How to Interpret a Forest Plot

A forest plot displays the result of each study in a meta-analysis and, usually, the pooled result, on one scale so you can see the size, direction, precision and consistency of effects at a glance.

Anatomy

• One row per study with its label and often the raw data (events/total).
• A square for each study's point estimate; the square's size reflects its weight in the pooled analysis.
• A horizontal line through each square showing the 95% confidence interval.
• A vertical line of no effect: 1 for ratios (odds, risk, hazard) and 0 for differences (mean difference).
• A diamond for the pooled estimate: its centre is the estimate and its width the confidence interval.
• Labels showing which side favours which treatment.
• Heterogeneity statistics (such as I² and the chi-squared test) and the model used (fixed or random effects).

How to read it

1. Note the effect measure and which direction favours the intervention.
2. See whether study confidence intervals cross the line of no effect.
3. Look at the diamond: does it cross the line? How wide is it?
4. Look at consistency: do studies point in the same direction and overlap?
5. Check weights: is one large study dominating?
6. Read heterogeneity and ask whether a single pooled number makes sense.

Illustrative example

Invented for teaching. A plot shows 8 trials with risk ratios between 0.60 and 1.05 and a pooled risk ratio of 0.78 (95% CI 0.66 to 0.92), I² = 20%. The diamond sits left of 1, so the pooled result favours the intervention, and the moderate spread of study results looks consistent. MRN's free forest plot generator draws plots from user-entered estimates.

Common pitfalls

• Reading a wide interval crossing 1 as "no effect" rather than "uncertain".
• Ignoring heterogeneity and risk of bias.

Key takeaways

• Squares = studies, lines = confidence intervals, diamond = pooled result.
• Interpret together with heterogeneity and quality.

This page is general educational information about research methods. It is not medical, legal or regulatory advice; follow your institution's ethics board and your target journal's instructions.