This paper reports the effects of Behavioral Family Systems Therapy (BFST) on directly observed interactions between parents and 119 adolescents with diabetes. Families received 3 months' treatment with BFST, an Education and Support Group (ES) or Current Therapy (CT). Multimodal assessments at baseline, posttreatment, and 6- and 12-month follow-ups included tape- recorded family problem-solving discussions scored with the Interaction Behavior Code. Compared with CT and ES, BFST yielded more improvement in family communication and problem-solving skills. Improvements in individual family members' communication were correlated with other improvements in family relationships and adaptation to diabetes, but not with changes in diabetic control. Increased positive and decreased negative reciprocity in family communication and improved family problem solving were associated with improved treatment adherence and adolescent adjustment to IDDM, but not with improved diabetic control.