Intermittent Fasting and Obesity-Related Health Outcomes: An Umbrella Review of Meta-analyses of Randomized Clinical Trials.
Level 1 - systematic review of randomized trials
Umbrella review of meta-analyses of randomized clinical trials
PubMed 34919135 · doi:10.1001/jamanetworkopen.2021.39558
What was done
An umbrella review searched PubMed, Embase, and the Cochrane Database of Systematic Reviews through January 12, 2021, for meta-analyses of randomized clinical trials (RCTs) assessing intermittent fasting (zero-calorie alternate-day fasting, modified alternate-day fasting, 5:2 diet, and time-restricted eating) on obesity-related health outcomes in adults. Effect sizes were recalculated using random-effects models, and evidence quality per association was assessed via GRADE criteria.
What was found
The review evaluated 11 meta-analyses comprising 130 RCTs (median sample size, 38 participants [IQR 24-69]; median follow-up, 3 months [IQR 2-5]) covering 104 unique associations (median 4 studies per association). Twenty-eight associations (27%) showed statistically significant benefits for BMI, body weight, fat mass, LDL cholesterol, total cholesterol, triglycerides, fasting glucose, fasting insulin, HOMA-IR, and blood pressure, while also being associated with reduced fat-free mass. Only 1 association (1%) was backed by high-quality evidence (modified alternate-day fasting for 1–2 months reducing BMI), 6 (6%) by moderate-quality evidence, 22 (21%) by low-quality evidence, and 75 (72%) by very-low-quality evidence.
Why it matters
While intermittent fasting—particularly modified alternate-day fasting—is supported as a weight-loss approach, the overwhelming majority (93%) of claimed health benefits currently rest on low- or very-low-quality trial evidence.
Limits
Individual trials within the synthesized meta-analyses featured small sample sizes (median n = 38) and short intervention periods (median 3 months). Most associations demonstrated low or very low GRADE certainty, and hard clinical outcomes like cardiovascular events and mortality were not measured.
Cited by
- contradicts There are no clinical trials in humans showing a clear lifespan or health benefit from fasting.