The goal of this study was to determine if vitamin A supplementation decreased the duration or severity of pneumonia. Children from 3 mo to 10 yr of age without underlying metabolic disease, tuberculosis or asthma were eligible. Forty-seven children were randomized to the vitamin A group and 48 to the placebo group. Subjects ≥ 1 yr received 200,000 IU of water-miscible retinol on admission and 100,000 IU the next day. Subjects < 1 yr received half this dose. Analysis of clinical variables during hospitalization by multivariate ANOVA and logistic regression revealed that oxygen saturation was lower in the vitamin A group (p = 0.003) while respiratory rate (p < 0.001), heart rate (p = 0.003), use of supplemental oxygen (p < 0.001) and an aggregate severity score (p < 0.001) were higher. Duration of hospitalization was not affected. When adjustments were made for severity of disease and weight-for-height on admission, duration of disease before admission, and clinical diagnosis of viral vs. bacterial pneumonia, oxygen saturation was again lower in the vitamin A group (p = 0.028) and the use of supplemental oxygen greater (p = 0.0008). High-dose vitamin A supplements are thus not recommended for children recovering from pneumonia.