Forest plot

A forest plot displays the results of the individual studies in a meta-analysis stacked in rows, with a combined estimate at the bottom. Each study's effect (such as a risk ratio or mean difference) is a box, the box area reflects its weight, and the horizontal line is its confidence interval. A vertical line marks no effect. The pooled result is a diamond whose width is its confidence interval. When reading one, check where each interval sits relative to the no-effect line, whether the studies point the same direction, and how wide the diamond is; scattered boxes or non-overlapping intervals signal heterogeneity, often reported as an I-squared value. For example, if eight trials of a fever drug each have wide, overlapping intervals crossing the no-effect line but the pooled diamond sits clearly to one side and does not cross it, pooling has revealed a consistent effect that single small trials could not detect on their own.