Forest plot

<h2 id="example">Example</h2>

<p>This blobbogram is from an iconic <a href="/facts/Systematic_review/dGvq3BGx">systematic review</a>; it shows <a href="/facts/Clinical_trial/q8LMr832">clinical trials</a> of the <a href="/facts/Preterm_birth/szMos8tu">use of corticosteroids to hasten lung development</a> in pregnancies where a baby is likely to be <a href="/facts/Premature_birth/szMos8tu">born prematurely</a>. Long <i>after</i> there was enough evidence to show that this treatment saved babies' lives, the evidence was not widely known and the treatment was not widely used. After a systematic review made the evidence better-known, the treatment was used more, preventing thousands of pre-term babies from dying of <a href="/facts/Infant_respiratory_distress_syndrome/eGBl4uow">infant respiratory distress syndrome</a>. However, when the treatment was rolled out in lower- and middle-income countries, it was found that more pre-term babies died. It is thought that this may be because of the higher risk of infection, which is more likely to kill a baby in places with lower-quality medical care.<a class="footnote-ref" id="fnref:6" href="#fn:6"><sup>6</sup></a>  A 2017 systematic review states that there is "little need" for further research into the usefulness of the treatment in higher-income countries, but further research is needed on how best to treat lower-income and higher-risk mothers, and optimal dosage.<a class="footnote-ref" id="fnref:7" href="#fn:7"><sup>7</sup></a>
</p>
<h2 id="reading-a-forest-plot">Reading a forest plot</h2>
<h3>Study identities</h3>
<p>Studies included in the meta-analysis and incorporated into the forest plot will generally be identified in <a href="/facts/Chronological_order/QnyvW82G">chronological order</a> on the left hand side by author and date.  There is no significance given to the vertical position assumed by a particular study.
</p>
<h3>Standardized mean difference</h3>
<p>The chart portion of the forest plot will be on the right hand side and will indicate the mean difference in effect between the <a href="/facts/Scientific_control/FXHfQwHx">test and control groups</a> in the studies.  A more precise rendering of the data shows up in number form in the text of each line, while a somewhat less precise graphic representation shows up in chart form on the right.  The vertical line (<a href="/facts/Y-axis/lkTqvMmB">y-axis</a>) indicates no effect.  The horizontal distance of a box from the y-axis demonstrates the difference between the test and control (the experimental data with control data subtracted out) in relation to no observable effect, otherwise known as the magnitude of the experimental effect.
</p>
<h4>Confidence interval whiskers</h4>
<p>The thin horizontal lines—sometimes referred to as whiskers—emerging from the box indicate the magnitude of the <a href="/facts/Confidence_interval/NS8mG5UE">confidence interval</a>.  The longer the lines, the wider the confidence interval, and the less reliable the data.  The shorter the lines, the narrower the confidence interval and the more reliable the data.
</p><p>If either the box or the confidence interval whiskers pass through the y-axis of no effect, the study data is said to be <a href="/facts/Statistical_significance/PMxjdlAu">statistically insignificant</a>.
</p>
<h4>Weight</h4>
<p>The meaningfulness of the study data, or <a href="/facts/Statistical_power/5KU54nW1">power</a>, is indicated by the weight (size) of the box.  More meaningful data, such as those from studies with greater <a href="/facts/Sample_size_determination/sf8dAFzq">sample sizes</a> and smaller confidence intervals, is indicated by a larger sized box than data from less meaningful studies, and they contribute to the pooled result to a greater degree.
</p>
<h3>Heterogeneity</h3>
<p>The forest plot is able to demonstrate the degree to which data from multiple studies observing the same effect overlap with one another.  Results that fail to overlap well are termed <a href="/facts/Homogeneity_(statistics)/gok3LHv3">heterogeneous</a> and is referred to as the <a href="/facts/Homogeneity_(statistics)/gok3LHv3">heterogeneity</a> of the data—such data is less conclusive.  If the results are similar between various studies, the data is said to be <a href="/facts/Homogeneity_(statistics)/gok3LHv3">homogeneous</a>, and the tendency is for these data to be more conclusive.
</p><p>The heterogeneity is indicated by the I2.  A heterogeneity of less than 50% is termed low, and indicates a greater degree of similarity between study data than an I2 value above 50%, which indicates more dissimilarity.
</p>
<h2 id="see-also">See also</h2>
<ul><li><a href="/facts/Galbraith_plot/AGS0V9wm">Galbraith plot</a></li>
<li><a href="/facts/Funnel_plot/F0yNHFrK">Funnel plot</a></li>
<li>The <a href="/facts/Cochrane_(organisation)/UlmdJ1H8">Cochrane (organisation)</a> logo is a forest plot</li>
<li>A forest plot helps determine whether <a href="/facts/Further_research_is_needed/Fycbz9jF">further research is needed</a></li></ul>

