NOTES: NA = not provided; PE = protective efficacy. Incidence of episodes (malaria and hospitalizations) analyzed “with a negative binomial regression to account for potential clustering within individuals. Robust standard errors were used to account for intra-cluster correlation at the community level for the Navrongo trial.” Clinical malaria: history of fever with any P. falciparum parasitemia. Risk of anemia (at least one episode) was analyzed using a Poisson regression model with a log-link and a robust error variance. To assess the sensitivity of the results to any single study, the SWG analyzed each outcome omitting one trial at a time. Estimates rounded to the nearest whole percentage.
aLocally cut-offs:>500μ/l for Ifakara, Manhiça, and Kumasi; >600μ/l for Lambaréné; >5000μ/l for Tamale; >8000μ/l for Navrongo.
bAnemia: PCV<25% for Ifakara, Manhiça, and Kumasi; Hb<8g/dl for Kumasi, Lambaréné, and Tamale.
cCombined estimates are basedon random-effects meta-analyses.
dRange: lowest and highest estimated protective efficacy for the sensitivity analyses.
SOURCE: Compiled from data in the Report of the Statistical Working Group (IPTi Consortium, 2007b).