Information on harm outcomes was insufficiently reported in systematic reviews and meta-analyses of individual participant data

Abstract

Meta-analysis is an important tool for summarizing findings from scientific research. But for harm outcomes,issues may arise due to the distribution problems, zero-events problems, as well as low statistical power. Individual participant data (IPD) meta-analysis represents a viable approach to overcome these shortcomings because all eligible small-scale studies can be treated as one large trial with multiple clusters through a one-stage analytic. An increasing number of IPD meta-analyses of healthcare interventions are published over the past years.6 Findings from these IPD meta-analyses contribute to the evolving body of best available evidence for informed decision-making. However, quality of safety reporting by the emerging IPD meta-analyses remains to be ascertained. In this study, we appraised the reporting of IPD meta-analyses, with a special focus on safety outcomes.