Pertussis is a common serious illness of childhood that can be controlled by immunization. It is a unique disease in that it is clinically manifested more often in females than in males. In the 20th century the mortality from pertussis has decreased steadily in both the United Kingdom and the United States. This decline in death rate was well underway prior to the introduction of pertussis vaccine but was accelerated after vaccine use became widespread. In recent years the case fatality rate in the United States has been considerably greater than that in the United Kingdom. One obvious reason for this difference is the difference in age-specific attack rates in the two nations. Available data also suggest that recent pertussis deaths in infants in England and Wales may frequently be reported as due to respiratory diseases other than pertussis. Although it is frequently suggested by some observers, there is no evidence that the incidence of pertussis was declining prior to the widespread use of vaccine. All available evidence indicates that pertussis vaccine use in both the United Kingdom and the United States was responsible for a drastic reduction in the magnitude of both endemic and epidemic pertussis. Decreased utilization of pertussis vaccine in England and Wales beginning in 1975 resulted in two major epidemics of pertussis in 1977-1979 and 1982-1983. Moderate local and systemic reactions commonly occur following pertussis immunization. These reactions appear to be less common and less severe in the United Kingdom than in the United States, but in contrast to recent studies in the United States, there are no recent quantitative studies in the United Kingdom. There are virtually no data available in the United States on the incidence of serious neurologic disease resulting from pertussis immunization. In contrast, the recently published findings of the NCES, a case-control study of national scope, have allowed attributable risk estimates of serious neurologic disease to be made. Although the NCES was a carefully controlled study, its findings in regard to attributable risk were surprisingly similar to those of several other studies in which no allowance was made for background encephalopathy due to other causes.(ABSTRACT TRUNCATED AT 400 WORDS)