The measles-mumps-rubella (MMR) vaccine has been postulated to cause a form of autism characterized by regression and bowel symptoms, and onset occurring shortly after vaccination. It is also claimed that, as a result, there has been a dramatic increase in autism prevalence. These hypotheses have now been tested in a number of epidemiologic studies that are reviewed in this article. None has found any evidence of the existence of a phenotypically distinct form of autism in children who received the MMR vaccine or of a clustering of onset symptoms in children who are autistic after receiving the MMR vaccine. There is no proof that the overall risk of autism is higher in children who were vaccinated with MMR or of an increase in autism prevalence associated with the use of the MMR vaccine. No epidemiologic evidence suggests an association between MMR vaccination and autism. Moreover, epidemiologic evidence against such an association is compelling.