Forest Plot Generator | Free Online Meta-Analysis Tool

Q: Should I use a fixed-effect or random-effects model?

Use random effects when studies differ in populations or methods and you expect the true effect to vary; use fixed effect only when the studies are estimating one common effect. The Cochrane Handbook discusses the choice in Chapter 10.

Q: Why are ratios shown on a log scale?

On the log scale a ratio of 0.5 and a ratio of 2 are the same distance from 1, so benefits and harms are displayed symmetrically.

Q: Can I make a forest plot from odds ratios and confidence intervals?

Yes. Choose Estimate & CI and type each study's odds ratio with its lower and upper 95% limits.

Q: Is this forest plot generator free?

Yes. It is free, needs no account and runs entirely in your browser.

Study

Events (treatment)

No event (treatment)

Events (control)

No event (control)

Pooled risk ratio	0.37 (95% CI 0.24 to 0.58)
Test for overall effect	z = -4.35, p < 0.001
I²	82.5%represents considerable heterogeneity (75-100%)
Cochran's Q	28.51 (df = 5, p < 0.001)
τ² (between-study variance)	0.1981τ = 0.4450 on the log scale
95% prediction interval	0.09 to 1.49Where the true effect in a new study is likely to lie

Pooled risk ratio

0.37 (95% CI 0.24 to 0.58)

Test for overall effect

z = -4.35, p < 0.001

I²

82.5%represents considerable heterogeneity (75-100%)

Cochran's Q

28.51 (df = 5, p < 0.001)

τ² (between-study variance)

0.1981τ = 0.4450 on the log scale

95% prediction interval

0.09 to 1.49Where the true effect in a new study is likely to lie

How to Use the Forest Plot Generator

Choose the effect measure

Odds ratio, risk ratio, hazard ratio, mean difference or SMD.

Enter your studies

Type 2×2 counts, or each study's estimate with its 95% confidence interval.

Pick the model

Random effects (DerSimonian-Laird) or fixed effect (inverse variance).

Download the plot

Export a high-resolution PNG or a vector SVG, and copy the results sentence.

How to read a forest plot

Each row of a forest plot is one study: the square is its estimate, the square's size reflects its weight and the horizontal line is its 95% confidence interval. The diamond at the bottom is the pooled result, and its width is the pooled confidence interval.

The vertical line marks no effect (1 for ratios, 0 for differences). Ratios are plotted on a log scale so that effects in either direction look symmetrical. With random effects, the dashed line shows the 95% prediction interval: where the true effect in a new, similar study is likely to lie.

How it is calculated

Ratios are analysed on the log scale; SE = (ln upper − ln lower) / (2 × 1.96) when entered as a confidence interval
Fixed effect: inverse-variance weights w = 1/SE²
Random effects: DerSimonian-Laird τ² = max(0, (Q − df) / (Σw − Σw²/Σw)), weights 1/(SE² + τ²)
I² = (Q − df) / Q; prediction interval = μ ± t(k−2) × √(τ² + SE(μ)²)
Zero cells in 2×2 tables get 0.5 added to every cell

Sources

Forest Plot Generator: FAQ

Should I use a fixed-effect or random-effects model?

Use random effects when studies differ in populations or methods and you expect the true effect to vary; use fixed effect only when the studies are estimating one common effect. The Cochrane Handbook discusses the choice in Chapter 10.

Why are ratios shown on a log scale?

On the log scale a ratio of 0.5 and a ratio of 2 are the same distance from 1, so benefits and harms are displayed symmetrically.

Can I make a forest plot from odds ratios and confidence intervals?

Yes. Choose Estimate & CI and type each study's odds ratio with its lower and upper 95% limits.

Is this forest plot generator free?