In a meta-analysis, the I-squared statistic does not tell us how much the effect size varies


      • The vast majority of meta-analyses published in the fields of medicine and epidemiology use the I-squared statistic to quantify the amount of heterogeneity.
      • While this interpretation of I-squared is ubiquitous, it is nevertheless a fundamental mistake. I-squared does not tell us how much the effect size varies.
      • The statistic that does tell us how much the effect size varies is the prediction interval.


      To read this article in full you will need to make a payment

      Purchase one-time access:

      Academic & Personal: 24 hour online accessCorporate R&D Professionals: 24 hour online access
      One-time access price info
      • For academic or personal research use, select 'Academic and Personal'
      • For corporate R&D use, select 'Corporate R&D Professionals'


      Subscribe to Journal of Clinical Epidemiology
      Already a print subscriber? Claim online access
      Already an online subscriber? Sign in
      Institutional Access: Sign in to ScienceDirect


        • Higgins J.P.
        • Thompson S.G.
        Quantifying heterogeneity in a meta-analysis.
        Stat Med. 2002; 21: 1539-1558
        • Higgins J.P.
        • Thompson S.G.
        • Deeks J.J.
        • Altman D.G.
        Measuring inconsistency in meta-analyses.
        BMJ. 2003; 327: 557-560
        • Sorita A.
        • Ahmed A.
        • Starr S.R.
        • Thompson K.M.
        • Reed D.A.
        • Prokop L.
        • et al.
        Off-hour presentation and outcomes in patients with acute myocardial infarction: systematic review and meta-analysis.
        BMJ. 2014; 348: f7393
        • Holst L.B.
        • Petersen M.W.
        • Haase N.
        • Perner A.
        • Wetterslev J.
        Restrictive versus liberal transfusion strategy for red blood cell transfusion: systematic review of randomised trials with meta-analysis and trial sequential analysis.
        BMJ. 2015; 350: h1354
        • Borenstein M.
        • Higgins J.P.
        • Hedges L.V.
        • Rothstein H.R.
        Basics of meta-analysis: I2 is not an absolute measure of heterogeneity.
        Res Synth Methods. 2017; 8: 5-18
        • Borenstein M.
        Research note: in a meta-analysis, the I2 index does not tell us how much the effect size varies across studies.
        J Phys. 2020; 66: 135-139
        • Higgins J.P.
        Commentary: heterogeneity in meta-analysis should be expected and appropriately quantified.
        Int J Epidemiol. 2008; 37: 1158-1160
        • IntHout J.
        • Ioannidis J.P.A.
        • Rovers M.M.
        • Goeman J.J.
        Plea for routinely presenting prediction intervals in meta-analysis.
        BMJ Open. 2016; 6: e010247
        • Bailar 3rd, J.C.
        The promise and problems of meta-analysis.
        N Engl J Med. 1997; 337: 559-561
        • Borenstein M.
        Common mistakes in meta-analysis and how to avoid them.
        Biostat, Inc., New Jersey2019