状态—特质抑郁题库构建及结构方程模型分析

发布时间：2018-06-11 17:23

  本文选题：状态抑郁 + 特质抑郁　； 参考：《第四军医大学》2016年硕士论文

【摘要】：随着时代快速发展,物质生活水平不断提高,人类的精神世界却随着外界的高速运行凸显出了一系列适应性问题。世界卫生组织(WHO)等其它卫生机构都根据近十几年的医疗现状预测,在可见的未来,抑郁症很可能会成为导致人类社会死亡和残疾的重要致病因素,成为最重要的卫生问题。尽管抑郁症的诊断和治疗显得十分急迫,但相关调查发现,由于抑郁症患者多数情况下选择的是综合医院,加之内科门诊医生对心理障碍的识别率不高(调查发现仅为21%),导致接受抗抑郁治疗的患者也为数不多(仅为5%)。因此,提升初级医疗机构中非精神科医务人员关于心理问题的识别能力就显得十分必要。本研究从抑郁症的症状出发,通过总结整理美国精神疾病诊断标准(第3版)、国际疾病分类(第10版)和中国精神障碍分类及诊断标准(第3版)中关于抑郁症的诊断描述,归纳总结了9大类抑郁症的常见症状,并按照状态-特质进行划分,搜集通用的抑郁测评量表,组建题库。本研究的主要研究结果有:1.根据抑郁症状的不同类别,将9大类症状区分为状态-特质抑郁,其中,状态抑郁包括:精神运动性问题、生理状况(躯体感受,睡眠情况,饮食情况)、抑郁心境、注意力减退;特质抑郁包括:负性态度、无力自责感、社会关系、死亡观念、兴趣缺失。2.对题库中所有条目进行项目反应理论的参数估计,各个症状维度的测验信息量都达到了理想位置,保留的97道条目的区分度均0.7,探索性因素分析(EFA)的结果都体现了良好的因子结构,验证性因素分析(CFA)的结果所有条目的载荷均在0.4以上,CFI和TLI均在0.95以上,RMSEA的估计值小于0.05,90%的置信区间在精确拟合的范围内,WRMR的估计值也在可接受的范围内。状态抑郁各维度Cronbach’sα系数平均值为0.768,特质抑郁各维度Cronbach’sα系数平均值为0.790,状态抑郁和特质抑郁分量表的α系数分别为0.949和0.956。3.结构方程模型(SEM)分析结果显示,最终有5个状态-特质抑郁症状维度相关模型,模型1:状态抑郁的抑郁心境(S3)和注意力减退(S4)症状维度对特质抑郁的负性态度(T1)症状维度的路径系数分别为0.804和0.175;模型2:状态抑郁的抑郁心境(S3)和注意力减退(S4)症状维度对特质抑郁的无力感(T2)症状维度的路径系数分别为0.71和0.27;模型3:状态抑郁的睡眠情况(S2.2)和抑郁心境(S3)症状维度对特质抑郁的社会关系(T3)症状维度的路径系数分别为0.187和0.725;模型4:状态抑郁的抑郁心境(S3)症状维度对特质抑郁的死亡观念(T4)症状维度的路径系数为0.87;模型5:状态抑郁的抑郁心境(S3)症状维度对特质抑郁的兴趣缺失(T5)症状维度的路径系数为0.95。研究结论:1.状态-特质抑郁题库有较好的信度和效度,具有良好的心理测量学特性,提示可应用于抑郁检测;2.状态-特质抑郁各症状维度的结构方程模型提示了各自症状的组合模式,预期可对临床诊断提供参考。
[Abstract]:With the rapid development of the times and the continuous improvement of material life, the spiritual world of human beings has highlighted a series of adaptability problems with the rapid operation of the outside world. The World Health Organization (WHO) and other health institutions have predicted that depression is likely to become an important cause of death and disability in human society and the most important health problem in the foreseeable future. Despite the urgency of diagnosis and treatment of depression, the survey found that patients with depression in most cases chose a general hospital. In addition, the rate of recognition of mental disorders by physicians was not high (the survey found that only 21 patients were diagnosed with antidepressants, and the number of patients receiving antidepressant treatment was only 5%. Therefore, it is necessary to improve the ability of non-psychiatric medical personnel to identify mental problems in primary medical institutions. Based on the symptoms of depression, this study summarized the diagnostic descriptions of depression in the US (3rd edition), the International Classification of Diseases (10th Edition) and the Chinese Classification and Diagnostic criteria of Mental Disorder (3rd Edition). The common symptoms of 9 kinds of depression were summarized and classified according to state-trait. The general depression evaluation scale was collected and the question bank was set up. The main results of this study are: 1: 1. According to the different types of depressive symptoms, 9 major symptoms are classified as state-trait depression, in which state depression includes: psychomotor problems, physiological conditions (somatosensory, sleep, diet, depression, depression), Attention loss; trait depression includes: negative attitudes, inability to blame, social relationships, death concepts, and lack of interest. 2. By estimating the parameters of the item response theory for all the items in the item bank, the test information of each symptom dimension has reached the ideal position. The differences of 97 items were 0.7.The results of exploratory factor analysis (EFAA) showed a good factor structure. The results of the confirmatory factor analysis (CFAA) show that the loads of all entries are above 0.4 and the estimated values of RMSEA and TLI are above 0.95. The confidence intervals of less than 0.0590% of the estimated values are within the range of accurate fitting and the estimated values of WRMR are also within the acceptable range. The average value of Cronbachs 伪 coefficient in each dimension of state depression was 0.768, and the average coefficient of Cronbachs 伪 coefficient in every dimension of trait depression was 0.790. The 伪 coefficients of state depression and trait depression component tables were 0.949 and 0.956.3 respectively. The results of structural equation model (SEM) showed that there were 5 status-trait depression dimension correlation models. Model 1: the path coefficients of symptom dimension of depression (S3) and attention deficit (S4) were 0.804 and 0.175, respectively; Model 2: depressive mood of state depression (S3) and attentional depression (S4) The path coefficients of symptom dimension to trait depression were 0.71 and 0.27, respectively; Model 3: sleep status of state depression and depression state of mind were 0.71 and 0.27, respectively. Model 3: the path of symptom dimension to social relationship of trait depression and symptom dimension of depression were 0.71 and 0.27, respectively; Model 3: sleep status of state depression and depression state of mind S3) the pathway of symptom dimension to social relationship of trait depression. Model 4: the path coefficient of symptom dimension of depressive state to death of trait depression and T4) symptom dimension was 0.87; model 5: symptom dimension of depressive state of depressive mood and S3) symptom dimension to trait depression was 0.87; Model 5: depression of state of depressive state of mind (S3) symptom dimension to idiosyncratic depression. The path coefficient of the symptom dimension was 0.95. Conclusion: 1. The state-trait depression question bank has good reliability and validity, and has good psychometric characteristics, suggesting that it can be used in depression test. The structural equation model of the dimensions of state-trait depression indicates the combination pattern of their symptoms, which is expected to provide reference for clinical diagnosis.
【学位授予单位】：第四军医大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：R749.4
，

本文编号：2006079