Mini-Mental State Examination: diagnostic test accuracy study in primary care referrals

Abstract

Aim: To undertake a diagnostic test accuracy study of the Mini-Mental State Examination (MMSE) administered in primary care to patients who were subsequently referred to a cognitive disorders clinic in secondary care (n = 72). Methods: MMSE scores from primary care were cross-classified with reference standard diagnoses made in secondary care, blind to MMSE score, in order to calculate standard measures of discrimination (including sensitivity and specificity, positive and negative predictive values). Results: MMSE showed poor sensitivity (0.64) but better specificity (0.80) for diagnosis of any cognitive impairment at the index paper specified cut-off, with little additional benefit using a more stringent cut-off. Conclusion: These data suggest MMSE is not suitable for screening for cognitive impairment in the low prevalence setting of primary care.