Meta-analysis & Forest Plot
Pool the results of several independent studies. Supports binary outcomes (OR odds ratio) and continuous outcomes (MD mean difference), provides fixed-effect and random-effects (DerSimonian-Laird) models, computes heterogeneity (I², Q, τ²), and draws a forest plot.
① Choose data type and model
② Enter the study data
One study per row: study name + treatment-group events + treatment-group non-events + control-group events + control-group non-events (the 2×2 cells a b c d), Tab/comma-separated.
How to use & methodology
What is meta-analysis for?
Meta-analysis quantitatively pools the results of several independent studies on the same question into a combined effect estimate and assesses consistency between studies. It is one of the highest-level evidence types in evidence-based medicine, and the forest plot is its standard presentation.
How do I choose between fixed and random effects?
Fixed effect assumes all studies estimate the same true effect with only sampling error; random effects assume the true effects themselves differ between studies. When heterogeneity is large (I²>50%), use random effects; most clinical meta-analyses report random effects by default, which is more robust.
How do I interpret I²?
I² is the percentage of total variation due to between-study variation: 0–40% heterogeneity may be unimportant, 30–60% moderate, 50–90% substantial, 75–100% considerable. The higher I², the less consistent the results and the more cautiously you should pool and explore causes (population, dose, design differences).
How do I read the forest plot?
Each study is a square (effect size, square size = weight) with a horizontal line (95% CI); the diamond at the bottom is the pooled effect, its width being its CI. The central dashed vertical line is the null-effect line (OR=1 or MD=0); a line or diamond crossing it means that result is not statistically significant.