A-225 The Pillbox Test: Does It Have Sufficient Construct Validity as a Measure of Instrumental Activities of Daily Living?

Abstract

Abstract Objective: The Pillbox Test has demonstrated adequate criterion validity and the ability to discriminate between patients with and without dementia. The purpose of this study is to establish the construct validity of the Pillbox Test as a measure of instrumental activities of daily living (IADL) abilities. Method: A clinic-referred sample of veterans who passed performance validity tests and completed the Pillbox Test, Texas Functional Living Scale (TFLS), and Geriatric Depression Scale, Short Form (GDS-SF) were selected for inclusion. This resulted in a sample of 62 veterans (93.5% male; 51.6% White, 32.3% Hispanic/Latinx, 16.1% Black; 71% Monolingual [English], 29% Bilingual [English/Spanish]). Spearman correlation and Mann-Whitney U tests were conducted due to non-normal distributions. Results: There was a moderate negative association between errors on the Pillbox Test and overall performance of functional living skills assessed by TFLS, r(60)=-.598, p<.001. There was not a significant relationship between Pillbox Test errors and self-reported mood symptoms on GDS-SF, r(60)=-.109, p=.401. Similarly, non-parametric analyses revealed a significant difference in performance on TFLS between those who passed the Pillbox Test relative to those who failed (p<.001) and no difference on GDS-SF report between the Pillbox Test pass/fail groups (p=.394). Conclusions: The Pillbox Test demonstrated adequate construct validity among a diverse older-adult sample. Specifically, the observed moderate correlation with TFLS supports convergent validity and the weak correlation with GDS-SF supports divergent validity. Compared to those who passed, those who failed the Pillbox Test had lower TFLS scores but similar GSD-SF scores. Findings support the Pillbox test as a measure of IADL performance.