Infantile autism was diagnosed by DSM-III criteria in 132 children (26 girls) who were outpatients of the Tsuchiura Child Guidance Center during the years 1977-1985. The children, all Japanese except for one Laotian boy born in Laos, were classified according to year and month of birth. The prevalence rate of infantile autism in southern Ibaraki, Japan, within the birth cohort born between 1972 and 1978 was 13.9/10,000 children. The month of birth for infantile autism increased in the second quarter of the year. The prevalence rate of infantile autism in each 1-year birth cohort fluctuated in a 4-year cycle, which was closely correlated (r = .92) with the number of children admitted with pneumonia and bronchiolitis in that area. These findings led us to postulate that infectious factors of children's pneumonia and bronchiolitis may have some role in the cause of infantile autism.

Since early weaning in infancy has been known to result in vulnerability to infection, weaning times of 145 children diagnosed as autistic by DSM-III were statistically compared with those of 224 normal children in the same catchment area: 24.8% of the patients and 7.5% of the controls were weaned by the end of 1 week, a significant difference. Early weaning because of the mother's rather than the child's condition occurred with 17.9% of the patients and 5.8% of the controls, also a significant difference. Historical studies on infantile autism revealed that the disease developed more prevalently in the socioeconomic status where the incidence of breast-feeding was less frequent. These results suggest that early weaning may contribute to the etiology of infantile autism.