北京市3—18岁人群腰围和腰围身高比分布特征及其适宜界值的研究

目的 研究北京市3～18岁人群腰围（WC）、腰围身高比（WHtR）的分布特征；在心血管疾病（CVD）危险因素评估的基础上提出6—18岁人群WC和WHtR的适宜界值。方法 对2004年北京市儿童青少年代谢综合征研究项目总人群中3—18岁人群的WC、WHtR进行描述性分析；基于CVD危险因素评估，采用工作者特征曲线分析方法，研究6—18岁人群WC和WHtR的适宜界值；并利用《北京市学校卫生防病工作规划》（2006）血脂健康调查中6—17岁人群生理、生化检测数据，对上述界值进行交叉验证。结果 基于CVD危险因素评估基础上的WC适宜界值为性别年龄组的第80百分位值，WHtR的适宜界值为0．46。交叉验证显示：按WC和WHtR适宜界值划分的肥胖组血压、血脂等CVD危险因素无论均值还是检出率均显著高于正常体重组。结论 上述WC、WHtR适宜界值可敏感地区分高血压、高三酰甘油和低高密度脂蛋白胆固醇等CVD危险因素，WC作为相对简单的测量方法应列入学生体检的常规测量项目。

Using waist circumference and waist-to-height ratio to access central obesity in children and adolescents

Objective To analyze the distribution of waist circumference and waist-to-height ratio for 3 - 18 year-old children and to explore the related optimal thresholds for schoolchildren in Beijing based on the assessment of cardiovascular risk factors. Methods As part of the Beijing Child and Adolescent Metabolic Syndrome study, a stratified cluster representative sample of 23 422 children aged 0 - 18 years was selected. Anthropometric measures including height, weight and waist circumference data were collected from about 20 000 subjects aged 3 - 18 years. Fasting finger-stick capillary whole blood levels of glucose, total cholesterol and triglyceride were measured by Accutrend GCT measuring system. Receiver operating characteristic analysis was used to determine the waist circumference and waist-to-height ratio to discriminate between the presence and absence of the abnormality of these cardiovascular risk factors and to determine the optimal thresholds by age and gender. The second sample was made up of 2 794 children and adolescents （1 456 male and 1 338 female）, aged 6 to 17 years, randomly selected from 18 districts in the Beijing area. Systolic and diastolic blood pressure data were collected. Fasting serum lipids including total cholesterol, triglyceride, high density lipoprotein cholesterol, low density lipoprotein cholesterol, fasting plasma glucose, anthropometric index as height and waist circumference were measured. According to waist circumference and waist-to-height ratio cut off points, test samples were fallen into two groups including a normal group and an obesity group. The average levels and abnormal rates of the cardiovascular risk factors were compared with general linear model analysis between the two groups. Results ① The waist circumference measurements were differed by age, gender and urban-rural residence, and the waist-to-height ratio remained comparatively steady among different groups. ② After adjusting for age, gender, urban-rural residence, the binary logistic regression model