Poisson-model analysis of the risk of vaccine-associated paralytic poliomyelitis in Japan between 1971 and 2000.

This study estimates the risk of vaccine-associated paralytic poliomyelitis (VAPP) in Japan between 1971 and 2000. We acquired data regarding the number of VAPP cases from the website of the Ministry of Health, Labour and Welfare, and we estimated the number of oral poliovirus vaccines (OPV) administered based on the reported immunization data. Risk was calculated as the ratio between the number of VAPP cases and the number of OPV doses administered. Both the Runs test and the Poisson model were used to analyze the occurrence of VAPP. Thirty-three cases of VAPP were recorded in Japan between 1971 and 2000; approximately one case occurred per year. There were no statistical changes in temporal trends as regards the occurrence of VAPP between 1971 and 2000. The overall risk for VAPP, including both recipient and contact VAPP, was one case per 2.0 million OPV doses administered. The risk of recipient VAPP was one per 3.7 million doses, among which the first dose posed a much higher risk of one per 2.3 million than that of the subsequent dose. These data indicated that the occurrence of VAPP is rare, but the risk has remained constant for as long as OPV has been used in Japan.