Forest Plot Interpretation: How to Read One

A forest plot is a graphical display used in meta-analysis to visualize individual study results and the pooled (combined) effect estimate. Each study is represented by a square (point estimate) with a horizontal line (95% confidence interval), and the overall result is shown as a diamond at the bottom. Forest plots are the standard visualization in Cochrane systematic reviews and are generated using R, Stata, or RevMan.

To read a forest plot, work through five checks in order: (1) whether the diamond crosses the vertical line of no effect, (2) whether the individual study squares sit on the same side of that line, (3) the I-squared heterogeneity value, (4) whether any single large square dominates the study weights, and (5) the overall p-value and confidence interval width. Each check is unpacked in the step-by-step section below.

Forest plot interpretation is one of the most essential skills for any researcher working with systematic reviews or meta-analyses. Whether you are reviewing published evidence, defending a thesis, or conducting your own quantitative synthesis, understanding how to read a forest plot determines whether you can evaluate the strength, consistency, and clinical relevance of pooled findings. This guide breaks down every component, then addresses the misinterpretations that most often undermine evidence appraisal.

Why Every Meta-Analysis Uses a Forest Plot

Forest plots appear in nearly every systematic review that includes quantitative synthesis. They are the primary output of meta-analysis software and the standard figure format required by the Cochrane Handbook for Systematic Reviews of Interventions (Higgins et al., 2023). Cochrane reviews, Campbell reviews, and most health sciences meta-analyses use forest plots as the definitive summary of evidence.

The fundamental purpose of a forest plot is to answer two questions simultaneously: what does each individual study show, and what do all studies show when combined? A single forest plot visualizes the pooled effect size, making it possible to see both the direction and precision of the overall finding at a glance. This dual function, individual and aggregate, makes forest plots uniquely powerful for evidence synthesis.

Anatomy of a Forest Plot, Component by Component

Every forest plot contains seven structural elements. Understanding each component is the prerequisite for accurate forest plot interpretation. The following breakdown covers each element from left to right as it appears on a standard plot.

Study Labels

The left column lists each included study, typically formatted as "Author (Year)." Studies may be ordered chronologically, alphabetically, or by subgroup. In Cochrane forest plots, studies within subgroups are listed together with subtotals before the overall pooled estimate.

Effect Size Squares

Each study's result is displayed as a square positioned on the horizontal axis. The center of the square represents the study's point estimate, the calculated effect size for that individual study. The position along the axis tells you the magnitude and direction of the effect.

Confidence Interval Lines

A horizontal line extends from each side of the square, representing the study's 95% confidence interval. The confidence interval width indicates the precision of the estimate. Narrow intervals mean more precise estimates (typically from larger studies), while wide intervals indicate greater uncertainty. If the horizontal line crosses the vertical line of no effect, that study's result alone is not statistically significant.

The Vertical Line of No Effect

A vertical reference line marks the point of no effect. For mean difference (MD) and standardized mean difference (SMD), this line sits at zero. For odds ratio forest plots, risk ratio forest plots, and hazard ratio forest plots, it sits at one. Any study or pooled result whose confidence interval crosses this line has not demonstrated a statistically significant effect.

The Diamond (Pooled Estimate)

The diamond at the bottom of the plot represents the diamond summary estimate, the overall pooled effect size calculated from all included studies. The center of the diamond is the combined point estimate, and the left and right tips show the 95% confidence interval. If the diamond does not cross the line of no effect, the overall result is statistically significant. The diamond is the single most important element on the plot because it represents the synthesis of all available evidence.

Weight Column

A column on the right side of many forest plots displays the study weight as a percentage. Study weight reflects how much each study contributes to the pooled estimate. In a fixed-effect model, weight is determined entirely by sample size and variance. In a random-effects model, weights are more evenly distributed because the model accounts for both within-study and between-study variance (DerSimonian & Laird, 1986). The size of each square on the plot is proportional to its weight, larger squares represent studies with greater influence on the pooled result.

Forest Plot Statistics

Below the diamond, most forest plots display key statistical summaries: the I-squared (I²) statistic for heterogeneity, tau-squared for between-study variance, the Q-test p-value, and the overall effect p-value. These numbers are critical for evaluating whether the pooled result is reliable and whether the included studies are measuring the same underlying effect.

Forest Plot Interpretation: A Step-by-Step Reading Guide

Read a forest plot in five steps: (1) check the diamond position relative to the line of no effect, (2) assess individual study consistency, (3) evaluate heterogeneity via I², (4) examine study weights, and (5) check the overall p-value. This framework ensures you extract every piece of information the plot provides without skipping critical details.

