The Menninger Clinic, 12301 Main Street, Houston, TX 77035, United States; Baylor College of Medicine, One Baylor Plaza, Houston, TX 77030, United States.
The Menninger Clinic, 12301 Main Street, Houston, TX 77035, United States; Baylor College of Medicine, One Baylor Plaza, Houston, TX 77030, United States.
Compr Psychiatry. 2018 Jan;80:97-103. doi: 10.1016/j.comppsych.2017.09.003. Epub 2017 Sep 12.
With the publication of DSM 5 alternative model for personality disorders it is critical to assess the components of the model against evidence-based models such as the five factor model and the DSM-IV-TR categorical model. This study explored the relative clinical utility of these models in screening for borderline personality disorder (BPD).
Receiver operator characteristics and diagnostic efficiency statistics were calculated for three personality measures to ascertain the relative diagnostic efficiency of each measure. A total of 1653 adult inpatients at a specialist psychiatric hospital completed SCID-II interviews. Sample 1 (n=653) completed the SCID-II interviews, SCID-II Questionnaire (SCID-II-PQ) and the Big Five Inventory (BFI), while Sample 2 (n=1,000) completed the SCID-II interviews, Personality Inventory for DSM5 (PID-5) and the BFI.
BFI measure evidenced moderate accuracy for two composites: High Neuroticism+ low agreeableness composite (AUC=0.72, SE=0.01, p<0.001) and High Neuroticism+ Low+Low Conscientiousness (AUC=0.73, SE=0.01, p<0.0001). The SCID-II-PQ evidenced moderate-to-excellent accuracy (AUC=0.86, SE=0.02, p<0.0001) with a good balance of specificity (SP=0.80) and sensitivity (SN=0.78). The PID-5 BPD algorithm (consisting of elevated emotional lability, anxiousness, separation insecurity, hostility, depressivity, impulsivity, and risk taking) evidenced moderate-to-excellent accuracy (AUC=0.87, SE=0.01, p<0.0001) with a good balance of specificity (SP=0.76) and sensitivity (SN=0.81).
Findings generally support the use of SCID-II-PQ and PID-5 BPD algorithm for screening purposes. Furthermore, findings support the accuracy of the DSM 5 alternative model Criteria B trait constellation for diagnosing BPD. Limitations of the study include the single inpatient setting and use of two discrete samples to assess PID-5 and SCID-II-PQ.
随着 DSM-5 人格障碍替代模型的出版，评估该模型的组成部分与基于证据的模型（如五因素模型和 DSM-IV-TR 分类模型）相对比至关重要。本研究探讨了这些模型在筛查边缘型人格障碍（BPD）方面的相对临床效用。
计算了三种人格测量工具的接收者操作特征和诊断效率统计数据，以确定每种测量工具的相对诊断效率。共有 1653 名专科精神病院的成年住院患者完成了 SCID-II 访谈。样本 1（n=653）完成了 SCID-II 访谈、SCID-II 问卷（SCID-II-PQ）和大五人格量表（BFI），而样本 2（n=1000）完成了 SCID-II 访谈、人格障碍诊断问卷-5（PID-5）和 BFI。
BFI 测量工具在两个组合方面表现出中等准确性：高神经质+低宜人性组合（AUC=0.72，SE=0.01，p<0.001）和高神经质+低+低尽责性组合（AUC=0.73，SE=0.01，p<0.0001）。SCID-II-PQ 表现出中等至良好的准确性（AUC=0.86，SE=0.02，p<0.0001），具有良好的特异性（SP=0.80）和敏感性（SN=0.78）平衡。PID-5 BPD 算法（包括情绪不稳定、焦虑、分离不安全感、敌意、抑郁、冲动和冒险行为）表现出中等至良好的准确性（AUC=0.87，SE=0.01，p<0.0001），具有良好的特异性（SP=0.76）和敏感性（SN=0.81）平衡。
研究结果普遍支持使用 SCID-II-PQ 和 PID-5 BPD 算法进行筛查。此外，研究结果支持 DSM-5 替代模型 B 特质群标准用于诊断 BPD 的准确性。研究的局限性包括单一住院环境和使用两个离散样本来评估 PID-5 和 SCID-II-PQ。
