Which is the Preferred Measure of Heterogeneity in Meta-Analysis and Why? A Revisit

Abstract

Heterogeneity assessment is unavoidable under meta-analysis as it guides the path to choose appropriate synthesizing method. Its consideration further strengthens the drawn inferences. In the literature, there are various measures of heterogeneity, such as, Q statistic, H2 statistic, HM2 statistic, R2 statistic and I2 statistic. The present work mainly aimed to appraise all existing methods and find the preferred measure of heterogeneity in meta-analysis. Various heterogeneity measures were compared on the basis of theoretical criteria as well as their analytical results on the observed data from a systematic review of randomized controlled trials comparing neoadjuvant chemotherapy with adjuvant chemotherapy in the treatment of breast cancer patients. Theoretical criteria involve dependence on the extent of heterogeneity, invariance from scale and sample size and its interpretation. I2 statistic is a sample size and scale invariant measure, which increases with increased heterogeneity. Unlike other methods of measuring extent of heterogeneity, I2 has finite upper bound as well. Further, I2 has precise confidence interval among all measures of heterogeneity assessment. In addition, it has appealing interpretation in terms of the proportion of the total variability that is because of heterogeneity alone. With all these virtues, I2 is a preferred measure of heterogeneity in meta-analysis.