COGNITIVE FUNCTION IN OLD AND VERY OLD RESIDENTS OF A RESIDENTIAL FACILITY - RELATIONSHIP TO AGE, EDUCATION, AND DEMENTIA

OBJECTIVE: To determine if age, education, and dementia status affect neuropsychological performance in old and very old frail residential c are subjects. DESIGN: Descriptive study of performance at the time of preadmission assessment. SETTING: Jewish Home for the Aging, Reseda, C alifornia. PARTICIPANTS: 201 applicants to the Jewish Home for the Agi ng residential care setting. Mean age was 84.7 years; SD was 5.6. Nine ty-five subjects were 84 years of age or younger, while 106 were age 8 5 and older. There were 141 nondemented, 21 demented, and 39 were poss ibly demented applicants. Levels of education were as follows: 0-4 yea rs: n = 25; 5-8: n = 69; 9-12: n = 77; and, 13-20: n = 23. MEASUREMENT S: Independent variables were age, education, and dementia status. Out come measures were Folstein MMSE, Inglis P-A Learning Test, Digit Span , Cube Copying, selected Boston Diagnostic Aphasia Exam subtests. RESU LTS: Subjects with 0 to 4 years of education scored more poorly on cog nitive tests than other subjects. The very old tended to score more po orly than the old. Neuropsychological tests discriminated between thos e with normal cognitive function, possible dementia, and established d ementia. About one-third of nondemented elderly scored below the tradi tional impairment cut-off of 24 points on the Mini-Mental State Exam. CONCLUSIONS: Questions are raised about how to interpret the poorer co gnitive performance of very old and often frail subjects, especially i n long-term-care settings where there are fewer demands upon residents whose impairments might otherwise cause them more functional difficul ty.