累积Logit模型对农村留守居民抑郁症状影响因素的分析

Application of Cumulative Logit Model in the Analysis of Influencing Factors on Depressive Symptoms in Left-behind Rural Residents

目的探讨累积Logit模型在农村留守居民抑郁症状影响因素分析中的应用。方法采用多阶段分层随机抽样的方法,使用中文版流调用抑郁量表对514名沈阳市农村留守居民的抑郁症状进行调查。非参数秩和检验用于农村留守居民抑郁症状的分层单因素分析,并以单因素分析有统计学意义的变量拟合农村留守居民抑郁症状影响因素的累积Logit模型。结果单因素分析显示,农村留守居民抑郁症状分类在不同年龄、文化程度、年人均收入、患慢性病、债务负担、睡眠状况和疲劳状况之间差异均有统计学意义(P<0.05)。多因素分析显示,对累积Logit模型贡献有统计学意义的变量是患慢性病、债务负担、睡眠状况和疲劳状况。结论累积Logit模型是有序反应变量有效的分析方法。患慢性病、债务负担、睡眠状况差和过度疲劳增加了沈阳市农村留守居民抑郁症状发生的风险。

Objective To explore the application of cumulative Logit model in the analysis of influencing factors on depressive symptoms in the left-behind rural residents.Methods The depressive symptoms of 514 left-behind rural residents in Shenyang selected by multistage stratified sampling were assessed by the Chinese Version of the Center for Epidemiologic Studies Depression Scale（CES-D）.Non-parametric rank sum tests were used for stratified univariate analysis,and the variables that were confirmed to have statistical significance in the univariate analysis were introduced to fit the cumulative Logit model of related factors of depressive symptoms in the rural left-behind residents.Results The univariate analysis showed that the depressive symptom classification displayed significant difference among groups of different age,education,annual per capita income,chronic diseases,debt burden,sleeping quality and fatigue status.Further analysis by cumulative Logit model showed that four factors,that is,chronic diseases,debt burden,sleeping quality and fatigue status,were identified as independent predictors of the depressive symptoms.Conclusions Cumulative Logit model is an effective analysis method of ordinal response variables.Chronic diseases,debt burden,poor sleep and over-fatigue may increase the risk of the occurrence of depressive symptoms in the left-behind rural residents in Shenyang.