Estimating the year of eradication of tuberculosis in Japan

Abstract

The time of eradication of tuberculosis has been discussed for several countries, and based on those results, a new strategic plan and goals have been elaborated. Considering such developments, and in order to make a new tuberculosis control strategy, it is important to determine the point at which eradication of tuberculosis would be achieved in Japan. Styblo proposed the two conventional definitions of eradication of tuberculosis, namely that the incidence of smear-positive tuberculosis has fallen below 1 per million population or that the prevalence of tuberculosis infection in the general population has fallen below 1% and continues to decrease. The bacteriological results of new cases have been reported since 1975 in Japan. However, those results are still of doubtful validity and reliability. Therefore, the author estimated the year of eradication of tuberculosis, according to the criterion that tuberculosis is eradicated when the proportion of the population infected with tubercle bacilli is less than 1%. If the risk of infection is changing at a regular rate, it is possible to estimate the risk of infection at any time in the past and in the future. Once the risk of infection is determined, it is also possible to calculate the age-specific prevalence of infection and the proportion of the population infected with tubercle bacilli at various times in the past and in the future. In Japan, the risk of infection before World War II was assumed to be around 4% and not to vary with calendar year. And based on the data from the prevalence surveys in Okinawa in 1968 and 1973, the risk of infection was estimated 0.3% in 1968 and has declined on average, by 10 to 11% annually. At that time, Okinawa was the only area free from BCG vaccination in Japan. The incidence rate in Japan also has declined, on average, by 10% annually. However, since late 1970s, the annual speed of decline of the incidence rate has been slowed down. Therefore, I assumed that the recent trend of the infection risk is the same as the trend of the recent incidence rate among the 0-29 year age-group. The size of the effect of age on the risk of infection has been discussed. The author also considered age-effects in the model. The weight applied to the risk of infection by age was determined by examining the age-specific positive rate in the 1930s before the era when BCG vaccination was widely used.(ABSTRACT TRUNCATED AT 400 WORDS)