Soldz S, Budman S, Demby A, Merry J
Mental Health Research Program, Harvard Community Health Plan, Brookline, MA 02146.
J Pers Assess. 1993 Jun;60(3):486-99. doi: 10.1207/s15327752jpa6003_6.
In an attempt to compare different methods for assessing personality disorder, this study compared the Millon Clinical Multiaxial Inventory-II (MCMI-II; Millon, 1987), a self-report questionnaire, and the Personality Disorder Examination (PDE; Loranger, 1988), a semistructured clinical interview. Subjects (N = 97) were mental health outpatients of a health maintenance organization in New England. The instruments were compared in terms of the presence of personality disorder, the number of diagnoses assigned to a patient, and agreement in specific diagnoses and in cluster assignment. Agreement between the two instruments was low; the two instruments exhibited greater agreement in predicting the absence of diagnoses than their presence. Agreement was best for the borderline and avoidant diagnoses. Correlations between scales exhibited somewhat better agreement than was evident for diagnoses. Analyses at the cluster level resulted in moderate correlations between the instruments. Very high intracluster correlations were found for the MCMI-II, but not for the PDE.
为了比较评估人格障碍的不同方法,本研究对比了自陈问卷《米隆临床多轴问卷第二版》(MCMI-II;米隆,1987年)和半结构化临床访谈《人格障碍检查》(PDE;洛兰热,1988年)。研究对象(N = 97)是新英格兰一家健康维护组织的心理健康门诊患者。从人格障碍的存在情况、给予患者的诊断数量以及特定诊断和聚类分配的一致性方面对这两种工具进行了比较。两种工具之间的一致性较低;在预测无诊断情况时,两种工具表现出的一致性比预测有诊断情况时更高。边缘型和回避型诊断的一致性最佳。量表之间的相关性表现出的一致性比诊断方面明显要好一些。聚类水平的分析得出两种工具之间存在中度相关性。发现《米隆临床多轴问卷第二版》存在非常高的聚类内相关性,但《人格障碍检查》则没有。