We conducted a randomized, double-blind, placebo-controlled trial to test the efficacy of the 14-valent pneumococcal capsular polysaccharide vaccine in 2295 high-risk patients (patients with one or more of the following: age above 55 years and the presence of chronic cardiac, pulmonary, renal, or hepatic disease, alcoholism, or diabetes mellitus). Seventy-one episodes of proved or probable pneumococcal pneumonia or bronchitis occurred among 63 of the patients (27 placebo recipients and 36 vaccine recipients). Vaccine-serotype Streptococcus pneumoniae strains were recovered in association with 11 infections in the placebo group and 14 infections in the vaccine group. Pneumococcal infections occurred most frequently among patients with chronic pulmonary, cardiac, or renal diseases. Among vaccine recipients who subsequently had vaccine-type pneumonia or bronchitis, the majority did not make or sustain serum antibodies against their infecting organism in concentrations that were twice as high as the base-line values, or more than 400 ng of antibody nitrogen per milliliter, although their base-line levels were higher than those in subjects in whom infection did not develop. We were unable to demonstrate any efficacy of the pneumococcal vaccine in preventing pneumonia or bronchitis in this population. Our data suggest that chronically ill patients, who are most susceptible to infection, may have an impaired immune response to the pneumococcal vaccine.