Step 1: Look at the diamond. Does the diamond cross the line of no effect? If no, the pooled result is statistically significant. Note the direction, is the effect favoring treatment or control, intervention or comparator? The position of the diamond summary estimate tells you both the direction and magnitude of the combined finding.

Step 2: Check individual study consistency. Are most study squares on the same side of the line of no effect? When studies cluster on one side, the evidence is more consistent. When studies scatter across both sides, the pooled estimate may be masking important differences between studies. Consistent results strengthen your confidence in the meta-analysis results.

Step 3: Assess heterogeneity. Look at the I-squared value below the plot. I-squared measures the percentage of variability across studies that is due to real differences rather than chance. I-squared interpretation follows Cochrane thresholds: below 25% is low, 25-50% is moderate, 50-75% is substantial, and above 75% is considerable heterogeneity. High heterogeneity means the studies are not measuring the same thing, and the pooled estimate should be interpreted with caution.

Step 4: Examine study weights. Are the squares roughly similar in size, or does one study dominate? A single large study contributing 40% or more of the total weight can drive the pooled estimate. If that dominant study has methodological limitations, the pooled result may not reflect the broader evidence base. The study weight distribution matters as much as the final number.

Step 5: Check the overall p-value. The p-value for the overall effect test tells you whether the pooled result is statistically distinguishable from zero (or one, for ratio measures). A p-value below 0.05 is conventionally considered statistically significant, but always interpret this alongside the confidence interval width and clinical relevance of the effect magnitude.

Need publication-ready forest plots for your meta-analysis? Our biostatisticians generate and interpret forest plots, subgroup analyses, and sensitivity analyses using validated statistical software. request a professional research quote, or explore our meta-analysis services.

Effect Size Measures on Forest Plots

The type of effect size displayed on a forest plot depends on the outcome being measured and the study designs included. Choosing the correct metric and scale is essential for accurate forest plot interpretation.

Mean Difference (MD) and Standardized Mean Difference (SMD)

When all studies measure outcomes on the same scale (e.g., blood pressure in mmHg), the forest plot displays mean differences with the line of no effect at zero. When studies use different scales to measure the same construct (e.g., pain measured by VAS and NRS), the standardized mean difference is used, typically reported as Cohen's d or Hedges' g. Hedges' g applies a correction for small sample bias, making it the preferred choice for meta-analyses with studies under 20 participants per group. Before creating a forest plot, calculate your effect sizes with our free our free effect size calculator.

Odds Ratio (OR) and Risk Ratio (RR)

For dichotomous outcomes (event occurred or did not), forest plots display odds ratios or risk ratios. An odds ratio compares the odds of an event in the treatment group versus the control group. A risk ratio compares the probability of an event. Both use a line of no effect at 1, and values are plotted on a logarithmic scale to ensure symmetry, an OR of 0.5 (halved odds) and an OR of 2.0 (doubled odds) appear equidistant from 1.

Hazard Ratio (HR)

For time-to-event outcomes (survival analysis), forest plots display hazard ratios. Like OR and RR, the line of no effect is at 1 and values are plotted on a log scale. A hazard ratio below 1 typically indicates a protective effect (slower event occurrence in the treatment group).

Understanding Heterogeneity on a Forest Plot

Heterogeneity on a forest plot reveals whether the included studies are measuring the same underlying effect or producing genuinely different results. Visual and statistical cues work together to quantify this variation and determine whether the pooled effect size is a meaningful summary or an oversimplification.

Visual cues provide the first indication. When confidence intervals overlap substantially and most squares cluster near the diamond, heterogeneity is low. When intervals are scattered across the plot with minimal overlap, or when some studies show strong positive effects while others show negative effects, heterogeneity is high. In our meta-analysis work, the most common misinterpretation we encounter is researchers focusing on the diamond while ignoring I² values above 75%.

I-squared interpretation follows established Cochrane thresholds for the heterogeneity assessment:

Cochrane classifies heterogeneity as low (I² < 25%), moderate (25-50%), substantial (50-75%), or considerable (>75%), these thresholds guide whether the pooled estimate should be interpreted with caution (Higgins et al., 2023).

