OBJECTIVE: We examined if active family nutritional support is associated with improved metabolic outcomes for Diné (Navajo) individuals living with type 2 diabetes. RESEARCH DESIGN AND METHODS: The presence of family support, using variables identified in earlier ethnographic research, was assessed via surveys in a convenience sample of 163 diabetic individuals. Diabetes outcome measures (HbA(1c), serum glucose, triglycerides, total cholesterol, creatinine, and systolic and diastolic blood pressure measures) were extracted from participants' medical records. Bivariate analyses and multiple logistic regressions were conducted. RESULTS: All measures of family support showed a relation with one or more indicators of metabolic control in bivariate analyses. In multivariate analyses, respondents were more likely to be in the best tertile for triglyceride (P < 0.05), cholesterol (P < 0.05), and HbA(1c) (P < 0.05) if another person cooked most of the meals. Respondents in families who bought/cooked "light" foods were more likely to be in the best tertile for triglyceride (P < 0.005) and cholesterol levels (P < 0.005), and those in families whose members ate "light" foods with them were more likely to be in the best tertile for triglycerides (P < 0.005). When all three support variables were entered into a multivariate model, only the variable "other family members cook the majority of the meals" was significantly associated with being in the lowest triglyceride (P = 0.05), HbA(1c) (P < 0.05), or cholesterol tertiles (P < 0.05). These relationships were most evident for women with diabetes. CONCLUSIONS: Active family nutritional support, as measured by culturally relevant categories, is significantly associated with control of triglyceride, cholesterol, and HbA(1c) levels. The findings suggest that the family is a more useful unit of intervention for Diné individuals than for the individual alone when designing diabetes care strategies.