Abstract

STUDY OBJECTIVE--To evaluate new national influenza vaccination practices which were started in 1987 under a revised law in Japan. DESIGN--This was a three year, nonrandomised cohort study with information collected by questionnaire between 1989 and 1991. SETTING--Eight primary schools in the city of Yonago, Tottori, Japan. These schools were selected from 23 schools in the city. PARTICIPANTS--Altogether 4251 pupils (1355 boys and 2896 girls) in years 1-4 of the eight primary schools were included in this study, and followed up. Three years later, data for 1619 pupils (768 boys and 851 girls) were obtained and analysed. MAIN RESULTS--The one-winter seasonal incidence rates of influenza-like disease were 13.4%, 29.9%, and 10.3% in 1989, 1990, and 1991 respectively. The incidence rate of influenza-like disease in fully vaccinated pupils was significantly lower than that in unvaccinated pupils in 1990, but not in 1989 or 1991. Stepwise multiple regression analysis showed that the incidence of influenza-like disease had a statistical relationship with the frequency of vaccination and the school year (R2 was 0.0148). Standardised parameters of the frequency of vaccination and the school frequency of vaccination and the school year were -0.089 and -0.080 respectively. CONCLUSIONS--The preventive effects of influenza vaccine are not strong. There must be some unknown factors that affect the incidence of influenza. This vaccine is useful for pupils in the early school years who seem to have less resistance. All pupils should not be inoculated with the vaccine to reduce influenza transmission in the community or school.