Season of birth and childhood psychosis

This paper reports a test of the hypothesis that winter birth is a risk factor for childhood psychosis. All psychotic children (n=208) admitted to a day-treatment facility in Paris from 1960 through 1986 were compared with 1040 controls matched by sex and year of birth. Risk ratio estimates were computed for each month, and a significant increase was observed for boys born in December. This association was stronger in some years than in others. The fact that there were no excesses in contiguous months detracts from the persuasiveness of this association.