LAUSR

INTRODUCTION The Mini-Mental State Examination (MMSE) is widely used for dementia screening. The purpose of this study was twofold. First, to cross-culturally adapt the Moroccan Arabic version of MMSE (MA-MMSE) to screen dementia among literate and illiterate elderly Moroccans, and second to analyze its psychometric properties. METHODS The MA-MMSE was administered to 80 elderly patients, 28 were females and 52 were males. Among all the participants, 22 showed evidence of dementia. Receiver Operator Characteristic (ROC), internal consistency, and test-retest reliability were investigated. The test-retest reliability was assessed by Spearman correlation analysis. Cronbach's α coefficients for internal consistency between total scores and subscores of the test were calculated. RESULTS Areas under the ROC curve in literate and illiterate elderly Moroccans participants corresponded to 0.932 and 0.936, respectively. The cutoff point of 21/22 on the MA-MMSE in literate elderly had the highest sensitivity (85.7%), specificity (94.7%), and a positive likelihood ratio of (16.29%), whereas the cutoff point of 18/19 of the screening test in illiterate elderly yielded the highest sensitivity (92.9%), specificity (95.0%), and a positive likelihood ratio of (18.57%). The Cronbach's α values of the MA-MMSE for literate and illiterate elderly Moroccans were higher than 0.7. Thirty subjects were randomly selected for participation in the test-retest protocol. Participants were retested after an interval of 4-6 weeks. The test-retest reliability for total scores was significant 0.78 (p < 0.01). CONCLUSION The findings provide preliminary evidence that the MA-MMSE is a feasible instrument for dementia screening.