The Mini‐Mental State Examination

To the Editor:—I read with interest the comprehensive review on the Mini-Mental State Examination by Tombaugh and McIntyre.1 At the memory clinic here, we have also assessed the objective value of the Mini-Mental State Examination (MMSE) by examining the correlation of the score on the MMSE in forty-seven elderly patients with dementia with two computerized automated psychometric tests (classification task and digit-recall test). The classification task tests recognition and discrimination as well as speed of performance; the digit-recall test examines short-term retention and learning ability. Both of these tests had been shown by us2 to correlate well with the latency of P300, an event-related potential, regarded as an objective measure. Correlation coefficient between the MMSE and the classification task in patients with dementia of Alzheimer's type (n = 27) was 0.42 (P < 0.03) and the digit-recall task was 0.79 (P < 0.001). In patients with multi-infarct dementia (n = 20), the correlation coefficient between the MMSE and the classification test was 0.627 (P < 0.002), and between MMSE and the digit-recall test it was 0.64 (P < 0.003). The data from this study have been published in the Journal of Clinical and Experimental Gerontology.3