Objective: To assess the effect of Nepal’s vitamin A supplementation programme on child mortality at age 12–59 months.

Materials and methods: Logistic regression, applied to retrospective data from Nepal’s 2001 Demographic and Health Survey.

Results: After a number of potentially confounding variables are controlled, the effect of 100% community-level vitamin A coverage since the child’s birth, relative to no coverage, is to reduce the odds of dying at age 12–59 months by slightly more than half (OR ¼ 0.47, P ¼ 0.03).

Conclusions: The estimated bene?cial effect of vitamin A supplementation on child mortality is larger than that found in most earlier clinical studies. This larger effect may be due mainly to the other healthrelated activities undertaken by the female community health volunteers who distribute vitamin A capsules.