OBJECTIVES: To pilot two new cognitive screening tools for use in an urban Malaysian population and to compare their criterion validity against a gold standard, the well-established Mini-Mental State Examination (MMSE).
METHODS: The IDEA cognitive screen, Picture-based Memory Impairment Scale (PMIS), and MMSE were administered to a convenience sample of elderly (≥ 65 years) from the community and outpatient clinics at an urban teaching hospital. Consensus diagnosis was performed by two geriatricians blinded to PMIS and IDEA cognitive screen scores using the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-V) clinical criteria. The MMSE performance was used as a reference.
RESULTS: The study enrolled 66 participants, with a median age of 78.5 years (interquartile range [IQR], 72.5-83.0) years and 11.0 median years of education (IQR, 9.0-13.0). Forty-three (65.2%) were female, and 32 (48.4%) were Chinese. The area under the receiver operating characteristic (AUROC) curve values were .962 (IDEA cognitive screen), .970 (PMIS), and .935 (MMSE). The optimal cutoff values for sensitivity and specificity were: IDEA cognitive screen: ≤ 11, 90.9% and 89.7%; PMIS: ≤ 6, 97.3% and 69.0%; and MMSE: ≤ 23, 84.6% and 76.0%. Although the sample size was small, multivariable logistic regression modelling suggested that all three screen scores did not appear to be educationally biased.
CONCLUSION: The IDEA and PMIS tools are potentially valid screening tools for dementia in urban Malaysia, and perform at least as well as the MMSE. Further work on larger representative, cohorts is needed to further assess the psychometric properties.
CLINICAL IMPLICATIONS: Study provides alternative screening tools for dementia for both non-specialists and specialists.
* Title and MeSH Headings from MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine.