Prospective study of prodromal features for bipolarity in well Amish children.

OBJECTIVE: A prospective study of psychiatrically well Amish children to determine differences in the frequency and pattern of clinical features that may be prodromal for bipolar I disorder. METHOD: Children with a bipolar I parent (n = 100) and children of well parents in a matched control sample (n = 110) were assessed annually for 7 years with semistructured interviews covering medical/developmental features and symptoms/behaviors that are possibly prodromal for bipolarity. Randomized histories of these 210 children were evaluated blindly by 4 clinicians for independent ratings of risk for bipolarity. RESULTS: Thirty-eight percent of the children of bipolar parents were rated as at risk compared with 17% of children in the control sample. Most control sample children with risk ratings had well parents with a bipolar sibling (i.e., family history positive). Children with family histories negative for mental illness rarely received even a low risk rating. Clinical features significantly (p