Abstract
Provision of precise estimates of number of cattle infected as accurate as possible is important for dealing with control measures of bovine spongiform encephalopathy (BSE) in Japan. We extended backcalculation in order not only to estimate but also to interpret its present epidemiology. Results in previous studies in the UK were used for critical probability density functions such as the incubation period of BSE, age dependent susceptibility, and survivorship function of cattle. Based on eleven confirmed BSE cases, including and excluding two atypical cases, four possible trajectories, which proposed between two to eighteen cases until the year 2010 with reasonable 95 % prediction intervals, were obtained under different assumptions. Through the use of the age-stratified method, it was considered the major feed risk had been concentrated mainly in 1996 and thereafter was at a much reduced level. Technical matters in the application of backcalculation for a highly sparse record were discussed.
