W 30: Implementing Neurocognitive Testing in Clinical Trials: Facilitating Rater Administration With an iPad-Based App
Brian K. Saxby
Vice President, Business Development
Neurocog Trials Inc. United States
To determine the validity of a newly-developed iPad-based App to facilitate test administration of an established neurocognitive test battery in clinical trials of challenging patient populations.
48 patients (23 female) with schizophrenia and 50 healthy controls (25 female) were recruited from 3 US academic sites. Participants were assessed with both the pen-and-paper Brief Assessment of Cognition (P&P BAC) and the iPad-based application (BAC App), in a counterbalanced order.
In both groups, the distributions of standardized composite scores for the P&P BAC and BAC App were indistinguishable, and the between-methods means were not significantly different. Discrimination was similarly robust, with the P&P BAC (d=1.24) and BAC App (d=1.34) both demonstrating large effect sizes of deficits in patients compared to controls. The between-methods correlations for individual measures in patients were r>.70 except Token Motor (r=.43) and Tower of London (r=.61). In patients, performance between the test methods was not significantly different on any test, except the Token Motor Test, where the mode of performance is qualitatively different. When data from the Token Motor Test were excluded, the correlation of composite scores improved to r=.88 (p<.001) in healthy controls and r=.89 (p<.001) in patients, consistent with the test-retest reliability of each measure.
To facilitate the administration of a widely-used pen-and-paper neurocognitive battery, the Brief Assessment of Cognition (BAC), we developed an iPad-based application, the BAC App. The purpose of the BAC App is to reduce administration burden on site raters by automating and standardizing the testing procedures and scoring, while maintaining the importance of the rater-patient interaction. Ensuring full understanding of the task at hand, motivation to try their best, and providing general encouragement to complete the testing is critical to obtaining meaningful data in impaired or behaviorally-challenging patient populations such as Alzheimer’s Disease, depression, ADHD and schizophrenia. The data from this validation study confirm the equivalence of the mode of administration between the P&P BAC and BAC App for all subtests, except Token Motor. The applicability of the established normative data is also confirmed. Therefore, tablet-assisted rater administration using the BAC App is a viable option for use in future clinical trials to reduce site burden while maintaining the validity of the neurocognitive outcomes.
Authors: Saxby, BK (1); Atkins, AS (1); Davis, VG (1); Tseng, T (1); Vaughan, A (1); Harvey, P (2); Narasimhan, M (3); Patterson, T (4) Keefe, RSE (1, 5)
(1) NeuroCog Trials, Durham, NC; (2) University of Miami Miller School of Medicine, Miami, FL; (3) University of South Carolina School of Medicine, Columbia, SC; (4) University of California, San Diego, School of Medicine, San Diego, CA; (5) Duke University Medical Center, Durham, NC