基于分类树模型对高尿酸血症的影响因素分析

Analysis on the associated factors of hyperuricemia: based on classification tree model

目的了解影响高尿酸血症(hyperuricemia,HUA)的生理、社会心理及膳食等生活习惯因素,为HUA的干预提供参考依据。方法对2016-2017年在大连医科大学附属第二医院健康管理中心体检的成年人进行问卷调查和体格测量,比较基本资料、行为习惯、心理因素等对HUA患病风险的影响。采取分类树模型分析各研究因素与HUA的关系。结果研究共纳入研究对象4118人,平均年龄为(52.8±7.9)岁。统计发现,30岁以下及40~50岁、吸烟、饮酒、静坐为主工作、对工作不满意、精神压力极大的人群HUA检出率更高。因子分析共提取出3种膳食模式:传统模式,油炸腌熏甜点模式,优质蛋白模式,累计方差贡献率53.886%。多因素模型显示,优质蛋白模式、体育锻炼和空腹血糖(fasting plasma glucose,FPG)与HUA呈负相关,男性、吸烟、饮酒、血脂异常及体重指数(body mass index,BMI)为HUA的危险因素。性别分析显示,男、女优质蛋白模式均与HUA呈负相关,油炸腌熏甜点模式为女性HUA的独立危险因素。分类树模型显示男性、血脂异常、高BMI等级、油炸腌熏甜点模式是HUA的危险因素。结论不同性别人群HUA的影响因素不同,应采取个性化的健康管理。超重或肥胖男性应控制吸烟、饮酒等行为,女性更应注意油炸腌熏甜点食物的摄入,提倡优质蛋白饮食模式及适当体育锻炼。

Objective To explore the associations between physiological factors,psychosocial factors,dietary habits,lifestyles and hyperuricemia(HUA)and provide the evidence for hyperuricemia intervention.Methods From 2016 to 2017,adults during their checkup in the health management center of the Second Affiliated Hospital of Dalian Medical University were interviewed with self-designed questionnaire and received the health examinations.To determine the influence of socio-economic data,life style habits and psychological factors on the risk of HUA,the classification tree model was adopted.Results A total of 4118 subjects were enrolled in the study,with an average age of(52.8±7.9)years.The results showed that the prevalence of HUA was higher among people under 30 years old and 40-50 years old who are smoking,drinking and doing the sedentary work,who unsatisfied with their work and under a great mental stress.Factor analysis model identified three dietary patterns,traditional model,fried and smoked food and dessert model and high quality protein model,the cumulative variance contribution rate was 53.886%.In multivariate model,it was found that high-quality protein pattern,physical exercise and fasting plasma glucose(FPG)were negatively associated with HUA.Male,smoking,drinking,having dyslipidemia,higher body mass index(BMI)were risk factors for HUA.Subgroup analysis showed that in both male and female,the high-quality protein patterns were negatively associated with HUA.The pattern of fried and smoked food and dessert patterns was the independent risk factors for HUA in female.Classification tree model showed that male,dyslipidemia,higher BMI grade,and the fried and smoked food and dessert pattern were the risk factors for HUA.Conclusions The influencing factors of HUA in different gender were different,which indicate the individualized health management should be adopted.Male who is overweight or obese should quit smoking and drinking.Women should pay more attention to avoid of the excess intake of fried and smoked food and dessert.Meanwhile,a high-quality protein diet and more physical exercise should be encouraged.