Forest Plot Generator

A forest plot is a graphical display used in meta-analysis to show the effect sizes and confidence intervals from multiple individual studies alongside a pooled (summary) estimate. Each study is represented by a horizontal line (the confidence interval) and a square (the point estimate). Forest plots are standard in systematic reviews to visually assess consistency across studies and the overall direction of evidence.

This tool supports Risk Ratio (RR), Odds Ratio (OR), Hazard Ratio (HR), and Mean Difference (MD). For ratio-based measures (RR, OR, HR), calculations are performed on the log scale to ensure symmetry, and the null value is 1. For Mean Difference, the null value is 0.

A fixed effect model assumes all studies are estimating the same true effect and that variability is due to sampling error alone. A random effects model (DerSimonian-Laird) assumes true effects vary between studies and incorporates between-study variance (tau²) into the pooled estimate. Random effects is generally preferred when heterogeneity is present (I² > 25–30%).

I² quantifies the percentage of total variation across studies due to heterogeneity rather than chance. Values of 0–25% suggest low heterogeneity, 25–50% moderate, 50–75% substantial, and above 75% considerable heterogeneity. High I² values may indicate that studies differ in population, intervention, or outcome definitions, and a random effects model is recommended.

This tool currently generates a single-group forest plot with individual study estimates and a summary diamond. Subgroup analysis — where studies are divided into pre-specified subsets — would require labeling studies by subgroup manually before entering them. Future versions may support explicit subgroup fields.

The diamond at the bottom of a forest plot represents the pooled (meta-analytic) estimate. Its horizontal center is the pooled effect size, and the left and right tips mark the lower and upper bounds of the pooled confidence interval. A diamond that does not cross the line of no effect (1 for ratios, 0 for differences) indicates a statistically significant overall result.

Most peer-reviewed journals accept forest plots as high-resolution PNG, SVG, or PDF files. Plots should clearly label the effect measure, confidence interval level, null line, study weights, and the pooled estimate with its CI. Some journals also require reporting of Q and I² statistics, which this tool calculates automatically.

Weights are calculated using the inverse-variance method. Each study's weight is proportional to the reciprocal of its variance (derived from the width of its confidence interval). Under a random effects model, between-study variance (tau²) is added to each study's within-study variance before computing weights, giving more even distribution of weights across studies compared to fixed effect.