应用因子分析和降秩回归评估膳食模式与血脂异常的关系  

Evaluating relationship between dietary pattern and dyslipidemia using factor analysis and reduced rank regression

作　　者：黄伦 任泽萍[2] 赵俊康[1] 令狐丽琴 梁洁[1] 杨晓娟 景佳蕊 王彤[1] HUANG Lun;REN Ze-ping;ZHAO Jun-kang;LINGHU Li-qin;LIANG Jie;YANG Xiao-juan;JING Jia-rui;WANG Tong(Department of Health Statistics,School of Public Health,Shanxi Medical University,Taiyuan,Shanxi 030001,China;不详)

机构地区：[1]山西医科大学公共卫生学院流行病与卫生统计学教研室,山西太原030001 [2]山西省疾病预防控制中心慢病科

出　　处：《现代预防医学》2021年第24期4442-4446,共5页Modern Preventive Medicine

基　　金：国家自然科学基金(81872715);全国医学专业学位研究生教育指导委员会研究课题(B1-YX20180302-04);山西省卫生计生委科研课题(2018GW20)。

摘　　要：目的研究山西省18岁以上居民膳食模式与血脂异常的关系,比较降秩回归(reduced rank regression,RRR)与因子分析(factor analysis,FA)两种方法的差异。方法通过多阶段分层整群抽样,在山西省抽取5649名对象,通过问卷调查获取调查对象一般情况和膳食摄入信息,抽取静脉血测量血脂。采用RRR和FA法获得膳食模式,使用多变量logistic回归分析两种方法所得膳食模式与血脂异常的关系。结果山西省18岁以上居民的血脂异常患病率为39.36%。FA得到5种膳食模式,分别是:肉食模式、主食谷物模式、绿色果蔬模式、类快餐模式、高蛋白模式。RRR得到一种模式s1。多因素logistic结果显示,FA所得肉食模式(Q4较Q1:OR=1.13,95%CI:1.12~1.55,P_(trend)<0.05)、绿色果蔬模式(Q4较Q1:OR=0.77,95%CI:0.65~0.90,P_(trend)<0.05)与RRR所得S1(Q4较Q1:OR=0.63,95%CI:0.54~0.74,P_(trend)<0.001)与血脂异常有关联。结论肉食模式和绿色果蔬模式分别是血脂异常的危险因素和保护因素,山西省居民在日常饮食中应注意控制肉食及碳水总量的摄入,减少饱和脂肪酸摄入,食用富含不饱和脂肪酸的食物,增加蔬菜水果的摄入来改善膳食结构,降低罹患血脂异常的风险。Objective To investigate the relationship between dietary patterns and dyslipidemia in residents aged 18 years or older in Shanxi Province,and to compare the differences between reduced rank regression(RRR)and factor analysis(FA).Methods Through multi-stage stratified whole-group sampling,5649 residents were selected in Shanxi Province,and data of their general conditions and dietary intake were obtained through questionnaires.In addition,the venous blood was drawn to measure blood lipids.The RRR and FA methods were used to obtain dietary patterns,and the relationship between dietary patterns and dyslipidemia was analyzed using multivariate logistic regression.Results The prevalence of dyslipidemia among residents aged 18 years or older in Shanxi Province was 39.36%.Five dietary patterns were obtained by F A:meat pattern,staple grain pattern,green fruit and vegetable pattern,fast-food-like pattern,and high protein pattern.The pattern SI was obtained by RRR.Multivariate logistic results showed that the meat pattern obtained by FA(Q4 compared to Q1:OR=1.13,95%CI:1.12-1.55,P_(trend)<0.05),green fruit and vegetable pattern(Q4 compared to Q1:OR=0.77,95%CI:0.65-0.90,P_(trend)<0.05)and S1 obtained by RRR(Q4 compared to Q1:OR=0.63,95%CI:0.54-0.74,Ptrend<0.001)were associated with dyslipidemia.Conclusion Meat pattern are risk factors for dyslipidemia,and green fruit and vegetable pattern protective factors.Residents in Shanxi Province should pay attention to control the intake of meat and total carbohydrate in their daily diet,reduce the intake of saturated fatty acids,consume foods rich in unsaturated fatty acids,and increase the intake of vegetables and fruits to improve the dietary structure and reduce dyslipidemia risk.

关 键 词：血脂异常 因子分析 降秩回归 膳食模式 

分 类 号：K589.2[历史地理—历史学] R154.41[历史地理—世界史]