OBJECTIVE: To determine the effectiveness of the Wyeth heptavalent pneumococcal conjugate vaccine against clinical and radiograph-confirmed pneumonia in children. METHODS: The heptavalent CRM(197) pneumococcal conjugate vaccine (PCV) was given to infants at 2, 4, 6 and 12 to 15 months of age in a randomized, double blind trial. Children were randomized to receive either the CRM(197) PCV (vaccine group) or the meningococcal type C CRM(197) conjugate vaccine (control group). The primary outcome of this trial was invasive pneumococcal disease. In addition children with the clinical diagnosis of pneumonia in the study population were identified through review of automated inpatient, emergency and outpatient databases. The subset of the cohort of these children who had chest radiographs obtained at the time of diagnosis was identified, and the original reading of their radiographs by the radiologist was obtained from automated databases. Rates of clinically diagnosed pneumonia, of pneumonia with a radiograph obtained regardless of result, of pneumonia with positive radiograph (consolidation, empyema or parenchymal infiltrate) and of pneumonia with only perihilar infiltrates were compared between vaccinated and nonvaccinated groups. In addition risk of disease pneumonia was evaluated by race and ethnicity. RESULTS: The incidence of a first pneumonia episode in the control group was 55.9 per 1000 person-years. A radiograph was obtained in 61% of episodes, a positive radiograph in 21% and perihilar findings in an additional 5%. In per protocol follow-up of children given PCV, first episodes of all clinically diagnosed pneumonia were reduced by 4.3% [95% confidence interval (CI), -3.5, 11.5%, = 0.27], episodes with a radiograph were reduced by 9.8% (CI 0.1, 18.5%, < 0.05) and episodes with a positive radiograph were reduced by 20.5% (CI 4.4, 34.0, = 0.02). In the intent to treat analysis including all episodes after randomization, episodes with a positive radiograph were reduced by 17.7%, =.01). The greatest impact was in the first year of life with a 32.2% reduction and a 23.4% reduction in the first 2 years, but only a 9.1% reduction in children >2 years of age. Asians, blacks and Hispanics were at higher risk of pneumonia than were whites, but there was no evidence of ethnic variation in PCV effectiveness. Ten of the 11 cases of pneumococcal pneumonia with a positive blood culture were in the control group. CONCLUSION: The pneumococcal conjugate vaccine tested was effective in reducing the risk of pneumonia in young children.