苏州市成年人身体形态指数和身体圆度指数与糖尿病的关系研究

Relationship of a body shape index and body roundness index with diabetes mellitus in adults in Suzhou

目的探讨苏州市成年人身体形态指数(ABSI)和身体圆度指数(BRI)与糖尿病的关系及其预测糖尿病的能力。方法采用横断面研究,收集2016年3月—4月苏州市10 089名常驻居民的人口学资料,测量身高、体重、腰围及血压,检测空腹血糖及血脂。通过身高和体重计算身体形态指数(ABSI)和身体圆度指数(BRI)。采用logistic回归模型分析ABSI、BRI与糖尿病的关联关系,并采用受试者工作特征曲线(ROC)计算ABSI、BRI、身体质量指数(BMI)、腰围(WC)和腰高比(WHtR)对糖尿病预测的曲线下面积(AUC)及95%可信区间(CI)。结果 9984名资料完整者被纳入本次分析,男性4026名,女性5958名,糖尿病发生率为14.18%(n=1416)。在调整多个因素后,以每项人体测量指标的下四分位为参比组,最高四分位组与糖尿病患病的关联性分析发现,男性中BMI的OR值最大(OR=2.552,95%CI:1.793~3.631),其次为BRI和WHtR。女性中,BRI和WHtR的OR值均比较高,且二者的OR值相似(BRI:OR=2.412,95%CI:1.813~3.209;WHtR:OR=2.406,95%CI:1.811~3.196)。对于男性和女性糖尿病患者而言,ABSI显示出最差的预测能力,其AUC值最低,而BRI与WHtR相当,对糖尿病预测的AUC均较高。结论预测糖尿病患病方面,ABSI、BRI并不优于传统的人体测量指标,ABSI显示出最弱的预测能力,而BRI显示出与WHtR类似的能力,可以用来预测糖尿病患病。

Objective To explore the relationship of a body shape index(ABSI)and body roundness index(BRI)with diabetes mellitus(DM)in adults in Suzhou,and to assess its predictive value of DM. Methods A cross-sectional study was conducted in Suzhou,Jiangsu province from March to April 2016,and demographic data were collected from 10089 participants. The height,weight,waist circumference and blood pressure were measured,and fasting blood glucose and blood lipid were detected. ABSI and BRI were calculated based on height and weight. The logistic regression model was used to estimate the association between ABSI,BRI and diabetes. The area under the curve(AUC)and 95% confidence interval(CI)of ABSI,BRI,body mass index(BMI),waist circumference(WC)and waist height ratio(WHtR)were calculated by receiver operating characteristic curve(ROC). Results A total of 9984 subjects with complete data were included in this analysis,including 4026 males and 5958 females. The prevalence of diabetes was 14.18%(n=1416). After adjusting for multiple factors,taking the lower quartile of each anthropometric index as the reference group,the association analysis between the highest quartile and diabetes mellitus showed that the Odds Ratio(OR)of BMI in men was the largest(OR=2.552,95%CI:1.793-3.631),followed by BRI and WHtR. In women,the OR values of BRI and WHtR were relatively higher,and they were similar(BRI:OR=2.412,95% CI:1.813-3.209;WHtR:OR=2.406,95% CI:1.811-3.196). For male and female patients with diabetes,ABSI showed the worst predictive ability,with the lowest AUC value,while BRI was similar to WHt R,with higher AUC for diabetes prediction. Conclusion ABSI and BRI are not superior to BMI,WC or WHt R in predicting the prevalence of diabetes. ABSI presents the weakest predictive ability,while BRI and WHtR show a similar ability and can be used to predict diabetes.