Objective: To study the sensitivity and specificity of the Abbreviated MMSE in Alzheimer patients with normal MMSE scores.

Materials and Methods: A retrospective analytic study of 89 normal subjects and Alzheimer’s patients with normal MMSE scores, aged 60 years or older, between 2015 and 2020, from the Memory Clinic, Ramathobodi Hospital. Pearson chi-square, Fisher’s exact test, and Mann-Whitney U test were used to compare demographic data and MMSE score of each item between the normal subjects and the Alzheimer Disease (AD) subjects. The ROC curve was analyzed to find the proper cut-off points in the three models of the Abbreviated form (p<0.05).

Results: AD subjects showed different scores from the normal subjects with statistical significance in four subtests of full MMSE-Thai-2002, namely, date, attention/calculation, recall, and repetition (p<0.001, 0.001, 0.018, and 0.023, respectively). The Model 2 of the Abbreviated MMSE consisting of recall, date, and attention/calculation, was the most proper test to be used in the clinical setting. [AUC (95% CI): 0.905 (0.831 to 0.979)]. The proper cut-off point was seven out of nine with a sensitivity of 89.4%, specificity of 78.3%, of PPV 92.2%, and NPV of 72.2%.

Conclusion: The models of short MMSE was an option in case of normal scores in the full version of MMSE with recall impairment. There should be further studies in larger number of Alzheimer’s patients, along with classification of daily life skills and other groups of brain diseases’ patients.

Keywords: Alzheimer’s disease; Abbreviated-MMSE; Normal MMSE scores

DOI: 10.35755/jmedassocthai.2022.12.13701

Received 20 June 2022 | Revised 20 September 2022 | Accepted 4 October 2022