OBJECTIVE: To assess the effects of BMI on progression to diabetes in Hong Kong Chinese and to analyze the optimal cutoff for overweight and obesity in Hong Kong Chinese. RESEARCH METHODS AND PROCEDURES: This is a prospective study with a mean follow-up of 2.1 years (median 1.4 years, range 0.9 to 8.4 years). We recruited 172 nondiabetic high-risk subjects, of whom 115 had normal glucose tolerance (NGT) and 57 had impaired glucose tolerance (IGT). BMI and 75-gram oral glucose tolerance tests were assessed at baseline and then at yearly intervals. RESULTS: The crude rates of progression to diabetes for subjects with NGT or IGT were 8.4% and 11.5% per year, respectively. For subjects with NGT, the progression rate to diabetes differed with different BMI ranges. For subjects with NGT and BMI > or = 25 kg/m2, the crude rates of progression to diabetes or glucose intolerance (diabetes or IGT) were 12.5% per year and 14.6% per year, respectively. The corresponding rates for subjects with NGT and BMI > or = 28 kg/m2 were 14.6% and 18.9% per year, respectively. Among subjects with NGT, those with BMI between 25 and 28 kg/m2 had the highest Youden index and likelihood ratio to predict the conversion to diabetes or glucose intolerance. DISCUSSION: Obese subjects with NGT had higher rates of progression to diabetes than nonobese subjects. We recommend redefining BMI cutoffs, with 23 kg/m2 for overweight and 28 kg/m2 for obesity. This definition may be more sensitive to identify at-risk subjects and more specific to identify "patients" for therapeutic management.
OBJECTIVE: To assess the effects of BMI on progression to diabetes in Hong Kong Chinese and to analyze the optimal cutoff for overweight and obesity in Hong Kong Chinese. RESEARCH METHODS AND PROCEDURES: This is a prospective study with a mean follow-up of 2.1 years (median 1.4 years, range 0.9 to 8.4 years). We recruited 172 nondiabetic high-risk subjects, of whom 115 had normal glucose tolerance (NGT) and 57 had impaired glucose tolerance (IGT). BMI and 75-gram oral glucose tolerance tests were assessed at baseline and then at yearly intervals. RESULTS: The crude rates of progression to diabetes for subjects with NGT or IGT were 8.4% and 11.5% per year, respectively. For subjects with NGT, the progression rate to diabetes differed with different BMI ranges. For subjects with NGT and BMI > or = 25 kg/m2, the crude rates of progression to diabetes or glucose intolerance (diabetes or IGT) were 12.5% per year and 14.6% per year, respectively. The corresponding rates for subjects with NGT and BMI > or = 28 kg/m2 were 14.6% and 18.9% per year, respectively. Among subjects with NGT, those with BMI between 25 and 28 kg/m2 had the highest Youden index and likelihood ratio to predict the conversion to diabetes or glucose intolerance. DISCUSSION: Obese subjects with NGT had higher rates of progression to diabetes than nonobese subjects. We recommend redefining BMI cutoffs, with 23 kg/m2 for overweight and 28 kg/m2 for obesity. This definition may be more sensitive to identify at-risk subjects and more specific to identify "patients" for therapeutic management.