Natural history of mild cognitive impairment in older persons.
Bennett DA, Wilson RS, Schneider JA, Evans DA, Beckett LA, Aggarwal NT, et al.;
Commented by , 29 Sep 2002
Aim of the study
Compare the risk of death, risk of incident Alzheimer’s disease (AD) and rates of change in cognition between persons with or without mild cognitive impairment (MCI).
Method
A cohort of older Catholic clergy was recruited from 40 groups across the USA between 1994 and 2000, with annual clinical and neuropsychological evaluations. Diagnosis for MCI was based on cognitive impairment by neuropsychologist but no dementia for the clinician. Longitudinal analysis were based on persons who completed an average of 5.5 clinical evaluations (range 2-8).
Results
Of the 798 persons in this study, 211 (26.4%) had MCI relative to persons without cognitive impairment they were slightly older (78.6 vs 74.3) and lesser educated (17.9 vs 18.3), and Mini Mental State Examination Scores were 27.4 vs 28.8.
In a proportional hazards model adjusted for age, sex and education, persons with MCI were 1.74 times more likely to die (95% CI, 1.22 to 2.48) and 3.17 times more likely to develop AD (95% CI, 2.16-4.64). Whereas the measures of episodic memory, semantic memory and perceptual speed showed a faster rate of decline in MCI, the rate of decline for measures of working memory and visuospatial ability did not differ from persons without cognitive impairment.
Discussion
The large sample size, relatively long follow-up and high rates of participation in this volunteer cohort add to the strength of the results, which may not be directly application to the aging population as a whole. As stated by the authors of this study, data on the proportion of elderly persons with MCI and their rates of development of AD will need to be estimated with confidence from longitudinal population-based studies.
th 2001 by Dr Robert Temple and summarized in the Pharmaceutical Approvals Monthly of April 1st 2001 that a community-based randomized clinical trial could a useful adjunct to results observed in research settings.