Getting More out of Meta-Analyses- A new approach to meta-analysis in light of unexplained heterogeneity

Meta-analyses sometimes summarize results in the presence of substantial unexplained between-study heterogeneity. As GRADE criteria highlight, unexplained heterogeneity reduces certainty in the evidence, resulting in limited confidence in the extent to which population average effect estimates apply to a specific group of patients. To address this problem, we propose that when researchers apply random-effect models, they should provide an analysis of the distribution of the effect that jointly considers the mean and standard-deviation parameters.

Source: Journal of Clinical Epidemiology - December 5, 2018 Category: Epidemiology Authors: Amit Saad, Daniel Yekutieli, Shaul Lev-Ran, Raz Gross, Gordon Guyatt Tags: Original Article Source Type: research