Supplementary Material for: Responsiveness to Change of the Montreal Cognitive Assessment, Mini-Mental State Examination, and SCOPA-Cog in Non-Demented Patients with Parkinson’s Disease

<b><i>Background:</i></b> Clinical monitoring of patients with Parkinson’s disease (PD) for cognitive decline is an important element of care. The Montreal Cognitive Assessment (MoCA) has been proposed to be a sensitive tool for assessing cognitive impairment in PD. The aim of our study was to compare the responsiveness of the MoCA to decline in cognition to the responsiveness of the Mini Mental State Examination (MMSE) and the Scales for Outcomes of Parkinson’s disease-cognition (SCOPA-Cog). <b><i>Methods:</i></b> PD patients without dementia were enrolled at 6 North American movement disorders centers between 2008 and 2011. Participants received annual evaluations including the MoCA, MMSE, and SCOPA-Cog followed by formal neuropsychological testing. The gold standard for change in cognition was defined as the change on the neuropsychological test scores over the annual assessments. The Reliable Change Method was used to provide an estimate of the probability that a given difference score would be obtained by chance. The sensitivity of the MoCA, MMSE, and SCOPA-Cog to change was quantified using receiver operating characteristics (ROC) curves. <b><i>Results:</i></b> One hundred seventeen patients were included in the analysis. Participants were followed at mean intervals of 11 ± 2 months for a median of 2 (maximum 5) visits. According to the reliable change index, 56 intervals of cognitive testing showed a decline in global cognition. ROC analysis of change in MoCA, MMSE, and SCOPA-Cog global scores compared to gold standard testing found an area under the curve (AUC) of 0.55 (95% CI 0.48–0.62), 0.56 (0.48–0.63), and 0.63 (0.55–0.70) respectively. There were no significant differences in the AUCs across the tests. The sensitivity of the MoCA, MMSE, and SCOPA-Cog to change at various thresholds for decline in scores reached a maximum of 71% for a cut-off of 1 point change on the SCOPA-Cog. <b><i>Conclusion:</i></b> Using neuropsychological testing as a gold standard comparator, the performance of the MoCA, MMSE, and SCOPA-Cog for detecting decline in non-demented PD patients over a 1-year interval is poor. This has implications for clinical practice; stable scores may not be taken as reassurance of the absence of cognitive decline.

<b><i>背景：</i></b> 对帕金森病（Parkinson’s disease, PD）患者开展认知减退的临床监测，是临床照护的关键环节。蒙特利尔认知评估量表（Montreal Cognitive Assessment, MoCA）已被推荐为评估帕金森病患者认知损害的敏感工具。本研究旨在比较MoCA与简易精神状态检查表（Mini Mental State Examination, MMSE）、帕金森病结局量表-认知分量表（Scales for Outcomes of Parkinson’s disease-cognition, SCOPA-Cog）在检测认知减退方面的响应性。<b><i>方法：</i></b> 2008年至2011年间，研究团队于北美6家运动障碍中心纳入无痴呆的帕金森病患者。受试者接受年度评估流程：依次完成MoCA、MMSE与SCOPA-Cog测评，随后接受正式神经心理学测试。认知变化的金标准定义为年度评估中神经心理学测试得分的变化量。本研究采用可靠变化法（Reliable Change Method），估算某一差异得分由偶然因素导致的概率。通过受试者工作特征（Receiver Operating Characteristics, ROC）曲线，量化MoCA、MMSE及SCOPA-Cog对认知变化的敏感性。<b><i>结果：</i></b> 共计117例患者纳入最终分析。受试者的平均随访间隔为11±2个月，中位随访次数为2次（最长随访5次）。根据可靠变化指数，共有56次认知测评间隔显示整体认知功能减退。以金标准神经心理学测试为参照，对MoCA、MMSE及SCOPA-Cog的整体得分变化进行ROC分析，其曲线下面积（Area Under the Curve, AUC）分别为0.55（95%置信区间CI 0.48~0.62）、0.56（0.48~0.63）及0.63（0.55~0.70）。三种测评工具的AUC无显著统计学差异。在不同得分减退阈值下，MoCA、MMSE及SCOPA-Cog的检测敏感性最高可达71%，该最优值对应SCOPA-Cog得分变化1分的截断值。<b><i>结论：</i></b> 以神经心理学测试作为金标准对照，MoCA、MMSE及SCOPA-Cog在1年随访间隔内，检测无痴呆帕金森病患者认知减退的效能较差。该结果对临床实践具有重要启示：稳定的测评得分不能作为排除认知减退的可靠依据。