A forest plot is the single most recognizable output of a meta-analysis. It displays each study's effect estimate alongside a pooled summary, letting readers instantly see where studies agree, where they diverge, and how much weight each one carries. If you are preparing a manuscript, a grant, or a thesis, knowing how to build and interpret this plot correctly is non-negotiable.

Try our free Forest Plot Generator to build publication-ready plots without writing a single line of code.

What You Need Before You Start

Before generating a forest plot, you need three things assembled in a clean spreadsheet or data file.

Effect size estimates for each study: odds ratios, risk ratios, standardized mean differences (Cohen's d or Hedges' g), or correlation coefficients depending on your outcome type.

Variance or confidence interval bounds for each estimate. Most software accepts either standard errors or the lower and upper bounds of the 95% confidence interval directly.

Study labels: author names, publication years, and sample sizes. These appear on the left axis and help readers locate familiar studies.

If you have not yet computed your effect sizes, use the Effect Size Calculator to convert raw group means, proportions, or correlation coefficients into a common metric before importing them.

Choosing Your Heterogeneity Model: REML vs DerSimonian-Laird

The choice of variance estimator determines how much weight your random-effects model assigns to smaller versus larger studies, and it affects the width of the pooled confidence interval.

DerSimonian-Laird (DL) is the classic method. It uses a method-of-moments estimator for between-study variance (tau-squared) that is fast and widely cited. However, it tends to underestimate tau-squared when the number of studies is small, which can produce confidence intervals that are too narrow.

REML (Restricted Maximum Likelihood) is the modern default in most software packages including R's metafor and Stata's metan. REML produces less biased tau-squared estimates, especially with fewer than 20 studies, and is now recommended by the Cochrane Handbook for most applications.

For a fixed-effect model, the choice does not apply because fixed-effect models assume no between-study variance. Use a fixed-effect model only when you have a strong theoretical reason to believe all studies are estimating exactly the same underlying true effect, which is rare in practice.

Practical guidance: default to REML unless a specific journal or method requires DL. When you switch estimators, recheck your I-squared, tau-squared, and the prediction interval width, because these will change.

Building the Plot: A Step-by-Step Walkthrough

Step 1: Import your data. In the Forest Plot Generator, paste your study labels, effect sizes, and confidence interval bounds into the data entry panel. The tool accepts Cohen's d, Hedges' g, odds ratios, risk ratios, and correlation coefficients.

Step 2: Select your effect measure and model. Choose the effect size type that matches your data. Select random-effects with REML as your starting model. The plot will render immediately with study squares scaled by inverse variance weight.

Step 3: Read the study squares and lines. Each square represents one study's point estimate. The square's size reflects the study's weight in the analysis: larger squares mean more weight. The horizontal line through the square is the 95% confidence interval. A line that crosses the null (zero for mean differences, one for ratios) indicates a non-statistically-significant result for that study individually.

Step 4: Interpret the diamond. The pooled effect appears as a diamond at the bottom. The center of the diamond is the summary estimate. The left and right tips are the confidence interval limits. A narrow diamond means more precise pooled evidence; a wide diamond indicates high heterogeneity or few studies.

Step 5: Add the prediction interval. The prediction interval is the range where you expect 95% of true effects to fall in future similar studies. It is wider than the confidence interval and is arguably more important for clinical decision-making. If your confidence interval excludes the null but your prediction interval crosses it, the pooled effect may not replicate in every population. Always report the prediction interval alongside I-squared.

Step 6: Check your heterogeneity statistics. I-squared above 50% suggests substantial heterogeneity. Tau-squared gives the absolute variance in true effects. The Q statistic p-value tests whether heterogeneity exceeds chance, though it has low power with few studies.

Subgroup Analysis on a Forest Plot

Subgroup analysis lets you test whether the pooled effect differs across pre-specified categories: intervention type, risk of bias level, patient age group, or geographic region.

To add subgroups in the Forest Plot Generator, assign each study a group label in the subgroup column. The plot will render each group as a separate mini-forest with its own diamond, then display a test for subgroup differences (the Q-between statistic with a p-value).

