Change to coverage with each additional round of MDA.

Coverage increased with successive rounds of MDA, from a 3.5 percentage point increase for a second round compared to the first, to approximately ten percentage points in rounds 10–12 relative to round one (Fig 2). A hiatus, where more than a year of no LF MDA passed before a given MDA began, was not significantly associated with coverage. This included districts implementing after a planned hiatus (e.g., a scheduled LF pre-Transmission Assessment Survey) or unplanned hiatus (e.g., a health worker strike preventing MDA implementation). Conducting the MDA in the same calendar month as the year before was associated with 1.6 percentage points of improved coverage. Where records showed an MDA was focal, (defined for this study as targeting less than 75% of the district population), this was associated with significantly lower epidemiological coverage (8.4 percentage points, p<0.001). Note that epidemiological coverage was calculated using a district’s full population in the denominator regardless of the number of persons targeted by MDA organizers.