Incidence and Risk Factors for Diabetes Mellitus in Korean Middle-aged Men : Seoul Cohort DM Follow-up Study

우리나라 성인 남성 당뇨병의 발생양상과 위험요인에 관한 전향적 코호트 연구

Objectives : It is known that the prevalence of diabetes mellitus(DM) appears to be rapidly increasing in recent times in Korea, presumably due to a westernized diet and change of life style followed by rapid economic growth. Based on the Seoul male cohort which was constructed in 1993, this study was conducted to estimate the annual incidence rates of DM through 4 years' follow up and to determine which factors are associated with DM risk in Korean middle-aged men. Methods : Among 14,533 men recruited at baseline, 559 were excluded because they reported a history of diabetes or were found to be diabetes at 1992 routine health examination. During 4 years' follow-up, 237 incident DM cases were ascertained through chart reviews and telephone contacts for those who have ever visited hospitals or clinics under suspicion of DM during 1993-1996 and the biennial routine health examinations in 1994 and 1996. Results : In this study the annual incidence of DM among the study population was estimated to be 0.5 per 100. This study showed that fasting glucose level at initial baseline examination was a powerful predictor of risk for diabetes several years later(fasting blood glucose of $\geq$ 110 mg/dl compared with $\leq$ 80 mg/dl, Hazard Ratio[HR]:15.6, 95% Confidence interval[CI]=9.1-26.6) after considering potential covariates such as age, family history, smoking and alcohol history, body mass index, physical activity, total energy intake, and total fiber intake. Adjusted hazard ratios of family history of diabetes was 1.95(95% CI=1.38-2.75); of obesity as measured by BMI(BMI $\geq$ 25.3 compared with $\leq$ 21.3) was 7.19(95% CI=3,75-13.8); of weight change during middle life(>10kg compared with $\leq$ 5) was 1.77(95% CI=1.16-2.69); of smoking(current vs none) was 1.93(95% CI=1.06-3.51); and fat intake(upper fertile compared with lower fertile) was 1.88(95% CI=1.01-3.49), while fiber intake was associated with the reduced risk(HR=0.36, 95% CI=0.19-0.67). Conclusion : The factors identified in this study indicate that the greatest reduction in risk of diabetes might be achieved through population-based efforts that promote fiber intake and reduce obesity, smoking, and fat intake.