How is missing data handled in cluster randomized controlled trials? A review of trials published in the NIHR...

Cluster randomized controlled trials are increasingly used to evaluate the effectiveness of interventions in clinical and public health research. However, missing data in cluster randomized controlled trials can lead to biased results and reduce statistical power if not handled appropriately. This study aimed to review, describe and summarize how missing primary outcome data are handled in reports of publicly funded cluster randomized controlled trials.

This study reviewed the handling of missing data in cluster randomized controlled trials published in the UK National Institute for Health and Care Research Journals Library from 1 January 1997 to 31 December 2024. Data extraction focused on trial design, missing data mechanisms, handling methods in primary analyses and sensitivity analyses.

Among the 110 identified cluster randomized controlled trials, 45% (50/110) did not report or take any action on missing data in either primary analysis or sensitivity analysis. In total, 75% (82/110) of the identified cluster randomized controlled trials did not impute missing values in their primary analysis. Advanced methods like multiple imputation were applied in only 15% (16/110) of primary analyses and 28% (31/110) of sensitivity analyses. On the contrary, the review highlighted that missing data handling methods have evolved over time, with an increasing adoption of multiple imputation since 2017. Overall, the reporting of how missing data is handled in cluster randomized controlled trials has improved in recent years, but there are still a large proportion of cluster randomized controlled trials lack of transparency in reporting missing data, where essential information such as the assumed missing mechanism could not be extracted from the reports.

Despite progress in adopting multiple imputation, inconsistent reporting and reliance on simplistic methods (e.g. complete case analysis) undermine cluster randomized controlled trial credibility. Recommendations include stricter adherence to CONSORT guidelines, routine sensitivity analyses for different missing mechanisms and enhanced training in advanced imputation techniques. This review provides updated insights into how missing data are handled in cluster randomized controlled trials and highlight the urgency for methodological transparency to ensure robust evidence generation in clustered trial designs.