OBJECTIVE: The aim of this study was to describe the development of the autopsy rate in stillbirths and infant deaths in an 11-year period and evaluate the information gained by performing an autopsy. METHODS: Included in the study were all stillbirths and infant deaths in Funen County, Denmark, in 1986-96. Data sources were death certificates and autopsy reports. RESULTS: The study included 273 stillbirths and 351 deaths in infancy. The rates of stillbirth and infant death did not change significantly during the period. The overall autopsy rate for stillbirths was 70% and for infant deaths 57%. There was a significant decline in autopsy rate during the years 1991-96 as compared with 1986-90 for stillbirths, infant deaths and infant deaths excluding sudden infant death syndrome. In stillbirth, the autopsy changed the diagnosis in 9% of the cases. In 22%, the clinical diagnosis was maintained, but additional information was obtained. In infant death, the numbers were 10% and 40%, respectively. CONCLUSION: In 10% of the autopsies the diagnosis was changed completely, with an impact on genetic counseling as well as on statistical records of causes of death in fetuses and infants. With additional information in 22-40% of the autopsies, the study emphasizes autopsy as a useful investigation.