[Mortality among individuals diagnosed with glucose intolerance by the 75 g oral glucose tolerance test in rural Japanese communities].

OBJECTIVE

To examine mortality among individuals confirmed to have glucose intolerance by the 75 g oral glucose tolerance test (OGTT) in rural communities.

METHODS

Subjects were 1,639 residents of seven rural communities in Japan (total population size = 54,000) who underwent an OGTT as a secondary screening test for diabetes during 1987-95 and had received no medication for diabetes at baseline. OGTT results were classified as diabetes (n = 471), impaired glucose tolerance (IGT) (n = 408), or normal glucose tolerance (NGT) (n = 760), according to the recommendations of the American Diabetes Association. Indirect age-adjusted mortality rates and standardized mortality ratios (SMRs) for diabetes and IGT groups were calculated using age-stratified mortality rates for people aged 40-89 in the communities as the standard population. Relative risks of death and 95% confidence intervals (CI) for diabetes and IGT were calculated using Cox's proportional models, adjusted for age and baseline factors such as body mass index (BMI), total cholesterol, hypertension, smoking, alcohol drinking, history of diabetes, and family history of diabetes.

RESULTS

During 9.4 years of follow-up, we recorded 70 deaths for the NGT, 46 for the IGT, and 71 for the diabetes group. The SMRs for IGT and diabetic men and women were not significantly increased in comparison with the standard population values. However, when analyzed in our cohort, age-adjusted relative risks of death for IGT and diabetes versus NGT men were 1.10 (9.5% CI: 0.72-1.67) and 1.54 (95% CI: 1.05-2.24), respectively, whereas, the risks among women did not significantly vary. Finally, multivariate analysis, further adjusted for the baseline factors, indicated the relative risk of mortality with diabetes in men to be significantly increased (relative risk = 1.74, 95% CI: 1.11-2.75).

CONCLUSION

During follow-up, mortality rates among screened IGT and diabetes were not significantly higher than those in general. Nonetheless, we found that diabetes elevated the risk of death in our cohort, primarily in men.