Rural community health-worker rotation cycles and tuberculosis treatment adherence in Malawian districts: a difference-in-differences estimation of rotation-timing thresholds

Community health-worker rotation is a routine administrative practice in rural Malawian districts, yet the timing at which rotation begins to disturb tuberculosis treatment adherence has not been formally bounded. the present analysis treats the question as a design problem rather than an estimation problem: before any district-level evaluation is mounted, what cluster counts and effect magnitudes can a before-after-control-impact (BACI) design actually detect under the variance structure that rural adherence data typically exhibit? Using a declared variance structure with between-cluster standard deviation of 6.0, residual standard deviation of 4.5 and intraclass correlation of 0.64, power was characterised across 2000 replications per design cell for cluster counts of 8, 16, 24 and 32 and effect sizes of 3, 5 and 8 percentage points. The analysis indicates that 54 clusters are required to detect a 5 percentage-point rotation effect at 80% power, a figure well above the cluster counts typically available to a single district health office. The contribution is a pre-evaluation feasibility bound: rotation-timing thresholds below roughly 5 percentage points cannot be credibly identified with the cluster counts that Malawian district evaluations can realistically assemble, and the paper specifies what such evaluations can and cannot claim.