Supplementary Material for: The Diagnostic Utility of the NINDS-CSN Neuropsychological Battery in Memory Clinics

Aims: To examine the diagnostic utility of the National Institute of Neurological Disorders and Stroke and the Canadian Stroke Network (NINDS-CSN) neuropsychological battery in memory clinics comparing controls with patients with no cognitive impairment (NCI), patients with cognitive impairment-no dementia (CIND) at varying severity levels (mild/moderate), and patients with dementia.Methods: A total of 405 participants with NCI, CIND or dementia were assessed with the NINDS-CSN battery. The discriminatory properties of all three protocols (5, 30 and 60 min) before and after education stratification (none/primary vs. secondary/above) were examined by receiver operating characteristic curves. Results: Overall, the shorter protocols are equivalent to the longer protocol in diagnosing dementia, regardless of education. To discriminate between nondementia groups, before education stratification, the 5-min protocol showed varied discriminatory properties between different diagnostic/severity groups. After stratification, the 5-min protocol was broadly equivalent to the longer protocols in lower-education groups [area under the curve (AUC) range: 0.77-0.87] but was less accurate in the higher-education groups (AUC range: 0.68-0.78). The 30- and 60-min protocol constantly showed moderate-to-excellent differentiating capacities regardless of education (AUC range: 0.80-0.90). Conclusion: The NINDS-CSN neuropsychological battery can be applied in memory clinics and effectively discriminate between cognitively intact individuals and those with cognitive impairments of varying severity. Furthermore, level of education should be taken into consideration when choosing protocols with different lengths for cognitive assessment.研究目的：探讨美国国家神经疾病和中风研究所（National Institute of Neurological Disorders and Stroke）与加拿大中风网络（Canadian Stroke Network）联合研发的NINDS-CSN神经心理成套测验在记忆门诊中的诊断应用价值，本研究以健康对照者、无认知损害（no cognitive impairment, NCI）患者、不同严重程度（轻度/中度）的无痴呆认知损害（cognitive impairment-no dementia, CIND）患者以及痴呆患者为比较对象。 研究方法：本研究共纳入405名认知正常、无痴呆认知损害或痴呆的参与者，采用NINDS-CSN神经心理成套测验对其进行评估。通过受试者工作特征曲线（Receiver Operating Characteristic, ROC），分析三种时长方案（5分钟、30分钟、60分钟）在教育程度分层（未接受教育/初等教育 vs 中等及以上教育）前后的区分性能。 研究结果：总体而言，无论受试者的教育程度如何，时长较短的测验方案在痴呆诊断中的效能与长时方案相当。未进行教育程度分层时，5分钟方案在区分不同诊断/严重程度组的性能存在差异。经教育程度分层后，5分钟方案在低教育程度组中的区分效能与长时方案基本相当[曲线下面积（Area Under the Curve, AUC）范围：0.77~0.87]，但在高教育程度组中准确性稍逊（AUC范围：0.68~0.78）。30分钟与60分钟方案无论教育程度如何，均持续表现出中等至优秀的区分能力（AUC范围：0.80~0.90）。 研究结论：NINDS-CSN神经心理成套测验可应用于记忆门诊，能够有效区分认知完好个体与不同严重程度的认知损害人群。此外，在选择不同时长的测验方案开展认知评估时，应将受试者的教育程度纳入考量因素。