Abstract
This paper examines the possibility of screening children of 6–59 months old for subsequent risk of dying by using various cut-off points of weight for age, weight for height, arm-circumference for age, and arm-circumference for height. The study was conducted in Kasongo, the main town of a rural area in Zaire. The risk of dying for Kasongo under-fives was quite uniformly distributed among groups defined by various cut-off points in the distribution of anthropometric parameters. It was not possible to demonstrate differential mortality risks or to identify a threshold level below which the risk would substantially increase. These results are in contrast with similar community-based studies from India and Bangladesh. Possible explanations for this contrast as well as the operational implications of the findings are discussed.