Date Submitted
8-2023
Faculty Advisor
Esperanza Anaya, PhD
Second Faculty Advisor
Leah Barreca, PhD
Third Faculty Advisor
Michael Kolessar, PsyD, ABPP
Abstract
This study examined the convergence validity and performance characteristics of the Trails-X assessment in a dementia population, using TMT A and B, a discriminability and difficulty index, and by exploring the clinical dementia population in the Trails-X manual and this studies dementia population. Hypothesis one proposed a minimal convergence between Trails-X and TMT A, indicated by a weak but statistically significant correlation and no relationship in error scores. Hypothesis two proposed a moderate or better convergence between TMT B and Trails-X, indicated by a moderate or better statistically significant correlation, and a weak relationship in the measure’s errors. Hypothesis one and two varied by population but demonstrated convergence in several groups. Particularly those on the opposite ends of cognitive ability. While those with no diagnosis or a diagnosis of major neurocognitive disorder demonstrated a more consistent relationship between the measures, the mild population demonstrated more variability and, therefore, a less direct relationship. Similarly, findings on the difficulty and discriminability indexes showed all nine trials of Trails-X do not differ in discriminability and are equally difficult for a dementia population. These results suggest Trails-X may not pick up on milder and more nuanced changes in cognition. Additionally, data from this study were compared to the clinical dementia population in the Trails-X manual and similarities were found between those in this clinical population with the Major Neurocognitive Disorder group. In contrast, the mild and no diagnosis groups were found to be significantly different from the clinical group, but similar to each other.
