Characteristics of communities in CHAT trial.

The objective of this study was to examine whether the effect of mass Azithromycin (AZ) distribution on all-cause mortality among children under 5 varies with seasonal malaria chemoprevention (SMC) administration season or coverage. This was a secondary analysis of the Community Health with Azithromycin Trial (CHAT), a cluster-randomized, placebo-controlled trial of twice-yearly AZ treatment in 341 communities in the Nouna District, Burkina Faso. All communities received SMC as standard-of-care. SMC administration and coverage data were provided from National Malaria Control Program. SMC season was defined as the period during and following SMC (July-December) versus the no SMC season (January-June). SMC coverage was assessed as proportion of the population covered and by whether it was below or above a threshold of 80%. We used Poisson regression models with person-time at risk as an offset and robust standard error to analyze mortality rates by treatment group and SMC subgroups and assessed interaction on both multiplicative and additive scales. Mortality was higher in SMC seasons for both arms. Compared to placebo, the mortality rate in AZ clusters was 0.77 (95% CI: 0.60 to 0.98) during SMC season, while it was 0.89 (95% CI: 0.68 to 1.15) during the non-SMC seasons. In clusters with