Low Accuracy of Brief Cognitive Tests in Tracking Longitudinal Cognitive Decline in an Asian Elderly Cohort

April Ka Sin Phua,Shaun Kuan Wei Hiu,Win King Goh,Mohammad Kamran Ikram,Narayanaswamy Venketasubramanian,Boon Yeow Tan,Christopher Li-Hsian Chen,Xin Xu
DOI: https://doi.org/10.3233/jad-170831
2018-01-01
Abstract:BACKGROUND:Researchers have questioned the utility of brief cognitive tests such as the Mini-Mental Status Examination (MMSE) and the Montreal Cognitive Assessment (MoCA) in serial administration and suggested that brief cognitive tests may not accurately track changes in Global Cognition.OBJECTIVE:To examine the accuracy of longitudinal changes on brief cognitive tests in reflecting progression in Global Cognition measured using comprehensive neuropsychological assessments.METHODS:Two hundred and seven participants were assessed with the MMSE, MoCA, and a validated comprehensive neuropsychological battery. Global z-scores on the battery were derived and used to assess overall and significant (≥0.5 standard deviation) decline on Global Cognition. Different patterns of decline on MMSE/MoCA were classified. Accuracy was examined using receiver operating characteristic curve, and sensitivity, specificity, positive (PPV) and negative (NPV) predictive values were reported.RESULTS:The overall ability of MMSE/MoCA change scores to discriminate participants who did and did not decline on Global Cognition was fair-to-moderate (AUC [95% CI] = 0.71 [0.64-0.78] & 0.73 [0.66-0.80] for overall decline; 0.78 [0.70-0.85] & 0.80 [0.73-0.86] for significant decline, respectively). Changes in MMSE/MoCA had low accuracy in identifying significant Global Cognitive Decline (PPV = 0.41 & 0.46, respectively) but high accuracy in ruling out significant decline and identifying cognitively stable participants (NPV = 0.89 & 0.88, respectively).CONCLUSION:There is limited utility in brief cognitive tests for tracking cognitive decline. Instead, they should be used for identifying participants who remain cognitively stable on follow up. These results accentuate the importance of acknowledging the limitations of brief cognitive tests when assessing cognitive change.
What problem does this paper attempt to address?