Estimates of efficacy for partial and full series of rotavirus vaccine against the most severe reported outcome of rotavirus gastroenteritis, by WHO mortality group.

aBecause vaccine efficacy against rotavirus deaths was not available, the model input was efficacy against the most severe reported form of rotavirus gastroenteritis in the clinical trial (≥11 denotes “severe” diarrhea and ≥15 denotes “very severe” diarrhea on 20 point Vesikari clinical scoring system).bNo decline in efficacy by age was reported by age for the very severe outcome, thus the efficacy estimate for children cThis trial measured efficacy during the first year of life. No estimates of efficacy were available against very severe disease that would serve as a better proxy for death (i.e., Vesikari ≥15) at these sites in Malawi and South Africa. Consistent with all other rotavirus efficacy trials where positive correlation exists between efficacy and severity score, it was assumed that efficacy in South Africa and Malawi would be higher against Vesikari score ≥15 than Vesikari ≥11. For the model, estimates of efficacy against “very severe” rotavirus diarrhea were from sites in Africa and Asia where these data were available [5],[6],[29].dBecause no partial vaccine efficacy estimates were available for Africa and Asia, we assumed that a proportional difference in efficacy between full and partial vaccination that was observed in high mortality country of Nicaragua [25].