Find in Library

<p>Abstract</p> <p>Background</p> <p>The heterogeneity statistic <it>I</it>², interpreted as the percentage of variability due to heterogeneity between studies rather than sampling error, depends on precision, that is, the size of the studies included.</p> <p>Methods</p> <p>Based on a real meta-analysis, we simulate artificially 'inflating' the sample size under the random effects model. For a given inflation factor <it>M </it>= 1, 2, 3,... and for each trial <it>i</it>, we create a <it>M</it>-inflated trial by drawing a treatment effect estimate from the random effects model, using <inline-formula><m:math name="1471-2288-8-79-i1" xmlns:m="http://www.w3.org/1998/Math/MathML"><m:semantics><m:mrow><m:msubsup><m:mi>s</m:mi><m:mi>i</m:mi><m:mn>2</m:mn></m:msubsup></m:mrow><m:annotation encoding="MathType-MTEF"> MathType@MTEF@5@5@+=feaagaart1ev2aaatCvAUfKttLearuWrP9MDH5MBPbIqV92AaeXatLxBI9gBaebbnrfifHhDYfgasaacPC6xNi=xH8viVGI8Gi=hEeeu0xXdbba9frFj0xb9qqpG0dXdb9aspeI8k8fiI+fsY=rqGqVepae9pg0db9vqaiVgFr0xfr=xfr=xc9adbaqaaeGaciGaaiaabeqaaeqabiWaaaGcbaGaem4Cam3aa0baaSqaaiabdMgaPbqaaiabikdaYaaaaaa@2FBE@</m:annotation></m:semantics></m:math></inline-formula>/<it>M </it>as within-trial sampling variance.</p> <p>Results</p> <p>As precision increases, while estimates of the heterogeneity variance <it>τ</it>²remain unchanged on average, estimates of <it>I</it>²increase rapidly to nearly 100%. A similar phenomenon is apparent in a sample of 157 meta-analyses.</p> <p>Conclusion</p> <p>When deciding whether or not to pool treatment estimates in a meta-analysis, the yard-stick should be the clinical relevance of any heterogeneity present. <it>τ</it>², rather than <it>I</it>², is the appropriate measure for this purpose.</p>