Finley John-Christopher A, Robinson Anthony D, Cerny Brian M, Ovsiew Gabriel P, Pliskin Neil H, Calamia Matthew, Ulrich Devin M, Phillips Matthew S, Soble Jason R
Department of Psychiatry and Behavioral Sciences, Northwestern University Feinberg School of Medicine, Chicago, IL, USA.
Department of Psychiatry, University of Illinois College of Medicine, Chicago, IL, USA.
Clin Neuropsychol. 2025 Aug;39(6):1581-1597. doi: 10.1080/13854046.2024.2420376. Epub 2024 Oct 28.
This study investigated whether extreme cut-scores on the Barkley Adult Attention-Deficit/Hyperactivity Disorder (ADHD) Rating Scale-Fourth Edition (BAARS-IV) self-report scales could serve as symptom overreporting indicators in adult ADHD evaluations. The sample consisted of 162 adults who underwent a neuropsychological evaluation for ADHD. Patients were classified into valid ( = 115) and invalid ( = 47) groups based on multiple criterion symptom validity tests. Overreporting cut-scores identified for each of the BAARS-IV scales demonstrated acceptable classification accuracy or better. The Current Total scale yielded the highest classification accuracy (area under the curve of .83). An optimal raw cut-score of ≥56 for this scale yielded 47% sensitivity when upholding ≥90% specificity. A cut-score of ≥31 for the Current Inattention scale (90% specificity; 47% sensitivity) and ≥29 for the Current Sluggish Cognitive Tempo scale (88% specificity; 46% sensitivity) were also relatively strong indicators of overreporting. The Current Total and Current Sluggish Cognitive Tempo scales captured nonredundant aspects of symptom validity, and using them together increased sensitivity to 57% while maintaining ≥90% specificity. Findings provide preliminary support for the criterion and construct validity of extreme cut-scores on the BAARS-IV Current Sluggish Cognitive Tempo, Current Inattention, and Current Total scales as indicators of ADHD symptom overreporting. However, employing the cut-scores from the Current Sluggish Cognitive Tempo (≥29) and Current Total (≥56) together may be the most promising way to detect overreporting. Scores above both of these cut-points should, at a minimum, prompt further investigation into the validity of a patient's reported symptoms.
本研究调查了《巴克利成人注意力缺陷/多动障碍评定量表（第四版）》（BAARS-IV）自陈量表上的极端划界分数是否可作为成人注意力缺陷多动障碍（ADHD）评估中症状过度报告的指标。样本包括162名接受ADHD神经心理学评估的成年人。根据多项标准症状效度测试，将患者分为有效组（n = 115）和无效组（n = 47）。为BAARS-IV各量表确定的过度报告划界分数显示出可接受的分类准确性或更高。当前总分量表的分类准确性最高（曲线下面积为0.83）。该量表的最佳原始划界分数≥56，在保持≥90%特异性时，敏感性为47%。当前注意力不集中量表划界分数≥31（特异性90%；敏感性47%）和当前认知迟缓量表划界分数≥29（特异性88%；敏感性46%）也是过度报告的相对有力指标。当前总分量表和当前认知迟缓量表捕捉了症状效度的非冗余方面，同时使用它们可将敏感性提高到57%，同时保持≥90%的特异性。研究结果为BAARS-IV当前认知迟缓量表、当前注意力不集中量表和当前总分量表上的极端划界分数作为ADHD症状过度报告指标的标准效度和结构效度提供了初步支持。然而，同时使用当前认知迟缓量表（≥29）和当前总分量表（≥56）的划界分数可能是检测过度报告最有前景的方法。至少，高于这两个切点的分数应促使进一步调查患者报告症状的效度。
