Pertussis was a major childhood disease in parts of West Germany in 1990. Before a prospective household contact trial into the efficacy of an acellular pertussis vaccine was undertaken, the epidemiology of the disease was studied in one area. This study showed that pertussis had an annual incidence of 4%-6% in the first 6 years of life. During the trial, a total of 1223 pertussis cases were diagnosed by laboratory methods. Isolation rates of Bordetella organisms varied by the pediatrician who did the nasopharyngeal swab and decreased with time required for swab transport. Analysis of the serologic results showed that if IgG and IgA anti-pertactin antibodies had been considered, overall sensitivity would have been only marginally increased. Several pertussis patients had antibodies to parapertussis antigens, indicating a lack of complete cross-protection. The serologic response in secondary household contacts differed by vaccination status. However, this difference was unlikely to have influenced the point estimate of efficacy of the acellular vaccine.