The concurrent accuracy of the modified telephone interview for cognitive status and mini-mental state examination tools in detection of cognitive impairment among older adults

ABSTRACT. Due to the need for face-to-face administration of many cognitive screening tests, it is not always feasible to screen large-scale samples. Objective: This study aimed to assess the discriminant validity of the Persian version of Telephone Interview for Cognitive Status (P-TICS-m) and Mini-Mental State Examination in the middle-aged Iranian population. Methods: The P-TICS-m and MMSE were administered to 210 randomly selected middle-aged community-dwelling adults who had been registered in the Neyshabur Longitudinal Study on Ageing. Participants also underwent psychological examination by two neurologists to assess cognitive impairment based on the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-V) criteria. To evaluate the discriminant validity of P-TICS-m and MMSE with DSM-V criteria, the sensitivity, specificity, positive and negative predictive values (PPV and NPV), and positive and negative likelihood ratios (LR+ and LR−) were calculated. Results: The mean age of the participants was 59.6±6.8 years. The TICS and MMSE were highly correlated (r=0.635, p<0.001). The sensitivity, specificity, PPV, NPV, LR+, and LR− to discriminate cognitive impairment were, respectively, 83%, 92%, 68%, 96%, 10, and 0.182 for MMSE and 100%, 13%, 19%, 100%, 1.16, and 0 for TICS-m. The receiver operating characteristic curve analysis results showed no statistically significant differences between P-TICS-m and MMSE. Conclusions: Our findings indicate that the TICS-m test can be used as a screening tool instead of the MMSE. Due to the low specificity and low PPV of the TICS-m compared to MMSE, the diagnosis should be confirmed using definitive diagnostic tests when a subject is classified as having cognitive impairment.

摘要。由于多项认知筛查测验需面对面施测，对大规模样本开展筛查往往不具备可行性。研究目的：本研究旨在针对伊朗中年人群，评估波斯语版电话认知状态访谈量表（Telephone Interview for Cognitive Status，P-TICS-m）与简易精神状态检查表（Mini-Mental State Examination，MMSE）的区分效度。研究方法：招募内沙布尔老龄化纵向研究（Neyshabur Longitudinal Study on Ageing）登记的210名随机选取的社区居住中年成年人，对其施测P-TICS-m与MMSE。参与者还接受了两名神经内科医师开展的精神检查，以依据《精神障碍诊断与统计手册第五版》（Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, DSM-V）标准评估认知损害情况。为评估P-TICS-m与MMSE相较于DSM-V标准的区分效度，本研究计算了灵敏度、特异度、阳性预测值与阴性预测值（PPV与NPV），以及阳性似然比与阴性似然比（LR+与LR−）。研究结果：参与者的平均年龄为59.6±6.8岁。P-TICS-m与MMSE评分呈高度相关（r=0.635，p<0.001）。以区分认知损害为目标，MMSE的灵敏度、特异度、阳性预测值、阴性预测值、阳性似然比与阴性似然比分别为83%、92%、68%、96%、10与0.182；P-TICS-m的对应指标则分别为100%、13%、19%、100%、1.16与0。受试者工作特征曲线（Receiver Operating Characteristic curve，ROC曲线）分析结果显示，P-TICS-m与MMSE的区分效度无统计学显著差异。研究结论：本研究结果表明，P-TICS-m可作为MMSE的替代筛查工具。但相较于MMSE，P-TICS-m的特异度与阳性预测值均较低，因此当受试者被判定为存在认知损害时，需通过确定性诊断试验进一步确认诊断。