<h2 id="external-links">External links</h2>
<ul><li><a href="http://www.meta-analysis-made-easy.com">MIX 2.0</a> – software to perform meta-analysis and create forest plots in Excel.</li>
<li><a href="http://www.epigear.com">MetaXL</a> – Software that can create forest plots and run bias adjusted meta-analyses</li></ul>

<h2 id="references">References</h2>

<ol>
<li id="fn:1"><p>Lalkhen, AG (2008). "Statistics V: Introduction to clinical trials and systematic reviews". Continuing Education in Anaesthesia, Critical Care & Pain. 8 (4): 143–146. doi:10.1093/bjaceaccp/mkn023. <a href="https://doi.org/10.1093%2Fbjaceaccp%2Fmkn023" target="_blank">https://doi.org/10.1093%2Fbjaceaccp%2Fmkn023</a> <a href="#fnref:1" class="footnote-back-ref">↩</a></p></li>
<li id="fn:2"><p>Larry V. Hedges and Ingram Olkin (1985). Statistical Methods for Meta-Analysis. Orlando: Academic Press. ISBN 978-0-12-336380-0. <a href="978-0-12-336380-0" target="_blank">978-0-12-336380-0</a> <a href="#fnref:2" class="footnote-back-ref">↩</a></p></li>
<li id="fn:3"><p>Bijnens L, Collette L, Ivanov A, Hoctin Boes G, Sylvester R (1996). Can the forest plot be simplified without losing relevant information in meta-analyses? Communication at the meeting of the SCT, Pittsburgh, Pennsylvania, 5–8 May 1996. Controlled Clinical Trials 17(2S): 124. <a href="#fnref:3" class="footnote-back-ref">↩</a></p></li>
<li id="fn:4"><p>Steff Lewis & Mike Clarke (June 2001). "Forest plots: trying to see the wood and the trees". BMJ. 322 (7300): 1479–1480. doi:10.1136/bmj.322.7300.1479. PMC 1120528. PMID 11408310. <a href="/w/index.php?title=Steff_Lewis&action=edit&redlink=1" target="_blank">/w/index.php?title=Steff_Lewis&action=edit&redlink=1</a> <a href="#fnref:4" class="footnote-back-ref">↩</a></p></li>
<li id="fn:5"><p>Steff Lewis & Mike Clarke (June 2001). "Forest plots: trying to see the wood and the trees". BMJ. 322 (7300): 1479–1480. doi:10.1136/bmj.322.7300.1479. PMC 1120528. PMID 11408310. <a href="/w/index.php?title=Steff_Lewis&action=edit&redlink=1" target="_blank">/w/index.php?title=Steff_Lewis&action=edit&redlink=1</a> <a href="#fnref:5" class="footnote-back-ref">↩</a></p></li>
<li id="fn:6"><p>Iain Chalmers (October 4, 2016). "Should the Cochrane logo be accompanied by a health warning?". Evidently Cochrane. <a href="http://www.evidentlycochrane.net/cochrane-logo-health-warning/" target="_blank">http://www.evidentlycochrane.net/cochrane-logo-health-warning/</a> <a href="#fnref:6" class="footnote-back-ref">↩</a></p></li>
<li id="fn:7"><p>Roberts, Devender; Brown, Julie; Medley, Nancy; Dalziel, Stuart R (21 March 2017). "Antenatal corticosteroids for accelerating fetal lung maturation for women at risk of preterm birth". Cochrane Database of Systematic Reviews. doi:10.1002/14651858.CD004454.pub3. hdl:2292/34738. <a href="/wiki/Doi_(identifier)" target="_blank">/wiki/Doi_(identifier)</a> <a href="#fnref:7" class="footnote-back-ref">↩</a></p></li>
</ol>

Forest plot open-in-new

Forest plot