Huan Hu1, Yohei Kawasaki2, Keisuke Kuwahara3, Tohru Nakagawa4, Toru Honda4, Shuichiro Yamamoto4, Masafumi Eguchi5, Takeshi Kochi5, Akiko Nishihara6, Teppei Imai6, Makoto Yamamoto7, Hiroko Okazaki8, Toshiaki Miyamoto9, Kentaro Tomita10, Akihiko Uehara11, Takayuki Ogasawara12, Naoko Sasaki12, Ai Hori13, Satsue Nagahama14, Makiko Shimizu15, Taizo Murakami15, Sanmei Chen16, Isamu Kabe5, Tetsuya Mizoue16, Tomofumi Sone17, Seitaro Dohi8. 1. Department of Epidemiology and Prevention, National Center for Global Health and Medicine, Tokyo, Japan. Electronic address: hu.huanhuan@yahoo.com. 2. Biostatistics Section, Clinical Research Center, Chiba University Hospital, Chiba, Japan. 3. Department of Epidemiology and Prevention, National Center for Global Health and Medicine, Tokyo, Japan; Teikyo University Graduate School of Public Health, Tokyo, Japan. 4. Hitachi, Ltd., Ibaraki, Japan. 5. Furukawa Electric Co., Ltd., Tokyo, Japan. 6. Azbil Corporation, Tokyo, Japan. 7. Yamaha Corporation, Shizuoka, Japan. 8. Mitsui Chemicals, Inc., Tokyo, Japan. 9. Nippon Steel Corporation Kimitsu Works, Chiba, Japan. 10. Mitsubishi Plastics, Inc., Tokyo, Japan. 11. Seijinkai Shizunai Hospital, Hokkaidō, Japan. 12. Mitsubishi Fuso Truck and Bus Corporation, Kanagawa, Japan. 13. Department of Global Public Health, University of Tsukuba, Ibaraki, Japan. 14. All Japan Labour Welfare Foundation, Tokyo, Japan. 15. Mizue Medical Clinic, Keihin Occupational Health Center, Kanagawa, Japan. 16. Department of Epidemiology and Prevention, National Center for Global Health and Medicine, Tokyo, Japan. 17. National Institute of Public Health, Saitama, Japan.
Abstract
BACKGROUND & AIMS: To investigate trajectories of body mass index (BMI) and waist circumference (WC) among prediabetic people who progressed to diabetes, people who remained with prediabetes, and those who returned to normoglycemia. METHODS: We used data from 22,945 prediabetic people who received an annual health checkup for up to eight years. The development of diabetes was defined using the American Diabetes Association criteria. People who did not progress to diabetes during the observation period were classified as 'remained with prediabetes' or 'returned to normoglycemia', based on their last health checkup data. Trajectories of BMI and WC were evaluated using linear mixed models for repeated measures, with adjustment for a wide range of covariates. RESULTS: During the study period, 2972 people progressed to diabetes, 4706 returned to normoglycemia, and 15,267 remained with prediabetes. People who progressed to diabetes had a larger increase in mean BMI from 7 years to 1 year prior to diagnosis, which was about three times that of people who remained with prediabetes (annual change rate, 0.20 [95% confidence interval; 0.15 to 0.24] vs 0.06 [0.04 to 0.08] kg/m2 per year, P < 0.001), regardless of their BMI levels at the initial health checkup. Among people who returned to normoglycemia, mean BMI remained almost the same over time (-0.04 [-0.09 to 0.002] kg/m2 per year), except for those with obesity (-0.16 [-0.28 to -0.05] kg/m2 per year). As for WC, the annual change rate among people who developed diabetes was about 7 times that of people who remained with prediabetes (0.38 [0.32 to 0.45] vs 0.05 [0.03 to 0.08] cm per year, P < 0.001). We also observed a constant mean WC over time among people who had no central obesity and later returned to normoglycemia (-0.02 [-0.06 to 0.03] cm per year), and an annual decrease in mean WC among those who had central obesity and later returned to normoglycemia (-0.40 [-0.47 to -0.32] cm per year). CONCLUSIONS: Our study provides strong evidence that avoiding weight gain could help prediabetic people minimize the risk of developing diabetes, regardless of whether they are obese. Losing weight could help obese people restore normoglycemia from a prediabetic state, whereas maintaining current weight may help nonobese people return to normoglycemia.
BACKGROUND & AIMS: To investigate trajectories of body mass index (BMI) and waist circumference (WC) among prediabetic people who progressed to diabetes, people who remained with prediabetes, and those who returned to normoglycemia. METHODS: We used data from 22,945 prediabetic people who received an annual health checkup for up to eight years. The development of diabetes was defined using the American Diabetes Association criteria. People who did not progress to diabetes during the observation period were classified as 'remained with prediabetes' or 'returned to normoglycemia', based on their last health checkup data. Trajectories of BMI and WC were evaluated using linear mixed models for repeated measures, with adjustment for a wide range of covariates. RESULTS: During the study period, 2972 people progressed to diabetes, 4706 returned to normoglycemia, and 15,267 remained with prediabetes. People who progressed to diabetes had a larger increase in mean BMI from 7 years to 1 year prior to diagnosis, which was about three times that of people who remained with prediabetes (annual change rate, 0.20 [95% confidence interval; 0.15 to 0.24] vs 0.06 [0.04 to 0.08] kg/m2 per year, P < 0.001), regardless of their BMI levels at the initial health checkup. Among people who returned to normoglycemia, mean BMI remained almost the same over time (-0.04 [-0.09 to 0.002] kg/m2 per year), except for those with obesity (-0.16 [-0.28 to -0.05] kg/m2 per year). As for WC, the annual change rate among people who developed diabetes was about 7 times that of people who remained with prediabetes (0.38 [0.32 to 0.45] vs 0.05 [0.03 to 0.08] cm per year, P < 0.001). We also observed a constant mean WC over time among people who had no central obesity and later returned to normoglycemia (-0.02 [-0.06 to 0.03] cm per year), and an annual decrease in mean WC among those who had central obesity and later returned to normoglycemia (-0.40 [-0.47 to -0.32] cm per year). CONCLUSIONS: Our study provides strong evidence that avoiding weight gain could help prediabetic people minimize the risk of developing diabetes, regardless of whether they are obese. Losing weight could help obese people restore normoglycemia from a prediabetic state, whereas maintaining current weight may help nonobese people return to normoglycemia.