In this preliminary study, we attempted to determine whether diabetic subjects with a positive family medical history for diabetes have a better understanding of their own disease (ie., of its course, treatment, and genetic characteristics) than diabetic subjects without diabetes in their histories. We interviewed 50 diabetic subjects and scored their responses to determine their understanding of diabetes in general and in their own cases in particular. The data were analyzed using the Student's t test and chi-square analysis. Overall, patients with positive histories did not have significantly higher understanding than patients with negative histories. However, if patients had extensive exposure to the affected relative, or if tht relative was a spouse or a parent, the patients' understanding was significantly higher (P less than 0.05). Educational level, age, and duration of diabetes did not affect patients' understanding. The data suggest that the family medical history can be a valuable teaching model, once we redefine it so as to reconcile patient concepts about illness with physician concepts of disease.