Reported are the results of a meta-analysis (12 large-scale field trials in seven countries) of the impact of vitamin A supplementation on pneumonia morbidity and mortality, undertaken as part of a wider review process of a range of possible potential interventions for the prevention of childhood pneumonia. The summary estimate of the relative risk for the impact of vitamin A supplementation on pneumonia incidence was 0.95 (95% confidence interval (CI) = 0.89, 1.01), and for pneumonia mortality, 0.98 (95% CI = 0.75, 1.28). This is in marked contrast to the substantial impact of vitamin A supplementation on all-cause mortality (combined rate ratio (RR) = 0.77, 95% CI = 0.71, 0.84), and on diarrhoea-specific and measles-specific mortality. There was no evidence for a differential impact on pneumonia mortality by age. Since the majority of pneumonia deaths occur in the first year of life, we complemented the paucity of data on pneumonia-specific mortality among this age group with a detailed examination of all-cause mortality among infants. The mortality reduction in the 6-11 month age group was consistent with that observed for older age groups (RR = 0.69; 95% CI = 0.54, 0.90), but there was no reduction for 0-5 month-olds (RR = 0.97; 95% CI = 0.73, 1.29).