||AIMS/HYPOTHESIS: Diabetes increases the risk of acute myocardial infarction (AMI) and effective means for primary prevention are warranted. We prospectively examined the joint association of diabetes and leisure-time physical activity, as well as of diabetes and BMI, with the risk of AMI. METHODS: A total of 55,534 men and women in the Norwegian HUNT Study were followed-up for first AMI by hospital admission registries and the Cause of Death Registry. Cox proportional adjusted HRs with 95% CIs were estimated. RESULTS: Overall, 1,887 incident AMIs occurred during 12.3 years. Compared with inactive people without diabetes, inactive people with diabetes had an HR of 2.37 (95% CI 1.58, 3.57), whereas the HR among highly active persons with diabetes was 1.04 (95% CI 0.62, 1.74). Normal-weight (BMI 18.5-25 kg/m2) persons with diabetes had an HR of 1.60 (95% CI 1.05, 2.44) and obese (BMI > 30 kg/m2) persons with diabetes had an HR of 2.55 (95% CI 1.97, 3.29) compared with normal-weight persons without diabetes. The data suggest biological interaction between diabetes and physical activity, with a relative excess risk of inactivity and diabetes of 1.43 (95% CI 0.08, 2.78). For obesity and diabetes, the excess risk due to interaction was smaller (0.67; 95% CI -0.24, 1.58). CONCLUSIONS/INTERPRETATION: Body weight and, in particular, physical activity modified the association between diabetes and risk of first AMI. This highlights the potential importance of physical activity and weight maintenance in primary prevention of AMI among people with diabetes.