A Seroepidemiological Study on Measles and Rubella in Japan Using Data Accumulated at a Commercial Diagnostic Laboratory：Effect of “the 4th Period”MR Vaccination

[Background]：In 2007-2008, measles outbreaks occurred country-wide mainly among young people in their teens and twenties. The government implemented a 5-year interim measles-rubella vaccination program to boost the immunity in the young, beginning from the 2008 fiscal year (FY, April to next yearʼs March). In the present paper, we made a seroepidemiological investigation into the effect among the vaccination target groups (third-year high school students) who were born in FY1990 to FY1994. By using the voluminous data of the antibody tests performed at a commercial diagnostic laboratory, we calculated “approximate” antibody prevalence rates (see below) for different birth cohorts, and were able to plot their yearly changes after the vaccination. [Methods]：The rubella antibody was determined with the hemagglutination-inhibition (HI) method, and the measles IgG antibody with an enzyme immunoassay (EIA) using a commercially available kit. The antibody data mainly from healthy college students and hospital workers were used for analysis. Antibodypositive proportions among each age group in each “summation year”(see below) were calculated. In addition, age-specific geometric mean rubella HI antibody titers were calculated. [Results]：1) Since the individual antibody data contain information on the personʼs age but not the month of birth, the antibody prevalence rate of a birth cohort in a certain year is not calculable. However,we found that, when age-specific antibody prevalence is calculated by summation of one-year data from the previous yearʼs October to September (summation year, SY), it reflects 75% of the data of the birth cohort. We then used this value as the birth cohort-specific “approximate” antibody prevalence in the one-year period. We then depicted the graphs for yearly changes in the approximate prevalence rates of each birth cohort. 2) We found that, with cutoff HI antibody titers of ≧8, the prevalence percentage for the FY1990-1994 birth cohorts (vaccination target) increased after the vaccination and remained at high levels. In contrast,the prevalence rates with cutoff HI titers of ≧32 soon decreased；this pattern of the yearly changes seemed similar to that of the changes in approximate mean antibody titers. However, the FY1980-1983 birth cohorts, who had experienced nationwide rubella epidemics in 1987, maintained relatively higher prevalence rates. We think that the latter group had stronger immune responses through natural infection. 3) We