OBJECTIVE - Obesity and physical inactivity are established risk factors for type 2 diabetes and cardiovascular comorbidities. Whether adiposity or fitness level is more important to health is controversial. The objective of this research is to determine the relative associations of physical activity and BMI with the prevalence of diabetes and diabetes-related cardiovascular comorbidities in the U.S. RESEARCH DESIGN AND METHODS - The Medical Expenditure Panel Survey (MEPS) is a nationally representative survey of the U.S. population. From 2000 to 2002, detailed information on sociodemographic characteristics and health conditions were collected for 68,500 adults. Normal weight was defined as BMI 18.5 to <25 kg/m2, overweight 25 to ≤30 kg/m2, obese (class 1 and 11) 30 to <40 kg/m2, and obese (class III) ≥40 kg/m2. Physical activity was defined as moderate/vigorous activity ≥30 min a ≥3 days per week. RESULTS - The likelihood of having diabetes and diabetes-related cardiovascular comorbidities increased with BMI regardless of physical activity and increased with physical inactivity regardless of BMI. Compared with normal-weight active adults, the multivariate-adjusted odds ratio (OR) for diabetes was 1.52 (95% Cl 1.25-1.86) for normal-weight inactive adults and 1.65 (1.40-1.96) for overweight inactive adults; the OR for diabetes and comorbid hypertension was 1.71 (1.32-2.19) for normal-weight inactive adults and 1.84 ( 1.47-2.32) for overweight inactive adults. CONCLUSIONS - Both physical inactivity and obesity seem to be strongly and independently associated with diabetes and diabetes-related comorbidities. These results support continued research investigating the independent causal nature of these factors. © 2005 by the American Diabetes Association.