A follow-up study of age-associated memory impairment: neuropsychological predictors of dementia

Abstract

Objective: To examine the clinical course of age-associated memory impairment (AAMI) and to evaluate the value of neuropsychological tests in predicting cognitive decline in AAMI subjects in a follow-up period of more than 3 years.

Design: Prospective cohort study.

Setting: The outpatient Memory Research Unit of the Department of Neurology at the University of Kuopio in Eastern Finland.

Participants: A sample of 229 subjects (mean age 71.7 years) identified in two screening studies as having AAMI.

Measurements: A battery of neuropsychological tests and a structured inquiry for health status and subjective memory complaints were performed at baseline and follow-up to diagnose AAMI according to the criteria proposed by a National Institute of Mental Health work group.

Results: Of the 229 subjects, 176 (76.9%) participated in the follow-up for, on average, 3.6 years after the baseline. Of the participants, 104 (59.1%) still met the AAMI criteria. Other subjects were classified into five subgroups: (1) subjects showing decline in cognition meeting dementia diagnosis (16, 9.1% (13 of them AD)); (2) subjects with mild cognitive decline meeting neither dementia nor AAMI criteria (13, 7.4%); (3) subjects with memory performance now superior to AAMI criteria (17, 9.7%); (4) subjects having a disease classified as exclusion in the criteria (15, 8.5%); (5) subjects not now reporting subjective memory loss in everyday life (9, 5.1%). Two subjects (1.1%) were not classified because of incomplete data. Neuropsychological tests predicted which subjects would develop dementia during the follow-up period. The best discriminators between these subjects and those who remained AAMI were memory and verbal fluency tests.

Conclusion: The study suggests that, in general, AAMI is nonprogressive, but the AAMI population also includes subjects with early dementia and subjects without genuine memory loss. However, these subjects can be differentiated with a more detailed neuropsychological evaluation.