Clinical trials to estimate the efficacy of preventive interventions against malaria in paediatric populations: a methodological review

BACKGROUND: Recent years have seen publication of a considerable number of clinical trials of preventive interventions against clinical malaria in children. There has been variability in the specification of end-points, case definitions, analysis methods and reporting and the relative lack of standardization complicates the ability to make comparative evaluations between trials. METHODS: To prepare for a WHO consultation on design issues in malaria vaccine trials, controlled trials of preventive interventions against malaria in children in endemic countries were identified in which clinical malaria, or death, had been one of the main end-points. Trials were included that evaluated the impact of vaccines, insecticide-treated bed nets (ITN), intermittent presumptive or preventive therapy in infants (IPTi) or, in one instance, vitamin A supplementation. Methods that had been used in these trials were summarized and compared in order to identify issues that were directly relevant to the design of malaria vaccine trials. RESULTS: 29 controlled trials of preventive malaria interventions were identified, of which eight were vaccine trials. Vaccine trials that were designed to detect an effect on clinical malaria all reported the incidence rate of first episodes of clinical malaria as their primary endpoint. Only one trial of a preventive intervention (of ITN) was identified that was designed to detect an effect on severe malaria. A group of larger trials were designed to detect an effect of impregnated bed nets or curtains on all-cause mortality as the primary end-point. Key methodological and reporting differences between trials are noted in the text. Two issues have been identified that are of some concern. Firstly, the choice of primary endpoint is not stated in the reports of a number of the trials and, secondly, the relationship between pre-specified analysis plans and trial reports is rarely made clear. CONCLUSION: This article reports an investigation into the ways in which trial design and reporting could be improved and standardized to enable comparative evaluation of the relative merits of malaria control measures, and specifically with respect to the design of malaria vaccine trials. The need for standardization of clinical trial design, conduct, analysis and reporting has been also affirmed as a priority area by the Malaria Vaccine Technology Roadmap.