Medical research conducted over the years has proven that well-conducted systematic reviews with meta-analysis of randomized controlled trials provide a good source of evidence of intervention effects. This study aims to examine the methodological quality of individual participant data (IPD) meta-analysis and identify the potential areas of improvement.
This systematic review includes 323 IPD meta-analyses comprising 21 clinical areas. These included 270 non-Cochrane reviews, of which 269 were published in top-quarter journals. Eligible reviews were systematic reviews with IPD meta-analyses of randomized controlled trials. The primary outcome of the study was the methodological quality of IPD meta-analyses.
The meta-analyses indicated low compliance in assessing the risk of bias in randomized controlled trials. This elevated the risk of bias when analyzing results. Researchers found an excluded list of studies with justifications, pre-specifying the methods for examining participant-intervention interactions and overall effects. The findings further suggested that 126 IPD meta-analyses (39%) failed to obtain IPD from 90% or more trial participants, of which only 60 (48%) provided reasons, and 21 (17%) accounted for unavailability.
This systematic review concluded that the methodological quality of IPD meta-analyses was unsatisfactory and required an established a priori protocol with pre-defined data analysis systems.
Ref: https://www.bmj.com/content/373/bmj.n736
