| Description |
Executive functioning (EF) is the neurocognitive domain most strongly associated with performance of instrumental activities of daily living (IADLs). However, performance-based tests of EF have long been criticized for having poor ecological validity, largely because of their imperfect association with functional outcomes and their low face validity. However, past research has not examined whether high face validity improves a test's ability to predict IADLs. The present study examined this question. We pitted the Pillbox Test (i.e., a measure with high verisimilitude) against the Delis-Kaplan Executive Function System (D-KEFS; a measure of EF that has low verisimilitude) with respect to their ability to (a) predict actual at-home management of medications among community-dwelling older adults and (b) classify participants based on the degree to which they mismanaged their medications at home across the 8 weeks. As a secondary aim, we sought to establish the degree to which a commonly used laboratory-based behavioral measure of IADLs, specifically Timed Instrumental Activities of Daily Living (TIADL), reflects actual daily functioning. Fifty community-dwelling, older adults completed the Pillbox Test, D-KEFS, and TIADL. Weekly pill counts were conducted as an index of real-world daily functioning. Analyses using medication management as the dependent variable revealed that (a)the D-KEFS, the completion time (but not errors) on the Pillbox Test, and thecompletion time (but not errors) on TIADL were significantly related to medication management after controlling for age, (b) although the time to complete the Pillbox Test accounted for additional variance in medication management beyond age and the D-KEFS, the D-KEFS proved to be superior in terms of classifying participants based on their difficulties managing their medications, and (c) time to complete TIADL did not account for additional variance in medication management beyond age and the D-KEFS, or beyond age and time to complete the Pillbox Test. The present study does not support the notion that face validity, in and of itself, improves the ability of performance-based EF tests to predict functional outcomes. The results also suggest that time to complete TIADL may be considered an equal predictor of at-home functioning as tests of EF. iv |