Interpret subgroup results conservatively. A statistically significant Q-between only confirms that subgroup effects differ from each other; it does not prove that the subgroup variable causes the difference. Always pre-register subgroup hypotheses before analysis to avoid data-dredging.

Diagnostic Plots: Galbraith and Baujat

Once you have your main forest plot, two additional diagnostic plots reveal information that the standard forest plot cannot.

The Galbraith plot (also called a radial plot) plots each study's z-score divided by its standard error on the y-axis against one divided by its standard error on the x-axis. Studies that sit far from the regression line are outliers contributing disproportionately to heterogeneity. See our detailed guide on Galbraith and Baujat diagnostic plots for a full walkthrough.

The Baujat plot places each study's contribution to the overall Q statistic on the x-axis and its influence on the pooled estimate on the y-axis. Studies in the upper-right quadrant are both heterogeneous and influential; these are the studies to examine first when your I-squared is unacceptably high.

Both plots are available as tabs inside the Forest Plot Generator.

Cumulative Meta-Analysis View

The cumulative meta-analysis tab re-runs the pooled estimate each time a new study is added, ordered by publication date. This shows when the evidence first crossed statistical significance and whether the summary effect has been stable or drifting. See the dedicated guide on cumulative meta-analysis for interpretation strategies.

Exporting Your Plot

The Forest Plot Generator exports publication-ready SVG and PNG files suitable for journal submission. It also provides the equivalent R code using the metafor package so you can reproduce the exact plot in your own environment, adjust axis labels, and add custom annotations required by specific journals.

Run a sensitivity analysis after finalizing your main plot to test how the pooled estimate changes when you remove each study one at a time. The Sensitivity Analysis Tool integrates directly with the same effect size data and highlights which studies are responsible for the bulk of your summary estimate.

For publication bias assessment, pair your forest plot with a funnel plot. The Funnel Plot Generator generates Egger's test and trim-and-fill estimates alongside the visual asymmetry check.

Key Takeaways

FAQ

What is a forest plot in meta-analysis?

A forest plot displays each included study as a square (effect estimate) with a horizontal line (confidence interval) and combines them into a pooled summary diamond at the bottom. It allows readers to see individual study results and the overall synthesis in one figure.

Should I use a fixed-effect or random-effects model for my forest plot?

Use a random-effects model in most cases. A fixed-effect model assumes all studies estimate an identical true effect, which is rarely justified when studies differ in population, intervention, or follow-up duration. Random-effects models account for between-study variance and produce more conservative, generalizable pooled estimates.

What does a wide prediction interval mean?

A wide prediction interval means the true effect is likely to vary considerably across different settings or populations. Even if the pooled estimate is statistically significant, a prediction interval that crosses the null line suggests the intervention may not work everywhere and clinical application should be cautious.

How many studies do I need for a forest plot?

There is no hard minimum, but a random-effects meta-analysis with fewer than 5 studies produces highly unstable tau-squared estimates. With 2 to 4 studies, consider a fixed-effect model or a narrative synthesis with a table instead of a pooled diamond.

Can I run subgroup analysis if I only have 10 studies total?

Yes, but power is very limited. A subgroup analysis comparing two groups of 5 studies each will rarely detect a real subgroup difference even if one exists. Pre-register your subgroups, report the Q-between p-value, and note the low power explicitly in your manuscript.

What is the difference between I-squared and tau-squared?

I-squared expresses heterogeneity as a percentage of total variance that is due to between-study differences rather than chance, making it a relative measure. Tau-squared is the absolute variance of true effects across studies. A small I-squared can accompany a clinically meaningful tau-squared if study precision is high, so always report both.

How do I handle outlier studies on a forest plot?

First, use the Galbraith or Baujat plot to identify which studies are outliers and whether they are also influential. Then perform a sensitivity analysis removing those studies one at a time. Report both the full-set and leave-one-out results; do not silently exclude studies without a pre-registered justification.

Need help with your systematic review or meta-analysis? Get a free quote from our team of PhD researchers.