The hypothesis that direct exposure to or clinical illness with measles, rubella, mumps or chickenpox may play a causal role in autism has been evaluated. The study subjects were 163 cases of autism and their 355 unaffected siblings. Medical data relating to prenatal life and early infancy were obtained by interviewing the parents, and from medical records. Medical records were found for only about half the instances In which a pertinent illness or exposure was reported in the home Interview. However, there was no evidence of systematic bias In recall on the part of parents when reporting on an index case. When medical records and parental statements were both available they were relatively consonant. Increased frequency of combined illness and exposure was seen for the cases relative to their sibling controls with regard to prenatal measles, rubella and mumps and to postnatal mumps. Yet the proportion of cases with histories of viral experience during the periods of interest was small, indicating that these four viruses were unlikely to have played a major role In any substantial proportion of the cases. Some associations were noted with parental histories of other infections during gestation or infancy. Again, however, the proportion of cases that could be accounted for by any of these associations, even if causal, is small. © 1979 by The Johns Hopkins University School of Hygiene and Public Health.
