Meta-analysis shows intermittent fasting lowers BMI and fasting glucose in women with overweight or obesity

This study aimed to assess the effects of intermittent fasting (IF) on prespecified clinically relevant outcomes, including BMI, fasting blood glucose (FBG), and blood pressure, in women with overweight or obesity, and to explore the dose–response relationship and moderating factors of the intervention. Following PRISMA guidelines, we systematically searched PubMed, Web of Science, PsycINFO, and the Cochrane Library (up to August 1, 2025) for eligible randomized controlled trials (RCTs). Twenty-two studies involving 1,287 women with overweight or obesity were included. Mean differences (MD) were pooled using a restricted maximum likelihood random-effects model (REML). Heterogeneity was evaluated via I2 and Q tests, while publication bias was assessed using sensitivity analyses, Egger's regression, and the trim-and-fill method. Subgroup analyses, meta-regression, and nonlinear dose–response models were used to identify moderators. IF significantly reduced BMI in the final BMI model after exclusion of one influential study (MD = −0.41, 95% CI: −0.81 to−0.02, P = 0.0396) and FBG (MD = −2.18, 95% CI:−3.20 to−1.16, P < 0.0001) in women with overweight or obesity, with no significant effect on blood pressure. The initial BMI model showed substantial heterogeneity (I2 = 80.2%). Subgroup analyses suggested variability across intervention characteristics and participant strata, but these findings should be interpreted cautiously because several subgroup estimates were based on limited evidence. Dose–response analysis suggested a relatively favorable modeled frequency around twice weekly, with effects remaining relatively stable across 1,285–2,895 total intervention hours. IF may improve BMI and FBG in women with overweight or obesity, whereas no significant effects were observed for blood pressure outcomes. The dose–response findings suggest potentially favorable intervention patterns, but these should be interpreted cautiously because of heterogeneity across studies, variation in intervention and comparator conditions, possible co-interventions, and the limited certainty of the available evidence. Further well-designed trials are needed to confirm these findings and refine practical recommendations. https://www.crd.york.ac.uk/PROSPERO/view/CRD420251178671, identifier: CRD420251178671.