Interannual variation of the incidence of Haemophilus influenzae type b meningitis.

We conducted a population-based retrospective cohort study to define the annual age-specific incidence of Haemophilus influenzae type b meningitis for birth to 5-year-old residents of King County, Washington, from January 1977 through December 1986. We found naturally occurring wide interannual variations in incidence. The standard deviations of the age-specific incidence during the eight years before the introduction of H influenzae vaccine varied from 26% to 115% of the mean. If short-term changes in incidence were used to assess the efficacy of an H influenzae vaccine (which is less than 100% efficacious and not administered to all susceptible children), the conclusions could be erroneous. To avoid this bias, long-term cohort studies, case/control studies using concurrent controls, or large clinical trials are better choices. We found no significant change in overall incidence during the ten-year study period.