Correlates of disability among elderly retirees with "subclinical" cognitive impairment

Background We assessed the effects of impaired Executive Control Function (ECF) on Instrumental Activities of Daily Living (IADL) and level of tare among noninstitutionalized elderly retirees with "subclinical" cognitive impairment. Methods. Subjects (N = 561; age 78.2 +/- 5.0 years) were residents of a single, 1500 bed, continuing care retirement community. Subjects were examined for cognitive impairment using the Executive Interview (EXIT25), Mini-Mental State Examination (MMSE), and an executive clock-drawing task (CLOX). The CLOX is divided into executively sensitive (CLOX1) and simple constructional (CLOX2) subtests. Results. Residents in congregate high-rises (n = 301) differed significantly From those in independent-living apartments (n = 260) with respect to age, gender, percent living alone, EXIT25, CLOX1, MMSE, and CLOX2 scores (all P < .03). Only differences in ECF measures persisted after adjusting for age and living alone (p < .004). The EXIT25 (P < .006) and CLOX2 (p = .02) were associated with the use of prostheses. The differences in EXIT25 scores persisted after adjusting for level and living alone (p = .01). All instruments distinguished residents with impairment in IADLs. However, only CLOX2 (p < .001), EXIT25 (p < .001), and age (p < .001) made significant independent contributions. Conclusions. ECF has statistically significant effects on level of care and IADL impairment, even among noninstitutionalized retirees. This emergent disability is not well detected by traditional global cognitive measures. Evaluation and treatment may he delayed unless ECF measures are employed.