2.12. Statistical Heterogeneity

Statistical heterogeneity refers to differences in the effects of the exposure on study outcomes. The heterogeneity between included studies will be determined by visual inspection of forest plots and the calculation of the I² statistic, which provides a numerical estimate of heterogeneity [42].

The I² statistic will be calculated using RevMan [43]. The I² statistic describes the percentage of the total variation across studies that is due to heterogeneity rather than chance, and results lie between 0% and 100%. The level of heterogeneity increases with an increase in value. Conventionally, a value of 0% indicates no heterogeneity, 25% low, 50% moderate, and 75% high heterogeneity [44].

If I² is between 50% and 75%, we will undertake a meta-analysis but include a caveat to let the reader know that we had concerns about the reliability of our results. For I² > 75%, the results will not be pooled [44].