Key result
BACI estimator requires 54 clusters to detect a ~5-percentage-point gain in hypertension control.
Why the study?
Community health worker programmes support hypertension control, but visit logs recording contacts are frequently and non-randomly incomplete because patients lost to follow-up, disengaged, or transferred between facilities are systematically less likely to have visits recorded.
Provides a formal causal identification framework and sample size design characterization to account for non-random visit log incompleteness when evaluating community health worker hypertension outreach programmes.
Community health worker (CHW) programmes are widely used to support hypertension control, but the visit logs that record CHW contacts are frequently incomplete, and this incompleteness is not random: patients who are lost to follow-up, disengaged, or transferred between facilities are systematically less likely to have their visits recorded. The framework combines an augmented inverse-probability-weighted (AIPW) estimator with a completeness model that reweights the observed visit records to the target population, and it is evaluated through a replication-based design study of the before-after-control-impact (BACI) estimator under a declared variance structure (between-cluster SD = 6.0, residual SD = 4.5, intraclass correlation = 0.64). Power was estimated across 2000 design cells for cluster counts of 8, 16, 24 and 32 and effect sizes of 3, 5 and 8 percentage points. The design required 54 clusters to detect a 5-percentage-point effect at 80% power. The contribution is a formal identification result for the CHW visit effect under log incompleteness, together with a design-analysis characterisation of the BACI estimator that quantifies how cluster count and effect size jointly determine detectable effects in outreach-targeting evaluations.
No takes yet. Share an insight, caveat, or question.
Ceesay et al. (2026) studied Hypertension. Community health worker (CHW) visits evaluated via an augmented inverse-probability-weighted (AIPW) and before-after-control-impact (BACI) estimator was evaluated on Hypertension control (statistical power to detect CHW visit effect under variable log completeness). The before-after-control-impact (BACI) estimator required 54 clusters to detect a 5-percentage-point community health worker visit effect on hypertension control at 80% power (ICC = 0.64).
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: