A double blind placebo-controlled efficacy trial of two acellular pertussis vaccines was conducted in 3801 6− to 11-month-old children. Four vaccinated children died during 7 to 9 months follow-up as a result of Haemophilus influenzae type b meningitis, heroin intoxication with concomitant pneumonia, suspected septicemia, and Neisseria meningitidis Group B septicemia. From the actual death rate in children belonging to the same birth cohort in Sweden that could have been eligible for the trial, one death was expected among vaccinated children. Several investigations were carried out to examine the possibility that the deaths could be causally related to the vaccination. The relative risk for hospitalization due to systemic or respiratory infections was 1.07 (95% confidence interval, 0.95 to 1.20) and 0.83 (95% confidence interval, 0.64 to 1.08) in the vaccine groups as compared with the placebo group. Subsets of the population were studied for signs of immunosuppression. There was no indication of immunoglobulin deficiency or any sign of clinically significant leukopenia or lymphocytosis in vaccine recipients. The results of this analysis provide no evidence for a causal relation between vaccination with the studied acellular pertussis vaccines and altered resistance to invasive disease caused by encapsulated bacteria. The hypothesis that the two variables are related, however, cannot be refuted from these data.