ObjectiveAlthough diabetes has a great impact on coronary heart disease (CHD) in Western populations, there is limited evidence that diabetes influences CHD in the Japanese population. MethodsThe Japan Public Health Center-based prospective (JPHC) study conducted a follow-up of 31,192 individuals aged 40–69 years with no history of cardiovascular disease or cancer. Subjects were classified at baseline as normal, borderline diabetic or diabetic based on fasting and non-fasting blood glucose levels and the use of medication to treat diabetes. A Cox proportional hazards model was used to determine the association between diabetes and the risk of fatal and non-fatal CHD events after adjustment for potential confounders. ResultsDuring 12.9 years of follow-up (1990–2006), we identified 266 fatal and non-fatal coronary events using validated criteria. With normal individuals serving as a control, the hazard ratios for total incident CHD events after adjusting for sex, age, study community and fasting were 1.65 (95% CI, 1.19–2.29) and 3.05 (2.03–4.59) in the borderline and diabetic groups, respectively. These associations remained significant after adjustment for conventional risk factors. The population attributable fractions (PAF) of borderline diabetes and diabetes for CHD events were 6.9% and 6.3%, respectively. Furthermore, there was a trend for an association between an impaired fasting glucose (5.6–6.9mmol/l) and an increased risk of CHD events. ConclusionsThis prospective study suggests that diabetes and elevated glucose levels are associated with incident CHD in the general Japanese population. The PAF of diabetes for fatal and non-fatal coronary events was estimated to be moderate.