Purdue University, West Lafayette, IN 47907, USA.
Assessment. 2013 Jun;20(3):353-61. doi: 10.1177/1073191113486182. Epub 2013 Apr 15.
The DSM-5 (Diagnostic and Statistical Manual of Mental Disorders, 5th ed.) Section III will include an alternative hybrid system for the diagnosis of personality disorder (PD). This alternative system defines PD types partly through specific combinations of maladaptive traits, rather than by using a set of polythetic diagnostic criteria. The current report utilizes a large sample of undergraduates (n = 1,159) to examine three dimensional methods for comparing an individual's trait profile to each PD type. We found that the sum of an individual's scores on the assigned traits obtained large convergent correlations (Mdn r =.61) and best reproduced the patterns of PD discriminant correlations observed within the DSM-IV measure. We also tested the DSM-5 Section III model algorithms and compared them with different thresholds for assigning categorical diagnoses. Frequency rates using the algorithms were greatly reduced, whereas requiring half of the assigned traits produced rates that more closely approximated current prevalence estimates. Our research suggests that DSM-5 Section III trait model can reproduce the DSM-IV-TR PD constructs and identifies effective methods of doing so.
《精神障碍诊断与统计手册（第五版）》（DSM-5）第三部分将包括一种人格障碍（PD）诊断的替代混合系统。该替代系统通过特定的适应不良特征组合来部分定义 PD 类型，而不是使用一组多元诊断标准。本报告利用一个大型本科生样本（n=1159），检验了将个体特质与每种 PD 类型进行比较的三维方法。我们发现，个体在指定特征上的得分总和获得了较大的收敛相关（中位数 r =.61），并且最能再现 DSM-IV 测量中观察到的 PD 判别相关模式。我们还测试了 DSM-5 第三部分的模型算法，并将其与分配分类诊断的不同阈值进行了比较。使用算法的频率大大降低，而要求一半的指定特征产生的频率更接近当前的流行率估计。我们的研究表明，DSM-5 第三部分特质模型可以再现 DSM-IV-TR PD 结构，并确定了有效的再现方法。
