Premium
P3‐202: Comparing The Psychometric Properties of Hong Kong Version of the Montreal Cognitive Assessment Applying Age and Education Corrected Normative Data and Single Cut‐Off in Diagnosing Cognitive Impairment in Hong Kong Elders
Author(s) -
Li Hiu Sze,
Tam Yuen Yee,
Lau Mei Ling,
Nin Tang Lap,
Chan Chun Chung,
Yung Cho Yiu,
Yeung Pui Yu
Publication year - 2016
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1016/j.jalz.2016.06.1864
Subject(s) - percentile , montreal cognitive assessment , dementia , neurocognitive , medicine , cognitive impairment , cutoff , cognition , normative , receiver operating characteristic , diagnostic accuracy , gerontology , pediatrics , psychiatry , statistics , disease , mathematics , philosophy , physics , epistemology , quantum mechanics
Background: The objective of this study is to evaluate the performance between single and age and education corrected cut off in classifying cognitive impairment (CI) by using Hong Kong version of the Montreal Cognitive Assessment(HK-MoCA). Methods:HKMoCA scores were collected from patients attending the cognitive clinic of United Christian Hospital in 2011-2013. A total of 274 patients (dementia, n1⁄4122; mild cognitive impairment(MCI), n1⁄485; normal, n1⁄442) were recruited. The MoCA scores were evaluated by different cutoffs. Age and education corrected cutoff scores were at 16, 7 and 2 percentile as according to the diagnostic criteria used in major (2 percentile) and mild (16 percentile) neurocognitive disorders in the Diagnostic and Statistical Manual of Mental Disorders 5 edition and Petersen’s criteria (7 percentile) for MCI. Comparison was made with the single cutoff of 21/22 for MCI and CI and 18/19 for dementia validated in a local study. Results:To differentiate MCI from normal with single cut off, the sensitivity was 83.5% and specificity 69.0%. Using age and education corrected cut off, sensitivity and specificity were 35.3% and 90.5% and 47.1% and 88.1% when the 7 percentile and 16 percentiles were selected respectively. To detect dementia from controls, single cut off had 91.8% sensitivity and 90.5% specificity while age and education corrected cut off at 2 percentile had sensitivity of 49.2% and specificity 97.6%. For identifying CI from normal, the sensitivity and specificity were 93.2% and 69.0% in single cutoff respectively, in comparing with 63.8% and 88.1% in age and education corrected cut off at 16 percentile (Table 1). The accuracy of single cut off in correctly identifying tested patients into appropriate groups was 77.4% but only 46.4% if age and education corrected cut off were used. Conclusions: HKMoCA is a handy tool in aiding diagnosis of cognitive impairment for its relatively short administration time and incorporation of essential cognitive domains. However, choosing an appropriate Table 3 Logistic regression model of NP test to predict amyloid accumulation in PiB PET (1 of 4 region)