OBJECTIVE: Diabetes has been one of the leading causes of death since 1983. The death rate increased sharply in past ten years. The mortality was 44.4 per 100,000 in 2003, about 72% increment from 25.7, the death rate in 1993. The mortality in 2003 was almost 10 times of Japan, over three times of US, and two times of Korea's mortality. The purpose of this study is to investigate the differences of diabetes in the US population and the Taiwanese population. METHODS: We used two national surveys to identify diabetes in the population, the 1999-2002 NHANES and the 2002 Taiwan Survey on Hypertension, Hyperglycemia and Hypercholesterolemia. The definition of diabetes was those (a) who reported to be told to have the disease; (b) who reported to take medications for the disease, or (c) with fasting glucose over 126 mg/dL. Their demographic characteristics, lipid profiles, blood pressures, health behaviors and complications were compared. Descript statistics were used to compare the diabetes patients directly. SUDAAN was used to estimate the design effect after accounting for the complex survey scheme. RESULTS: There were 350 whites, 594 blacks and 478 Taiwanese were identified as diabetes in the surveys. The average ages for males were around 55-58 years, while those for females were 55-65 years. The HbA1c was best control in US whites. Both white males and females were under 8%. On the other hand, it was over 8% for the Taiwanese females and all blacks. Taiwanese had the lowest BMI and highest DBP among these three groups of people. The proportions of exercise, smoking and hypertension in Taiwanese were the lowest. CONCLUSIONS: The possible reasons for the differences between these populations are genetic makeup, public health education and treatments.