Performance and Receiver Operating Characteristics of the Mini

To examine the performance of the Mini-Mental State Examination (MMSE) in community-dwelling older persons in a developing country (Jamaica) undergoing rapid population aging. An embedded validity study was conducted utilizing participants from a nationally representative sample of 2782 older persons. Standardized MMSE scores were obtained for study participants. A random selection of 170 persons with MMSE scores greater than 20 and 170 persons with scores 20 or less was done. Field staff were trained to apply the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, criteria for the diagnosis of dementia. In total, 300 participants (167 participants with MMSE score of 20 or less, 133 participants with scores greater than 20) were assessed and categorized according to dementia status. Performance characteristics of the MMSE tool were determined for study participants and appropriate adjustment and analyses subsequently applied to facilitate extrapolation to the nationally representative sample. The mean MMSE scores for participants with score of 20 and less was 17.1 (standard deviation [SD] = 3.2) and 24.5 (SD = 2.8) for those with scores greater than 20. Dementia was identified in 34 participants. The receiver operating characteristic curve for MMSE scores in relation to dementia diagnosis had an area under the curve value of 0.935 (95% confidence interval, 0.893-0.977). The optimal MMSE cut-point was 18/19 and was consistently so regardless of age category, gender, educational level, and number of chronic illnesses. There is merit in using the MMSE examination as a screening tool for dementia in Jamaica. The findings of this study coupled with widespread use and familiarity among practitioners give credence to the MMSE as a reasonable screening tool for dementia in Jamaica—rapidly aging society