A Forest Plot is the name of a statistical information display created as part of a systematic review with meta-analysis of the literature. Interestingly, no one is sure why this statistical output is called a Forest Plot. Some think it is because the output resembles a tree. This article provides information about how to interpret a Forest Plot.

What is a Forest Plot? – What is it used for? – How do you read it?

This article describes the purpose of including a Forest Plot in a systematic review and meta-analysis. It lists the principal elements of a Forest Plot and describes how to interpret the results using an example of a continuous variable and an example of a binary (dichotomous) variable.

Forest Plot: A Forest Plot (See Figure 1 below) is the name of a specific statistical information display created as part of a systematic review of the literature. The Forest Plot presents a quantitative summary of the results of multiple studies on a specific topic. The advantage of a systematic review with a Forest Plot is that many small-sized studies are combined to provide a larger sample size and answer clinical questions with greater power (power = more chance of finding a statistical result if one exists).

The following provides specific examples to guide the reader in use of a systematic review that includes a meta-analysis and forest plot.

Systematic Review: The following are examples of a systematic review of the literature that focus  on specific topics such as “phlebitis and duration of a peripheral intravenous (IV) line” or “alcohol wipes versus alcohol barrier caps to prevent catheter associated central-line infection (CLABSI)”. The systematic review preferentially includes research studies that contributes data from randomized controlled trials (RCT) because randomization makes the groups more equitable and findings more generalizable. However, a systematic review may also include data from quasi-experimental studies where there were multiple groups, but randomization was not used to create the groups.

Study Inclusion Criteria: In the systematic review, the authors initially present the background and describe the criteria that were used to select the studies that will be combined into the Forest Plot. The data from the studies (sample size, characteristics, results) are entered in a specialized statistical program to combine the information for comparison.

Meta-analysis: Meta-analysis is the statistical method used to analyze the individual study data and create a Forest Plot.

Figure 1: Schematic of a Forest Plot showing a list of studies; the vertical line of “no effect”; label of effect for each side of the Forest Plot: squares and confidence interval lines (Box and Whiskers); and a diamond summary statistic.

The Forest Plot has a middle-vertical line is described as the “line of no effect” that divides the result into two sections (favors the intervention or test) or (does not support the intervention or test). In Figure 1, this vertical line is listed as the “label of no effect”.

Individual Studies: Each individual study is listed on the left side, with the authors’ names and year of publication (Figure 1).  Moving to the middle of the Forest Plot, each study is represented by a black square with a horizontal line through the square. The squares are different sizes, related to the sample size in the study. The bigger squares have more “weight”, which means the data is more meaningful due to a larger sample size with smaller confidence interval. The squares are positioned in the mid-section of the horizontal line as these lines represent the confidence intervals of the study result. If the horizontal line crosses the middle upright line, this means the result is not statistically significant.

  • Diamond Summary Statistic: At the bottom of the panel there will be a diamond that graphically represents a summary statistic for all the included studies.
  • Line of No Effect: The middle vertical line is labeled at the base as either 1 or 0.

1 indicates a binary (dichotomous) variable / result – “yes or no”, “positive or negative”

0 indicates a continuous variable / result – blood pressure, heart rate, height, hours, or similar

How to Interpret the Forest Plot for a Binary Result (Figure 2): When 1 is at shown at the base of the middle line, indicating a binary result, everything to the negative side is below 1 and everything to the positive side is above 1.

  • A non-significant result: A non-significant result would be a CI of -0.98 -to- 1.12.  Because the confidence interval includes 1, this result IS NOT significant (see Figure 2).
  • A significant result: A significant result would be a CI of 1.12 -to- 2.57.  Because the confidence interval does not include 1, this result IS significant

How to Interpret the Forest Plot for a Continuous Result: When 0 is in the middle this indicates a continuous result. Everything to the negative side is below 0. Everything to the positive side is above 0. A confidence interval (CI) and a p value are listed on the right side.

  • A non-significant result: A non-significant result would be a CI of -0.42 -to- 1.68.  Because the confidence interval includes 0, this result IS NOT significant (see Figure 2).
  • A significant result: A significant result would be a CI of 1.68 -to- 2.95.  Because the confidence interval does not include 0, this result IS significant (see Figure 2).

Figure 2: Schematic Elements of a Forest Plot


Published Example of a Forest Plot: Figure 3 is a Forest Plot from a recently published meta-analysis on the impact of early mobility in the ICU. The Forest Plot compares the number of ventilator-free days for patients who received early mobility in the ICU (intervention treatment label), compared with no early mobility (control treatment label).

Sample Size and Weight: In Figure 3, three of the six studies (50%) were extremely small with sample sizes under 25. The meta-analysis pools the findings from all these studies to give more power. Studies with larger sample sizes have greater weight in the analysis.

Interpreting the Result: In Figure 3 the mid-line number is 0, indicating measurement is a continuous variable, in this example the researchers are measuring the number of days the patient is not on the ventilator shortened to “ventilator-free days”.

Diamond Statistic and Final Result: In Figure 3 below, the Confidence Interval is 0.02-to-0.31 which does not include zero (it is above zero) so this result IS statistically significant. This is reflected in the position of the diamond which is to the right of the vertical line and does not cross it.

Figure 3: Forest Plot from Zhang et al.  (2019) Early mobilization of critically ill patients in the ICU – A systematic review and meta-analysis. PLoS ONE 14(10):e0223185.  Open Access.

Summary: This published Forest Plot (Figure 3) contains all the elements found in the schematic examples (Figures 1 and 2). Comparing the published and schematic versions will enhance knowledge of how a Forest Plot is presented in the literature.

This article presented a brief overview of a complex subject. To gain further insights it may be helpful to review some of the resources listed below. The two U-tube videos below are a great place to start.

Please contact Mary E. Lough PhD, RN if you have questions on this topic:



  1. T. Shaneyfelt How to Interpret a Forest Plot – Utube video


  1. Clinical Information Services: Utube video:


Article By: Mary E. Lough

Book an appointment or get a consultation.