Abstract Background The perinatal period spans from the onset of pregnancy to one year postpartum. Sleep quality is particularly critical during this period. An appropriate exercise regimen can significantly improve their sleep quality. Previous studies have demonstrated that exercise effectively enhances sleep quality; however, the optimal regimen remains unclear. To address this limitation, this study employs a network meta-analysis to systematically evaluate the effects of different types of exercise and dosage (including duration, frequency, and session length) on the sleep quality of perinatal women. This analysis provides evidence-based recommendations to improve sleep quality in perinatal women. Methods Using a combination of manual and computer-assisted search strategies, we searched PubMed, Web of Science, EBSCO, Cochrane Library, Scopus, Embase, ProQuest, CNKI, and Wanfang databases to identify randomized controlled trials on the effects of different exercise interventions on the sleep quality of perinatal women. The search was conducted up to August 5, 2025.The risk of bias was assessed using RevMan 5.4.1 software, and traditional meta-analysis and network meta-analysis were conducted using Stata 16.0 software to generate forest plots, network evidence plots, funnel plots, Surface Under the Cumulative Ranking Curve (SUCRA) plots, and GRADE evaluation plots. Results were reported using standardized mean difference (SMD) and 95% confidence intervals (CI). Results This study included 23 RCTs with 1,862 participants. Five different types of exercise were compared: Aerobic Exercise (AE), Aerobic Combined With Resistance Training (AE + RT), Yoga, Pilates, and Relaxation Exercises(RE). The highest-ranked intervention was RE SMD = -2.71, 95% CI (-3.96, -1.45), SUCRA = 88.5. Four intervention durations (≤ 4 weeks, 5–8 weeks, 9–12 weeks, 13–16 weeks), three frequencies (1–2 times per week, 3 times per week, 5 times per week), and three session durations (≤ 30 min, 30–60 min, > 60 min) were analyzed. Network meta-analysis revealed that RE was relatively effective SMD = -2.71, 95% CI (-3.96, -1.45), SUCRA = 88.5. For intervention duration, ≤ 4 weeks showed greater effectiveness SMD = -3.13, 95% CI (-4.21, -2.05), SUCRA = 97.2. Regarding frequency, 1–2 times per week was more effective SMD = -2.54, 95% CI (-3.57, -2.05), SUCRA = 97.7. For session duration, 30–60 min was the relatively effective SMD = -2.54, 95% CI (-3.41, -1.67), SUCRA = 99.0. Conclusion This network meta-analysis systematically evaluated different exercise types, durations, frequencies, and session lengths on the sleep quality of perinatal women, with a particular focus on the combination of exercise type and dosage. The findings revealed that RE lasting ≤ 4 weeks, performed 1–2 times per week, and 30–60 min per session, was the relatively effective. It is recommended that this regimen be prioritized in perinatal healthcare to optimize sleep and overall health. Future research should further explore the types, dosages, and combinations of exercise to provide more evidence for targeted interventions.
Guo et al. (2026) studied this question.