Institute of Psychiatry, Department of Biomedical and Specialty Surgical Sciences, University of Ferrara, Ferrara, Italy.
University Unit of Hospital Psychiatry, S. Anna Hospital and Health Authorities, Ferrara, Italy.
Cancer. 2018 Jun 1;124(11):2415-2426. doi: 10.1002/cncr.31340. Epub 2018 Apr 16.
Given the adverse consequences of psychiatric and psychosocial morbidity on the quality of life for patients with cancer, prompt detection of psychological symptoms is mandatory. The authors examined the properties and accuracy of the Brief Symptom Inventory (the 53-item version [BSI] and the 18-item version [BSI-18]) for the detection of psychiatric morbidity compared with the World Health Organization Composite International Diagnostic Interview (CIDI) for International Classification of Diseases-10th Revision psychiatric diagnoses.
A convenience sample of 498 patients with newly diagnosed cancer who were recruited in cancer outpatient services participated in the CIDI interview and in BSI and BSI-18 assessments.
The prevalence of psychiatric morbidity was 39.75%. When participants were classified as cases using the BSI standard case rule, agreement with the CIDI was potentially acceptable (sensitivity, 72.7%; specificity, 88.7%). In contrast, the accuracy of the BSI-18 in identifying cases was poor according to the standard case rule, with very low sensitivity (29.3%) (misclassification rate, 28.7%). By using a first alternative case-rule system (a BSI-18 global severity index [GSI] T-score ≥57), sensitivity marginally improved (45%), whereas a second alternative case-rule system (a GSI T-score ≥50) significantly increased sensitivity (77.3%). In receiver operating characteristic curve analysis, a further cutoff GSI T-score ≥48 exhibited good discrimination levels (sensitivity, 82.3%; specificity, 72.4%). There were some differences in GSI cutoff T-scores according to the International Classification of Diseases-10th Revision diagnosis and sex.
The BSI appeared to have acceptable diagnostic accuracy compared with a standardized psychiatric interview. For the BSI-18, it is mandatory to use alternative case-rule systems, to identify patients with psychiatric morbidity. Cancer 2018;124:2415-26. © 2018 American Cancer Society.
鉴于精神和心理社会发病率对癌症患者生活质量的不良影响，及时发现心理症状是强制性的。作者检查了Brief Symptom Inventory（53 项版本[BSI]和 18 项版本[BSI-18]）用于检测与世界卫生组织国际疾病分类第 10 版综合国际诊断访谈（CIDI）精神疾病诊断的精神发病率的特性和准确性。
便利抽样 498 例新诊断癌症患者在癌症门诊服务中参与了 CIDI 访谈以及 BSI 和 BSI-18 评估。
精神发病率的患病率为 39.75%。当使用 BSI 标准病例规则对参与者进行分类时，与 CIDI 的一致性具有潜在的可接受性（敏感性，72.7%；特异性，88.7%）。相比之下，根据标准病例规则，BSI-18 识别病例的准确性较差，敏感性非常低（29.3%）（错误分类率，28.7%）。通过使用第一个替代病例规则系统（BSI-18 全球严重程度指数[GSI]T 评分≥57），敏感性略有提高（45%），而第二个替代病例规则系统（GSI T 评分≥50）则显著提高了敏感性（77.3%）。在接收器操作特征曲线分析中，进一步的 GSI T 评分≥48 的截止值表现出良好的区分水平（敏感性，82.3%；特异性，72.4%）。GSI 截止 T 评分根据国际疾病分类第 10 版诊断和性别存在一些差异。
BSI 与标准化精神病访谈相比，似乎具有可接受的诊断准确性。对于 BSI-18，必须使用替代病例规则系统，以识别患有精神疾病的患者。癌症 2018;124:2415-26。©2018 美国癌症协会。
