Background: Systematic reviews and meta-analyses of randomised controlled trials (RCTs) are as strong as the quality of the included studies. Study quality can be impaired in numerous ways. The data might not be available, as the trial results were never published (publication bias). Reports from identified trials might present an incomplete information or provide it in a format not useful for meta-analysis (reporting bias). Access to individual participant data (IPD) might not be a panacea to all the problems in the meta-analysis. Nevertheless, meta-analysis using IPD has the potential to reduce the bias due to selective or incomplete outcome reporting considerably.

Objectives: For the IPD meta-analysis on the effect of diet and physical activity-based interventions in pregnancy (i-WIP), we gained access to IPD from 36 RCTs. The aim of our work was to investigate the relationship between the outcome data published in the trials’ report and the data contributed to the i-WIP IPD meta-analysis.

Methods: We evaluated the availability of information for the main outcomes for the i-WIP IPD meta-analysis. In our work, we focused on the reporting of nine outcomes (five maternal and four fetal/neonatal) and compared it with the data available in the datasets from the relevant trial. The amount of information between two sources was compared formally.

Result and discussion: Access to IPD allows a reduction in bias arising from limited outcome reporting in the aggregate meta-analysis. We will provide a detailed description of our findings and their consequences based on experience in the i-WIP IPD meta-analysis.