Back

Micro-costing for national-scale azithromycin mass drug administration to improve child survival in Niger

Peterson, B.; Nguyen, W.; Haroun, L. M.; Oumarou, F.; Arzika, A. M.; Maliki, R.; Amza, A.; Alio, K.; Gallo, N.; Aichatou, B.; Sara, I. I.; Beidi, D.; Kahn, J. G.; Bertozzi, S. M.; Lebas, E.; Arnold, B. F.; Lietman, T. M.; OBrien, K. S.; Fitzpatrick, M. C.

2026-02-11 health economics
10.64898/2026.02.09.26345952 medRxiv
Show abstract

As programs for azithromycin mass drug administration to reduce child mortality have begun in some parts of West Africa, it is imperative to understand their financial costs. We combined a micro-costing framework and observations from an implementation-focused sub-study within the AVENIR trials in 80 communities in the Dosso region of Niger to estimate the national health sector costs of a scaled-up programmatic approach for azithromycin biannual distribution to children aged 1-59 months of age living in nonurban areas, using the door-to-door modality. Our outcomes of interest were the annual budget at the regional and national levels for Niger and the cost per dose delivered. We found that the annual national budget required for azithromycin mass drug administration (MDA) achieving 90% average coverage would be $12.5M (95% Uncertainty Interval (UI) $12.2M, $13.0M) translating to $1.59 (95% UI $1.40, $2.30) per dose delivered. Across regions, cost per dose would vary from $1.17 (95% UI $1.03, $1.69) to $3.61 (95% UI $3.20, $5.16), with higher cost per dose expected for more sparsely populated regions. Training costs represented a large fraction (16.4%) of total costs, and integration of training with that for existing health interventions may provide opportunities for efficiency.

Published in PLOS Global Public Health (predicted rank #1) · training set

Matching journals

The top 2 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.