Fig. 2: Heterogeneous effects of MDA during the intervention and postintervention years.

Point estimates represent adjusted incidence ratios from subgroup analyses comparing the impact of MDA on malaria incidence during the intervention year (left) and the postintervention year (right) across potential spatial, environmental and demographic effect modifiers. For commune subgroup analyses, incidence ratios compare the incidence between the MDA arm and the control arm. Other effect modifiers were coded as binary indicators for values above or below the median in a given year. Analyses were performed at the cluster level. The analysis dataset included the preintervention, intervention and postintervention years (n = 3 years), with 60 clusters (30 per arm) per year (n = 180 clusters in total). Horizontal bars indicate 95% CIs computed using robust standard errors. In the postintervention year, the lower bound for Sinthiou Maleme was imputed for clarity of visualization; the estimated value was −104%. Incidence ratios were adjusted for the timing of case detection relative to the full PECADOM+ scale-up, as in the original trial analysis.