Interpreting meta-analysis graphs: The forest plot

A Summary of

Ried, K. (2006). Interpreting and understanding meta-analysis graphs: A practical guide. Australian Family Physician, 35(8), 635-638.

Relevance for Public Health

Meta-analysis can provide a very useful and powerful summary statistic of intervention effectiveness from research evidence. This practical resource outlines how to interpret the results of a meta-analysis and forest plots. This is a critical step in the process of using the results of a meta-analysis to inform program and policy decisions.

Description

A systematic review is a rigorous summary of research evidence using systematic, explicit and reproducible methods. In some cases, the results of similar studies in a systematic review can be quantitatively combined as a meta-analysis. The forest plot graphically displays the results of a meta-analysis to show the overall effect for the question of interest. This practical method provides an overview of critically appraising systematic reviews and meta-analysis, and outlines how to interpret forest plots. To learn more about forest plots and meta-analysis, see the videos in NCCMT’s video series “Understanding Research Evidence.”

By using a rigorous process to search, appraise and synthesize research, the systematic review provides a summary of knowledge on a question that assesses the validity and quality of included primary research studies (Ciliska, Cullum & Marks, 2001). Meta-analysis statistically combines the samples of each contributing study to create an overall summary statistic that is more precise than the effect size in the individual studies (Ciliska, Cullum & Marks, 2001). To learn more about interpreting and applying the results of a meta-analysis, see Freemantle & Geedes, 1998.

This resource includes:

  • Using criteria to assess systematic reviews and meta-analyses (Table 1)
  • Interpreting forest plots with a binary outcome variable (Figure 1)
  • Interpreting forest plots with a continous outcome variable (Figure 2)
  • Comparing meta-analyses of binary and continuous variables and outcome effect measures (Table 2)

Accessing the Method/Tool


Language(s)

Not specified

Format(s)

Not specified

Cost

Not Specified

Implementing the Method/Tool


Time for Participation/Completion

Information not available

Additional Resources and/or Skills Needed for Implementation

Not Specified

Steps for Using Method/Tool

Some useful definitions and principles for this method include:

  • When outcomes are binary (dichotomous), such as mortality, meta-analyses use relative risk (RR) or odds ratios (OR) for reporting the summary statistic.
  • Relative risk is the proportion of people experiencing an outcome in the intervention group divided by the proportion of people experiencing the outcome in the control group (see NCCMT Glossary).
  • With continuous outcome variables, such as blood pressure, mean effect size or mean difference is calculated, which may be weighted or unweighted (i.e., weighted mean difference WMD).
  • If there is no difference between the intervention and control group, OR or RR=1 or WMD=0 (the line of no effect).
  • Confidence intervals (CI) provides the range in which the true value for the population occurs (see NCCMT Glossary).
  • Studies with smaller CIs are more precise (have smaller ranges).
  • Studies are weighted by their sample size and CI.
  • Those studies with larger sample sizes and smaller CIs have greater weights (and bigger boxes).
  • The diamond is the overall result of the meta-analysis, with the centre of the diamond representing the overall effect estimate and the width of the diamond representing the overall CI.
  • When the diamond does not cross the line of effect, the difference between intervention and control groups is statistically significant.

Who is involved

Anyone who reads and interprets systematic reviews and meta-analyses would benefit from this method.

Conditions for Use

Not specified

Evaluation and Measurement Characteristics


Evaluation

Information not available

Validity

Not applicable

Reliability

Not applicable

Methodological Rating

Not applicable

Method/Tool Development


Developer(s)

Karin Ried

Method of Development

Information not available

Release Date

2006

Contact Person/Source

Karin Ried
The University of Adelaide
National Institute of Integrative Medicine
Australia
Email: karin.ried@adelaide.edu.au
Email: karinried@niim.com.au

These summaries are written by the NCCMT to condense and to provide an overview of the resources listed in the Registry of Methods and Tools and to give suggestions for their use in a public health context. For more information on individual methods and tools included in the review, please consult the authors/developers of the original resources.

Resources

Title of Primary Resource Interpreting and understanding meta-analysis graphs: A practical guide.
File Attachment None
Web-link
Reference Ried, K. (2006). Interpreting and understanding meta-analysis graphs: A practical guide. Australian Family Physician, 35(8), 635-638.
Type of Material Journal article
Format Periodical
Cost to Access None.
Language English
Conditions for Use Copyright © 2006 Australian Family Physician
Title of Supplementary Resource Evaluation of systematic reviews of treatment or prevention interventions.
File Attachment None
Web-link
Reference Ciliska, D., Cullum, N. & Marks, S. (2001). Evaluation of systematic reviews of treatment or prevention interventions. Evidence Based Nursing, 4, 100-104. doi: 10.1136/ebn.4.4.100
Type of Material Journal article
Format Periodical
Cost to Access Journal article purchase
Language English
Conditions for Use Not specified
Title of Supplementary Resource Understanding and interpreting systematic reviews and meta-analyses. Part 2: meta-analyses.
File Attachment None
Web-link
Reference Freemantle, N. & Geddes, J. (1998). Understanding and interpreting systematic reviews and meta-analyses. Part 2: meta-analyses. Evidence-Based Mental Health, 1(4), 102-104. doi: 10.1136/ebmh.1.4.102
Type of Material Journal article
Format Periodical
Cost to Access Journal article purchase
Language English
Conditions for Use Not specified

Have you used this resource? Share your story!