Intention-to-treat analysis of South African e-publications

Abstract

Background: The aim of this study was to determine whether randomised controlled trials in the South African (SA) e-publications utilised the intention-to-treat (ITT) principle as recommended by the CONSORT statement and the Cochrane Handbook. ITT analysis is a strategy that is adopted in order to address the issue of protocol violators where subjects are analyzed according to their assigned treatment group regardless of actual adherence to the treatment protocol.\ Methods: Since the SA e-publication service is the most comprehensive searchable collection of full text electronic South African Journals that are made available online, a search strategy was designed to identify randomized controlled trials in all areas of clinical research. An inclusion and exclusion criteria similar to the systematic review approach was then developed to select the final set of publications to be considered in the study. Results: Of the 84 studies that were identified online in the SA e-Publications, 26 randomised controlled trials met the inclusion/exclusion criteria. ITT analysis was mentioned in 4 (15%) of the studies. 12 (46%) utilised the principle but did not mention it since all cases initially randomised were analysed. 10 (39%) did not mention or use this principle. Conclusion: The general trend of reporting in the SA e-publications with regard to the ITT principle and handling of the protocol violators is explored in this study with the above findings. From these findings, many researchers face challenges in incorporating protocol violators in their statistical analysis. As researchers keen in improving the quality of reporting, we recommend the adoption of the CONSORT statement in the analysis of randomized clinical trials.