亚健康状态症状群的探索分析.docVIP

亚健康状态症状群的探索分析 何丽云1 李霞2 白文静1 张艳宏1 刘保延*1 （1中国中医科学院 北京100700 ，2中国人民大学统计学院 北京100872） 【摘要】目的：分析北京地区亚健康人群常见的临床症状特征。方法：利用北京地区亚健康状态流行病学调查数据，应用探索性因子分析，研究各症状的特征、性质及其之间的潜在关系，寻找代表亚健康状态的常见症状群，并用结构方程模型（Structural Equation Modeling,SEM）验证各症状群之间及因子与其各组成因素之间的关系。结果：取特征值大于1.0的因子共计13个，其累计贡献率达55.67％，结合专业知识，合并为10个因子。以疲劳为中心潜变量建立结构方程模型，结果：X2 / DF=3.65，拟合优度指标、调整拟合优度指标、相对拟合指数、tucher-lewis指数均在0.9以上。潜变量之间的直接效用说明：疲劳与视疲劳、睡眠、疼痛关系最为密切，权重系数在0.7-1.0之间，疲劳与二便情况、口咽部症状、出汗症状之间关系较大，权重系数在0.5左右。疲劳与潜变量之间的间接效用说明：疲劳间接对情绪失调起作用，其相关因子之间的多阶间接作用总和为0.702和0.112。结论：亚健康状态的症状表现复杂多样，常以某一症状为主，伴有多种症状同时出现；最常见的代表症状群依次为情绪失调，疲劳，睡眠异常，二便失调，心态不平和，视疲劳，疼痛等。结构方程模型表明，各症状群之间、症状群内症状与症状之间均存在强弱不等的因果关系，说明亚健康状态症状表现有其群集性和复杂性等特征。 关键词：亚健康状态 症状特征 因子分析 结构方程模型 The exploratory analysis of clinical symptoms of the sub-health in BeiJing He Li-yun1 Li Xia2 Bai Wen-jing1 Zhang Yan-hong1 Liu Bao-yan*1 1.China Academy of Chinese Medicine sciences, Beijing 100700，China 2.School of Statistics Renmin University of China，Beijing 100872，China 【abstract】Object：To analyze the common clinical symptoms of sub-health in Beijing district. Methods: With the sub-health epidemiology data and the method of factors analysis, we study the features and properties of various kinds of symptoms and the relationships between the symptoms, looking for the common symptom groups that are the representatives of various kinds of sub-health. Furthermore, we use the Structural Equation Modeling to validate the relationship between the symptom groups and the relationship between the symptom groups and their components. Results: The amount of the factors whose weight values are larger than 1.0 is 13, the cumulative of all the factors is 55.67%. However, with the domain knowledge the total factors are merged into 10. We take the fatigue as the main hidden variables to build the structural equation model, the function is X2 / DF=3.65, GFI，AGFI，CFI，tucher-lewis indexes are all above 0.9. The direct effects between the hidden variables show that fatigue has the most