The epidemiological study on registered cases of sudden infant death syndrome (SIDS) in Tokyo: examination of the effect of autopsy on diagnosis of SIDS and the mortality statistics in Japan.

In the United States and most of European countries, a diagnosis of sudden infant death syndrome (SIDS) may be given only after an autopsy has been performed. Under the new definition of SIDS in Japan, an autopsy is now mandatory for the diagnosis of SIDS. However, according to the official records on autopsies, the proportion of autopsy for sudden infant death in Japan is still low (less than 30%). If a physician suspects SIDS from a review of the patient's medical history and medical findings, he can write 'suspected SIDS' as the cause of death on the death certificate without performing an autopsy. Such a clinical diagnosis is entered in the Vital Statistics section by the Japanese Ministry of Health and Welfare. In this report, a comparative epidemiological survey of registered cases of SIDS--after autopsy and with no autopsy--was carried out by examining the data from the death certificates registered by the Japanese Ministry of Health and Welfare (vital statistics in Tokyo from January 1979 to December 1996). There were 369 cases of SIDS registered in Tokyo. We found 247 diagnosed after autopsy (66.9%) and 122 with no autopsy (33.1%). The following epidemiological variables were used: address of the deceased (a specific area in Tokyo), sex, year of death, time of death, month of death, age at death, occupation of householders, and place of death. There were epidemiological differences at the 0.05 significance level between registered cases diagnosed after autopsy and those diagnosed without autopsies, as follows: year (P=0.016) and place of death (P=0.037). In addition, there were slight epidemiological differences at the 0.10 significance level between registered cases diagnosed after autopsy and with no autopsy, as follows: month of death (P=0.076) and age at death (P=0.082). This suggests that the quality of diagnosis of SIDS is not completely guaranteed. With respect to the area of residence, the incidence of SIDS is high in those areas where autopsy is performed frequently. In Tokyo, the medical examiner system is enforced only in the urban area and there is a possibility that SIDS is being underdiagnosed in the rural area of the Metropolitan Tokyo. It is likely that the diagnosis of SIDS without autopsy will influence the quality of SIDS diagnoses. The administrative inadequacy in the autopsy system in Japan should be corrected to improve the accuracy of SIDS diagnosis.