# Meta-analysis

A meta-analysis is a statistical method that quantitatively combines results from multiple independent studies on the same question. It yields a pooled effect estimate, with tighter confidence intervals than any single study alone. Gene V. Glass coined the term in 1976. The process: pull an effect size from each study, weight it (usually by inverse variance), and combine into a weighted average. It is shown in a 'forest plot', where each horizontal line is one study, and the diamond at the bottom is the pooled result. A key issue is heterogeneity: how much the true effects vary across studies, beyond chance. It is quantified by I². Below 25% is low, 50 to 75% is moderate, and above 75% is high, though I² is sensitive to the number of studies. Publication bias is another risk. It is checked with funnel plots, Egger's test, or trim-and-fill. In longevity research, meta-analyses pool observational cohorts or trials to detect modest effects, like the survival benefit of physical activity (something you can act on), or the link between telomere length and death, that single studies cannot resolve on their own. But their conclusions are only as good as the studies they pool. Shared systematic biases get amplified, not corrected.

## Sources

- Glass GV. (1976). Primary, Secondary, and Meta-Analysis of Research. Educational Researcher. https://doi.org/10.3102/0013189x005010003
- Higgins JP, Thompson SG, Deeks JJ, et al.. (2003). Measuring inconsistency in meta-analyses. BMJ. https://doi.org/10.1136/bmj.327.7414.557
- Page MJ, McKenzie JE, Bossuyt PM, et al.. (2021). The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. International Journal of Surgery. https://doi.org/10.1016/j.ijsu.2021.105906

---

_Canonical: https://longevity-switzerland.com/en/glossary/meta-analysis · Part of Longevity Cities · Updated 2026-06-22_