Improving the diagnostic accuracy of the Mini-Mental State Examination.

Overview

INTRODUCTION: We determined the diagnostic accuracy of the Mini-Mental State Examination (MMSE) for dementia of the Alzheimer type (DAT) in an outpatient geriatric referral center in Switzerland. MATERIAL & METHODS: DAT patients and elderly controls were assigned to two groups: a validation sample (70 DAT patients; 50 controls) and a cross-validation sample (133 DAT patients; 43 controls). A Receiver Operating Characteristic curve was generated to derive the optimal MMSE cut-off score in the validation sample. RESULTS: The optimal MMSE cut-off was < 26/30 (sensitivity of 74%, specificity of 100%). Adjustments for age and education were necessary. The cross-validation confirmed these findings. CONCLUSION: iN A clinical setting the MMSE cut-off should be increased to < 26/30. A thorough neurobehavioral assessment is still necessary for a complete evaluation.