CI) used to quantify effect sizes. Subgroup analysis, meta-regression, and nonlinear dose-response modeling were conducted. RESULTS: A total of 26 RCTs involving 2,189 elderly participants were included. The meta-analysis revealed that exercise interventions significantly improved subjective sleep quality [SMD = -2.46, 95% CI (-2.99, -1.93), p < 0.001]. The most pronounced effects were observed in interventions with session durations ≤30 min [WMD = -4.25, 95% CI (-5.49, -3.02)], low intensity [WMD = -2.79, 95% CI (-3.44, -2.14)], twice-weekly frequency [WMD = -2.52, 95% CI (-3.00, -2.04)], and intervention durations ≤8 weeks [WMD = -2.45, 95% CI (-2.99, -1.91)]. Meta-regression showed no significant linear associations between sleep outcomes and intervention duration, intensity, frequency, or length. A nonlinear "U-shaped" dose-response relationship was identified, with the optimal effect observed at approximately 527 MET·min/week [Hedges' g = -0.82, 95% CI (-1.12, -0.52)]. CONCLUSION: Low-frequency, short-duration, and low-to-moderate intensity exercise interventions can effectively improve subjective sleep quality in older adults. Notably, even low-dose exercise can yield significant benefits.