OBJECTIVES: To meta-analyze research testing interventions to improve health behaviors including exercise among adults with type 1 diabetes. METHODS: Multiple literature-searching strategies identified published and unpublished studies reporting glycated hemoglobin outcomes. Fixed- and random-effects meta-analyses included moderator effects. RESULTS: Data were synthesized across 1435 subjects. The overall effect size was 0.26. This translates into a reduction of 0.33 in mean HbA1c levels between treatment (8.47%) and control (8.80%) groups. Effect sizes were larger with more females, lower body mass index, higher baseline HbA1c, and interventions targeting multiple behaviors. CONCLUSIONS: Behavior change interventions improve metabolic control in type 1 diabetes.