When heterogeneity is substantial or considerable, subgroup analysis and meta-regression can explore potential sources. Common sources include differences in study populations, interventions, outcome measurement, and follow-up duration. The Q-test provides a p-value for heterogeneity, but it has low statistical power with few studies and excessive power with many studies, I² is generally more informative. Tau-squared quantifies the absolute amount of between-study variance in a random-effects model. For a deeper discussion, see our guide on understanding heterogeneity in meta-analysis.

In a random-effects meta-analysis, study weights are more evenly distributed than in a fixed-effect model, because the random-effects model accounts for both within-study and between-study variance (DerSimonian & Laird, 1986). This distinction matters because the model choice affects both the width of the diamond and the relative influence of each study.

How Forest Plot Weights and the Pooled Diamond Are Calculated

Most guides describe the diamond without showing where it comes from. Understanding the arithmetic is what separates reading a forest plot from auditing one, and it is exactly what a thesis committee or peer reviewer will probe. Every pooled estimate on a forest plot is an inverse-variance weighted average, and you can reproduce it by hand.

Step 1: Convert each study to an effect size and its variance. For ratio measures, work on the log scale. A study with effect size y_i (for example, ln(OR)) has a within-study variance v_i, and its standard error is SE_i = sqrt(v_i). The inverse-variance weight is simply:

w_i = 1 / v_i

Studies with small variance (large, precise studies) receive large weights. This is why the square sizes on the plot are proportional to weight, not to sample size directly.

Step 2: Compute the fixed-effect pooled estimate. The pooled point estimate is the weighted mean, and its standard error comes directly from the summed weights:

• Pooled effect: M = (Σ w_i · y_i) / (Σ w_i)
• Standard error: SE(M) = sqrt(1 / Σ w_i)
• 95% confidence interval: M ± 1.96 · SE(M)

The tips of the diamond sit at the lower and upper limits of that confidence interval; the center sits at M.

Step 3: Quantify heterogeneity before trusting a fixed-effect diamond. Cochran's Q and I² come from the same weights:

• Q = Σ w_i · (y_i − M)²
• I² = max(0, (Q − df) / Q) × 100%, with df = k − 1
• Between-study variance (DerSimonian-Laird): τ² = max(0, (Q − df) / C), where C = Σ w_i − (Σ w_i² / Σ w_i)

Step 4: Re-weight for a random-effects diamond. A random-effects model adds τ² to every within-study variance, so the random-effects weight becomes:

w*_i = 1 / (v_i + τ²)

The pooled estimate and confidence interval use the same formulas as Step 2 but substitute w*_i for w_i. Because τ² is added to every study, the largest studies lose some of their dominance and the diamond widens. That is the visual signature of a random-effects plot.

A Four-Study Worked Example

Consider four studies reporting log odds ratios:

Fixed-effect result. Σ w_i = 105 and Σ w_i·y_i = 28.0, so M = 28.0 / 105 = 0.267 (OR = e^0.267 = 1.31). SE(M) = sqrt(1/105) = 0.098, giving a 95% CI of 0.267 ± 0.191 = [0.075, 0.458] on the log scale, or OR 1.08 to 1.58. The diamond center sits at 1.31 with tips at 1.08 and 1.58.

Heterogeneity. Q = 25(0.133)² + 20(0.033)² + 10(0.433)² + 50(−0.167)² = 3.73, with df = 3. I² = (3.73 − 3) / 3.73 = 19.6% (low). For τ², Σ w_i² = 3625, so C = 105 − (3625/105) = 70.48 and τ² = 0.73 / 70.48 = 0.0104.

Random-effects result. Adding τ² = 0.0104 to each variance gives w*_i of 19.8, 16.6, 9.1, and 32.9 (Σ = 78.4). Study D's share falls from 47.6% to 42.0% while the smaller studies gain influence. The pooled estimate becomes M = 22.54 / 78.4 = 0.288 (OR 1.33), SE = sqrt(1/78.4) = 0.113, and the 95% CI widens to [0.066, 0.509], or OR 1.07 to 1.66. The random-effects diamond is visibly wider than the fixed-effect one even though heterogeneity here is only modest.

You can reproduce this in R with metafor:

library(metafor)
yi <- c(0.40, 0.30, 0.70, 0.10)   # log odds ratios
vi <- c(0.04, 0.05, 0.10, 0.02)   # within-study variances

# Fixed-effect and random-effects (DerSimonian-Laird) models
rma(yi, vi, method = "FE")
rma(yi, vi, method = "DL")

# Forest plot with exponentiated axis for odds ratios
res <- rma(yi, vi, method = "DL")
forest(res, atransf = exp, showweights = TRUE)

When a reviewer asks why your diamond differs from a naive average of the study effects, this is the answer: the diamond is a precision-weighted average, and the random-effects version deliberately pulls toward the smaller studies by inflating every variance by τ². For the estimator choice behind τ², see our guide on REML versus DerSimonian-Laird.

