ABSTRACT
<p><b>OBJECTIVE</b>To describe the design and application of cluster randomized controlled method on typhoid Vi vaccine trial, and to assess the effect of implementation.</p><p><b>METHODS</b>Simple size calculation of cluster-randomized trial was used to determine the sample size of the two groups and a vaccination campaign was conducted. The study group was given typhoid Vi vaccine and the control group was given meningococcal A vaccine.</p><p><b>RESULTS</b>According to sample size calculation, a total sample of 96,121 participants was required and the study areas were divided into 108 clusters. In practice, 53 study clusters with 44,054 participants and 54 control clusters with 48,422 participants were stratified and matched according to size, location (urban or rural), characteristics (school, department, factory, demography) were randomized respectively. Confounding factors of two groups including age, sex, resident area, income, level of education were compared. It was found that the ratio of all confounding factors between the two groups were comparable and balanced.</p><p><b>CONCLUSION</b>Confounding factors can be better controlled between study group and the control group by applying cluster-randomized method on vaccine trail which enabled the intervention to be more scientifically evaluated; The implementation of cluster randomization trial was simple and easy to be accepted.</p>