Trial Sequential Analysis for dichotomous outcomes - a practical guide for systematic review protocols.

BACKGROUND

Trial Sequential Analysis (TSA) is a statistical method to control random errors in systematic reviews with meta-analyses of randomised clinical trials. In our results from the Major Mistakes and Errors in Trial Sequential Analysis (METSA) project, we systematically assessed the use of TSA across all medical fields and found significant mistakes in the preplanning and reporting of most TSAs. This article provides a practical guide for authors of systematic review protocols on what to consider when planning Trial Sequential Analysis for dichotomous outcomes.

METHODS

This practical guide has been developed based on the TSA manual, the recommendations published previously by Jakobsen and colleagues and Wetterslev and colleagues along with the findings from our recently published results from the METSA project.

RESULTS

The following five parameters should be clearly defined in a publicly available protocol before the review is undertaken: 1) the proportion of participants with an event in the control group; 2) the relative risk reduction or increase in the experimental group; 3) the risk of type I errors (alpha); 4) the risk of type II errors (beta); and 5) the diversity of the meta-analysis. Improving the planning and reporting of these parameters will improve the interpretation, reproducibility, and validity of Trial Sequential Analysis results used in systematic reviews.

CONCLUSIONS

We hope this practical guide will aid in improving pre-registration and reporting of TSAs of dichotomous outcomes within systematic review protocols with meta-analysis of randomised clinical trials in the future.