Common Forest Plot Misinterpretations

Five recurring errors compromise forest plot interpretation across disciplines. Recognizing these pitfalls prevents flawed conclusions that can misguide clinical practice and research direction.

Confusing statistical significance with clinical significance. A diamond that does not cross the line of no effect indicates statistical significance, but the magnitude of the effect may be too small to matter clinically. A pooled standardized mean difference of 0.1 might be statistically significant with thousands of participants yet meaningless in practice. Always evaluate whether the effect size is clinically relevant, not just statistically distinguishable from zero.

Ignoring heterogeneity when the diamond looks favorable. A well-positioned diamond can create false confidence. If I² is 80%, the studies fundamentally disagree about the magnitude or even the direction of the effect. The diamond in that scenario is an average of conflicting findings, not a robust summary. Always check I-squared before drawing conclusions from the pooled effect size.

Over-interpreting results from few studies. A meta-analysis with two or three studies produces a forest plot that looks like any other. However, the pooled estimate from two studies is highly sensitive to each individual result, the heterogeneity statistics are unreliable, and the confidence interval may be misleadingly narrow if both studies happen to agree. Sensitivity analysis tests result robustness by removing one study at a time, this is critical when the forest plot includes few studies.

Mistaking fixed-effect for random-effects forest plots. The same data can produce different diamonds depending on the model. A fixed-effect model assumes one true underlying effect, producing a narrower diamond driven by the largest studies. A random-effects model accounts for between-study heterogeneity, producing a wider diamond with more evenly distributed weights. Check which model was used before interpreting the result, most systematic reviews use random-effects models, but not all.

Reading odds ratios or risk ratios on a linear scale instead of a log scale. Ratio measures (OR, RR, HR) should be plotted on a logarithmic scale. On a log scale, an OR of 0.5 and an OR of 2.0 are equidistant from 1. On a linear scale, they are not, creating a visual distortion that makes effects appear asymmetric. Properly constructed forest plots use log scales for ratio measures, but always verify this when reading unfamiliar plots.

How to Create a Forest Plot

Several software options produce publication-ready forest plots, ranging from free tools requiring no installation to statistical packages used by biostatisticians worldwide. Your choice depends on your technical comfort level and the complexity of your analysis. If you would rather have the pooled analysis run and the plot produced for you, our meta-analysis biostatistics support delivers publication-ready forest plots with reproducible code.

R (metafor and meta packages), R is the most flexible platform for forest plot generation. The metafor package by Viechtbauer (2010) provides full control over every visual element, from label formatting to color coding by subgroup. The meta package by Balduzzi et al. offers a simpler interface with sensible defaults. Both packages handle effect size visualization for all common measures (MD, SMD, OR, RR, HR) and support subgroup, cumulative, and leave-one-out forest plots.

Stata (metan command), Stata's metan and admetan commands generate forest plots with options for random-effects and fixed-effect models, subgroup analysis, and prediction intervals. Stata is widely used in epidemiology and health services research, and its forest plot output is accepted by most journals without modification.

RevMan (Cochrane's tool), Review Manager is Cochrane's free software for conducting systematic reviews. It produces standardized forest plots that match Cochrane Handbook formatting requirements. RevMan is the required tool for Cochrane reviews and provides a guided interface that does not require programming knowledge.

Our free online forest plot generator, Create your own publication-ready forest plot with our free our forest plot generator. Enter your study data, select the effect size measure, and download a high-resolution figure suitable for journal submission, no software installation, no coding, no subscription required.

For researchers who want full control over their analysis, our complete meta-analysis guide covers the end-to-end process from data extraction through forest plot generation and interpretation. When your forest plot is ready, assess reporting quality using the GRADE summary of findings framework to contextualize your results within a broader evidence evaluation.

Different programs produce forest plots with varying customization. Explore software options for generating high-quality forest plots beyond RevMan.

For step-by-step instructions on building your own forest plot with subgroups and cumulative analysis, see our forest plot creation guide.

A specialized variant, the cumulative meta-analysis forest plot, shows how the pooled estimate changes as each study is added chronologically.

When the figure itself needs polishing rather than re-running the model, our scientific figure preparation turns raw statistical output into journal-grade charts.