Screening for mild cognitive impairment (MCI) utilizing combined mini-mental-cognitive capacity examinations for identifying dementia prodromes.Int J Geriatr Psychiatry 2002; 17(11):1027-33IJ
To test correctness of results when combining Mini-Mental State Examination (MMSE) and Cognitive Capacity Screening Examination (CCSE) for identifying mild cognitive impairment (MCI) among non-demented elderly subjects at risk for developing dementia.
A retrospective study was conducted among consecutively referred volunteers with memory complaints to a research out-patient clinic. Two cognitive screening tests (MMSE and CCSE) were performed according to established protocol. Resulting combined screening test (termed by acronym as CMC) combined the non-overlapping test items derived from both MMSE and CCSE. Conversion to dementia at follow-up served as the 'gold-standard' for evaluating correctness of CMC for identifying MCI.
Of 351 subjects completing cognitive assessments and meeting requirements for study protocol, 84 (23.9%) developed dementia of different types within 3-6 years (3.89 +/- 2.17) of follow-up. Among these, 47 met criteria for probable Alzheimer disease (AD), 22 for probable vascular dementia (VaD), 12 for mixed AD/VaD and three for probable frontotemporal dementia. When final diagnosis of AD was used as the 'gold standard' for testing correctness of MCI identified by cognitive screening tests, sensitivities of MMSE, CCSE and CMC for identifying MCI were relatively 61.0%, 74.3% and 83.1% with minimum specificity set at 80%. When diagnosis of all types of dementia was used as the standard for testing predictive correctness of MCI, CCSE emerged as an optimal MCI screening test.
Combining the CCSE and MMSE screening tests resulted in higher sensitivity than was achieved by MMSE alone and maintained specificity at comparable levels for identifying MCI. The results confirmed that CMC has optimal correctness and utility as a brief cognitive test for screening MCI as a prodrome for dementia among non-demented elderly populations.