The Haemophilus influenzae type b polysaccharide vaccine was licensed for use in the United States in April 1985. Postlicensure case-control efficacy studies have yielded markedly different estimates of efficacy, leading to contradictory recommendations to practicing physicians. To obtain additional information about the efficacy of the vaccine, we studied cases of invasive Haemophilus influenzae type b disease ascertained through active surveillance in areas with a total population of 34 million. We enrolled children 24 to 59 months of age who did not attend day-care centers. (Data from our day-care study have been published elsewhere.) For each case child, as many as three 24- to 59-month-old control children were chosen from a roster of acquaintances supplied by the child's parent. Conditional logistic regression was used, and vaccine efficacy was estimated to be 62% (95% confidence interval = 0%, 85%), which did not change significantly after adjusting for age and parental smoking, variables that were significantly different for case and control children. Results of this study support our previous finding of a positive protective efficacy, albeit lower than the efficacy of 90% found in children 18 to 71 months of age in the Finnish prelicensure trial.