Do nutrition programs make a difference? The case of Brazil.

Four Brazilian food and nutrition programs operating during some part of 1974-86 are evaluated for their effectiveness in curing or preventing infant and child malnutrition, including low birth weight when pregnant women were beneficiaries. Two programs distributed free food to identified clients: traditional commercial foods in one case and specially formulated supplements in the other. The other two programs subsidized four or more basic foodstuffs: one experiment quantitatively restricted a subsidy to identified families, and the other was unrestricted and open to all families patronizing certain shops. The programs were more effective at curing than at preventing malnutrition, and more effective at increasing weight than height. Many beneficiaries, even when initially underweight, showed no change, and some deteriorated despite the food transfer. Results were better after than during the first year of life, when deterioration is most likely. Donation programs including medical and educational components proved more effective than pure subsidies, showing that while poverty may be the chief cause of malnutrition, the problem should be seen as poor health rather than simply low food consumption. Evaluation also shows that programs were inefficient in transferring benefits, and that clients were deterred from participating by the costs of obtaining the food and its poor quality and small volume. Longer participation improved results, but more frequent participation in a given interval did not necessarily do so.