Sensitivity and specificity of the CANTAB Paired Associates Learning test for identifying mild cognitive impairment

Abstract

Mild cognitive impairment (MCI) is a stage between normal cognition and dementia, and timely diagnosis of this condition offers the opportunity for effective interventions to slow down the progression to dementia. The Mini-Mental Status Exam (MMSE) and the Montreal Cognitive Assessment (MoCA) are widely used tests for MCI screening. The Cambridge Neuropsychological Test Automated Battery (CANTAB) is a computer-based cognitive battery designed to reduce the influence of language and the possibility of examiner error. This study aimed to investigate the diagnostic accuracy of CANTAB in patients with MCI. Patients with a MoCA score of 25 or less were considered to be in the MoCA-defined MCI group, which comprised 23.1% of participants (15 out of 65). The Paired Associates Learning (PAL)-mean-error-to-success had the most area under the curve in the ROC curve (0.80; 95%CI: 0.66–0.95; p -value<0.01) and optimum cut-off for MCI diagnosis was 5.8 (Sensitivity: 77%; Specificity: 74%) based on the Youden index. Although the small sample size may have influenced the results, the findings suggest that the PAL demonstrated good diagnostic accuracy for distinguishing participants with MoCA-defined MCI.