Comparison of two scoring systems of the Mini-Mental State Examination as a screening test for dementia

Abstract

The difference in validity in detecting dementia of two different scoring systems (age- and education-adjusted versus raw scores) of the Mini-Mental State Examination (MMSE), was assessed in an epidemiological study on the prevalence of dementia. A second, independent, psychometric screening tool, the Milan Overall Dementia Assessment (MODA), was also used. A group of 829 subjects over age 59 took part in the study. Subjects scoring below the cut-off point of the MMSE and/or MODA were clinically evaluated. Clinical diagnoses were adopted as the gold standard. Sensitivity, specificity, positive, and negative predictive values of the two tests were calculated. Our findings show that the correction for age and education of the raw score of the MMSE causes the loss of the mildest demented patients and heavily reduces the sensitivity of the screening test (from 85.7% to 71.4%), while it accounts for a moderate improvement in the specificity (from 90.0% to 96.3%). A possible explanation for this seeming paradox is proposed.