广西长寿地区老年人代谢综合征的患病情况分析及预测模型建立

Prevalence analysis and predictive model establishment of metabolic syndrome in the elderly in the longevity area of Guangxi

目的了解广西长寿地区——桂林市恭城县老年人群代谢综合征(MS)的患病率并建立预测模型。方法2018年12月—2019年11月采用整群抽样方法,在广西桂林市恭城县选取莲花镇和栗木镇开展横断面研究。通过体检、问卷调查收集信息,将纳入的1703例研究对象分为MS组和非MS组,利用logistic回归分析MS的影响因素,列线图预测MS发生情况,建立预测模型。结果恭城县MS患病率为11.1%。不同性别、BMI、吸烟情况、工作强度、饮酒情况、食用油类型、肉类日摄入量、日步行时长、尿酸水平的老年人MS患病率有统计学差异(P均<0.05)。logistic回归结果显示女性(OR=1.870,95%CI为1.211~2.889,P=0.005)、超重(OR=6.667,95%CI为4.622~9.616,P<0.05)、肥胖(OR=20.690,95%CI为11.570~37.020,P<0.05)、肉类摄入≤56.58 g/d(OR=1.698,95%CI为1.075~2.683,P=0.023)、高尿酸(OR=2.454,95%CI为1.669~3.608,P<0.05)是发生MS的危险因素;每天步行>4.5 h(OR=0.521,95%CI为0.314~0.865,P=0.012)是其保护因素。建立了MS发生的列线图预测模型。列线图的C-index为0.803(95%CI为0.768~0.838),校准曲线拟合较好。结论广西长寿地区恭城县老年人的MS患病率较低,女性、超重、肥胖、肉类摄入≤56.58 g/d、高尿酸是MS发生的危险因素,建立的列线图可较好预测MS发生。

Objective To understand the prevalence and establish a predictive model of metabolic syndrome(MS)in the elderly in Gongcheng county,longevity area,Guilin City,Guangxi province.Methods From December 2018 to November 2019,the cluster sampling method was used to conduct a cross-sectional research in Lianhua town and Limu town in Gongcheng county,Guilin city,Guangxi.Information was collected through physical examination and question⁃naire surveys.A total of 1703 study subjects were divided into MS group and non-MS group.Logistic regression was used to analyze the influencing factors of MS,and to establish a prediction model of the nomogram to predict the occurrence of MS.Results The prevalence of MS in Gongcheng county was 11.1%.The prevalence of MS was different in the elderly with different gender,BMI,smoking status,work intensity,drinking status,edible oil type,daily meat intake,daily walking time,and uric acid level(all P<0.05).The logistic regression results showed that women(OR=1.870,95%CI:1.211~2.889,P=0.005),overweight(OR=6.667,95%CI:4.622~9.616,P<0.05),obesity(OR=20.690,95%CI:11.570~37.020,P<0.05),meat intake≤56.58 g/d(OR=1.698,95%CI:1.075~2.683,P=0.023),high uric acid(OR=2.454,95%CI:1.669~3.608,P<0.05)were risk factors for MS;walking>4.5 h/d(OR=0.521,95%CI:0.314~0.865,P=0.012)was a protective factor of MS.A nomogram prediction model was established.The C-index of the nomo⁃gram is 0.803(95%CI:0.768~0.838),and the calibration curve fits well.Conclusion The prevalence of MS in the elderly in Gongcheng County,longevity area,Guangxi province is relatively low.Females,overweight,obesity,meat in⁃take≤56.58 g/d,and high uric acid are risk factors for MS.The established nomogram is good to predict the occurrence of MS.