Background: Several published studies have recently shown a higher incidence of aseptic meningitis associated with Urabe AM-9 vaccine strain than that estimated previously. In France, where all mumps vaccines produced and marketed contain the Urabe AM-9 vaccine strain, little was known about the incidence of these side effects although several surveillance systems exist.
Methods: The capture-recapture method can be used to estimate the total number of cases of a disease when at least two independent data sources are available. Applying the capture-recapture method, data provided by the national network of hospital virology laboratories (EPIVIR) and by the Pharmacovigilance Department of the vaccine manufacturer (P.M.sv.) during a 3-year period between 1 January 1991 and 31 December 1993 have been used to obtain an estimate of the total number of aseptic meningitis cases associated with Urabe AM-9 vaccine.
Results: The number of such aseptic meningitis cases was estimated at 116 (95% confidence interval [CI]: 49-183). The risk of aseptic meningitis can thus be assessed as 1 case per 28,400 doses of Urabe AM-9 sold (95% confidence interval [CI]: 1 case per 18,000-1 case per 67,200). The sensitivity of the systems was 21% for EPIVIR and 23% for P.M.sv. The low number of cases recorded by either system makes it difficult to test the statistical independence of the two systems by the calculation of the correlation coefficient of the sensitivity of the systems after stratification. However, the stratified data by age, by sex, by geographical zone and by type of vaccine show that the totals after stratification are not very different from the crude total.
Conclusions: The French Vaccination Committee has recommended promoting the practice of measles-mumps-rubella vaccination in order to limit the circulation of the wild virus and to prevent an epidemic in insufficiently protected groups of the population, whilst awaiting vaccine containing the Jeryl-Lynn strain.