Takeuchi Hiroyoshi, Fervaha Gagan, Remington Gary
From the *Schizophrenia Division, Centre for Addiction and Mental Health; and †Department of Psychiatry, University of Toronto, Toronto, Ontario, Canada; ‡Department of Neuropsychiatry, Keio University School of Medicine, Tokyo, Japan; and §Institute of Medical Science, University of Toronto; and ‖Campbell Family Mental Health Research Institute, Centre for Addiction and Mental Health, Toronto, Ontario, Canada.
J Clin Psychopharmacol. 2016 Dec;36(6):593-596. doi: 10.1097/JCP.0000000000000569.
This study aimed to compare (1) the detection rates of antipsychotic-associated side effects between clinician and patient ratings and (2) differences as a function of change and absolute score definitions.
Data from phase 1 of the Clinical Antipsychotic Trials of Intervention Effectiveness (N = 1460) were analyzed. In this trial, 18 adverse events were systematically and concurrently assessed by clinicians and patients using a 4-point severity scale ranging from 0 (absent) to 3 (severe). The incidence of antipsychotic-associated side effects was calculated according to 2 definitions: change score (ie, higher score on the scale versus baseline) and absolute score (a score of 2 or 3 on the scale). In addition, patient and clinician concurrent detection rates were examined.
The differences in incidence of antipsychotic-associated side effects between clinician and patient ratings were as small as 5.7% across the 2 definitions. The incidence of all side effects across clinician and patient ratings was approximately 2 times higher when using the change versus absolute score definition. Among the side effects detected by patients, 11 side effects were identified more frequently by clinicians, with 14.3% to 30.2% differences when using the change versus absolute score definition. Conversely, there was no difference of 10% or greater in patient or clinician concurrent detection rate on any item when using the absolute versus change score definition.
Our findings suggest that patient ratings are in line with clinician ratings and that the change score definition may be superior for the assessment of antipsychotic-associated side effects in clinical studies.
本研究旨在比较（1）临床医生与患者评分中抗精神病药物相关副作用的检出率，以及（2）根据变化和绝对评分定义的差异。
对干预有效性临床抗精神病药物试验第1阶段的数据（N = 1460）进行分析。在该试验中，临床医生和患者使用从0（无）到3（严重）的4级严重程度量表，对18种不良事件进行系统且同时的评估。抗精神病药物相关副作用的发生率根据两种定义计算：变化评分（即量表上高于基线的分数）和绝对评分（量表上得分为2或3）。此外，还检查了患者和临床医生的同时检出率。
在两种定义下，临床医生与患者评分中抗精神病药物相关副作用发生率的差异小至5.7%。使用变化评分定义时，临床医生和患者评分中所有副作用的发生率约为使用绝对评分定义时的2倍。在患者检测出的副作用中，临床医生更频繁地识别出11种副作用，使用变化评分与绝对评分定义时，差异在14.3%至30.2%之间。相反，使用绝对评分与变化评分定义时，任何项目上患者或临床医生的同时检出率差异均未达到10%或更大。
我们的研究结果表明，患者评分与临床医生评分一致，并且变化评分定义在临床研究中评估抗精神病药物相关副作用方面可能更具优势。
