摘要
目的探讨颈围在社区人群诊断中心性肥胖的准确性及其切点。方法选取2010年10月至2011年1月,采用横断面普查设计对广州市石牌社区50岁以上的原始居民(n=1 494)进行调查。所有调查对象测量身高、体质量、颈围和腰围,中心性肥胖定义为男性腰围≥90 cm,女性腰围≥85 cm,颈围诊断中心性肥胖及超重的准确性及最佳切点通过绘制ROC曲线来评价。结果 1 473例数据完整者中位年龄为61(55~68)岁,颈围与腰围及体质指数的Pearson相关系数分别是0.639和0.565(均P<0.001)。颈围在男性人群中诊断中心性肥胖ROC曲线下面积分别为0.884(95%CI:0.855~0.909,P<0.000 1),最佳切点值分别为38.5 cm(敏感性和特异性分别为75.2%和85.4%)。颈围在女性人群中诊断中心性肥胖的ROC曲线下面积为0.849(95%CI:0.824~0.872,P<0.000 1),其最佳切点值分别为33.4 cm(敏感性和特异性分别为84.8%和69.4%)。结论在该社区50岁以上人群,颈围与中心性肥胖密切相关,其诊断中心性肥胖有较高的准确性,可作为腰围和体质指数的替代指标。
Objective To explore the role of neck circumference (NC) in identifying central obesity. Methods A community-based cross-sectional survey was conducted from October 2010 to January 2011 in Shipai community, Guangzhou. A total of 1 494 subjects (72.8%) aged over 50 were investigated. Height, weight, waist circumference (WC) and NC were measured in all subjects. Central obesity was defined by the standard in the 2013 China Guideline(WC ≥ 90.0 cm for men or WC ≥85.0 cm for women). Results The median age of 1 473 subjects with complete data was 61 years (55 - 68 years). NC was significantly positively correlated with WC and body mass index (correlation coefficient was 0.639 and 0.565, respectively, P 〈 0.001 ). ROC curves showed that the area under the curve for the ability of neck circumference to determine visceral adiposity was 0.884 (95%C1: 0.855 - 0.909, P 〈 0.000 1 ) for men and 0.849 (95%CI: 0.824 - 0.872, P 〈 0.000 1 ) for women. Moreover, in men, a NC of 38.5 cm had a sensitivity of 75.2% and specificity of 85.4 %, and in women, a NC of 33.4 em had a sensitivity of 84.8% and specificity of 69.4% , which were the optimal cutoffs for identifying central obesity. Conclusions NC could be used as an alternative for WC and body mass index to identify central obesity in Chi- nese subjects aged over 50 in community-based setting.
出处
《实用医学杂志》
CAS
北大核心
2017年第18期3127-3130,共4页
The Journal of Practical Medicine
基金
广州市科技计划项目科技惠民专项(编号:2014Y2-00072)
关键词
颈围
中心性肥胖
腰围
ROC曲线
neck circumference
central obesity
waist circumference
ROC curve