Despite dramatic advances in human health that have occurred during the 20th century, the end of the century still sees many places in the world with high child mortality rates. This is made worse by increasing inequity, such that there are still many communities in the world in which over 30% of children die before their fifth birthday. Estimates of the global burden of childhood pneumonia are based on the assumption that there is a predictable relationship between the childhood mortality rate and the proportion of that mortality that is attributable to pneumonia. As most child deaths occur at home and can only be investigated by verbal autopsy techniques, these estimates are very crude and provide only a guide to the overall burden of pneumonia. Recent estimates from the World Health Organization suggest that 1.9 million children die as a result of acute respiratory infection (ARI), mainly pneumonia, each year. For a number of reasons, this is likely to be an underestimate. Estimates of the morbidity burden attributable to pneumonia are also very approximate, as studies have used different and nonstandardized definitions of pneumonia. These estimates were originally used to assist with planning of ARI intervention activities and for advocacy to draw attention to the problem of ARI. Recently, the introduction of new vaccines against Haemophilus influenzae type b (Hib) and Streptococcus pneumoniae (pneumococcus) raised the prospect of prevention of pneumonia by vaccination. For reasons outlined in this paper, great caution must be exercised before using existing pneumonia burden estimates to predict mortality savings that may accompany the introduction of these vaccines into developing countries.