Department of Psychology, Macquarie University, Sydney, Australia.
Advanced Neuropsychological Treatment Services, Sydney, Australia.
Appl Neuropsychol Adult. 2022 Nov-Dec;29(6):1450-1457. doi: 10.1080/23279095.2021.1888727. Epub 2021 Mar 1.
The objective of this study was to determine the test-retest reliability; construct and criterion validity; and test operating characteristics of a newly developed cognitive impairment risk factor screening instrument, the Alcohol and Drug Cognitive Enhancement (ACE) Screening Tool. Participants in the validation study were 129 adults with substance use disorder (SUD) enrolled in residential SUD treatment services and 209 normal controls. Test and retest data were available for 36 participants with SUD and 40 normal control individuals on the ACE Screening Tool. Test-retest reliability was excellent (ICC = 0.97). The ACE Screening Tool was significantly correlated with the Montreal Cognitive Assessment (MoCA), Behavior Rating Inventory of Executive Functioning-Adult Version (BRIEF-A), Test of Premorbid Functioning (TOPF) and Five Point Test, establishing construct validity. Criterion validity was established using a ternary severity variable constructed using results obtained on the MoCA and BRIEF-A. Test operating characteristics analysis showed 93% sensitivity, 46% specificity, 33% positive predictive power, and 96% negative predictive power using a cut-score of >3. Those high levels of sensitivity and negative predictive power indicated that the tool would likely detect cognitive impairment when present and should therefore be considered suitable as an initial screening tool for cognitive impairment in individuals attending SUD services.
本研究旨在确定新开发的认知障碍风险因素筛查工具——酒精和药物认知增强（ACE）筛查工具的重测信度、结构效度和测试工作特征。验证研究的参与者包括 129 名患有物质使用障碍（SUD）的成年人和 209 名正常对照者，他们被纳入住院 SUD 治疗服务。36 名 SUD 患者和 40 名正常对照者有 ACE 筛查工具的测试和重测数据。重测信度非常高（ICC=0.97）。ACE 筛查工具与蒙特利尔认知评估（MoCA）、行为评定量表认知执行功能成人版（BRIEF-A）、预患病能力测验（TOPF）和五分测验显著相关，确立了结构效度。使用 MoCA 和 BRIEF-A 结果构建的三分严重程度变量建立了标准效度。使用>3 的截断分数进行测试工作特征分析显示，敏感性为 93%，特异性为 46%，阳性预测值为 33%，阴性预测值为 96%。这些高敏感性和阴性预测值表明，该工具在存在认知障碍时可能会检测到认知障碍，因此应被视为 SUD 服务中认知障碍的初始筛查工具。
