When applied post hoc to an existing population, the Mini-Cog was as effective in detecting dementia as longer screening and assessment instruments. Its brevity is a distinct advantage when the goal is to improve identification of older adults in a population who may be cognitively impaired. Prior evidence of good performance in a multiethnic community-based sample further supports its validity in the ethnolinguistically diverse populations of the United States in which widely used cognitive screens often fail.

Impaired vision and depression are both associated strongly with functional impairment in this community population of older adults. Depression, however, increases the odds of functional impairment independent of vision impairment. Treating depression may reduce excess disability associated with impaired vision.

Although Alzheimer's disease is widely regarded as a leading cause of death, dementias are reported on the death certificates of only a quarter of demented individuals in the population at large. Reporting is more likely in those with more advanced dementia, with Probable Alzheimer's disease, and those who die in long-term care institutions.

This paper reports on a comparison of the two alternative tests of attention in the Mini-Mental State Examination (MMSE), a well-known cognitive screening tool. The two tests, serial subtraction by seven and backwards spelling of the word world, are often used interchangeably. In a large population-based sample, the two tests were found to be weakly associated with each other, with the former test appearing more difficult, although both were strongly associated with educational level. The authors discuss the implications of this finding in clinical and research settings, and make recommendations for more consistent use of the instrument.

A bstractIn an epidemiological survey of a rural, largely blue-collar, community, 1,363 randomly selected adults, aged 65 1 years, were administered a cognitive screening battery (including in part the CE RAD neuropsychological tests): M ini-Mental State Examination; W ord List Learning, Recall, and Recognition; Story, Immediate and Delayed Recall; Boston Naming Test; Verbal Fluency; Tem poral Orientation; Constructional Praxis; Draw a Clock; and Trailmaking. Cognitively impaired subjects and cognitively intact controls underwent independent standardized diagnostic assessments and were rated on C linical Dementia Rating (CD R) scale. Overall, subjects at higher CD R levels (m ore severe dementia) had worse scores on all tests; showing that standard neuropsychological tests are valid for characterizing the cognitive impairments seen in dem entia, even in comm unity settings. However, non-demented scores on the CERAD tests in this community-based sample were lower than reported from CERAD' s pooled healthy controls from Alzheimer' s Disease Centers (ADCs) nationwide. Thus,`normal' scores from specialty dem entia clinics, where there m ay be a selection bias, m ay differ from normative scores from rural and/or less-educated populations. Patients from such populations m ay be functionally intact despite low test scores. C omm unity-based studies are required to complem ent specialty clinic-based studies of dementia and cognitive functioning.