Velozo CA (Rehabilition Outcomes Research Center REAP - North Florida/South Georgia Veterans Health System), Wen PS
(University of Florida), Heaton S
(University of Florida), Waid-Ebbs K
(North Florida/South Georgia Veterans Health System), Donovan N
(Louisiana State University), Garretson K
(University of Florida)
Objectives:
The increased number of OEF/OIF veterans returning with TBI demands a battery of assessments that captures the scope of cognitive deficits and the impact of these deficits on everyday life. The primary objective of this study was to do the foundational work for the development of a Computer Adaptive Measure of Functional Cognition for Traumatic Brain Injury (CAMFC-TBI).
Methods:
An item pool of 228 items across six domains of functional cognition (attention, memory, processing speed, executive function, emotional management, and social communication) were administered to 227 participants (90 patient self-ratings – 47 outpatient and 43 one-year post injury, 89 caregiver ratings, and 48 healthcare professional ratings). Patients also were administered a battery of neuropsychological/functional measures. Unidimensionality was assessed using confirmatory factor analysis (CFA), and item-level psychometrics were assessed with Rasch analysis. Relationships across raters and concurrent validity were examined using correlations.
Results:
Unidimensionality could not be confirmed for the domains. Findings included: 79-97% items fit the Rasch model; person separation 1.92-4.92; Cronbach’s alpha 0.92-1.00; no floor effects; only emotional management showed a ceiling effect. Patient ratings showed moderate correlations with caregiver ratings for all domains except attention and caregiver ratings showed moderate correlations with healthcare professionals for memory and executive function. Patient ratings showed weak correlations with healthcare professional ratings. While the attention domain was poorly correlated with Digit Span for all raters, memory and processing speed domains showed stronger correlations with analogous neuropsychological measures, especially for healthcare professionals. Moderate to strong correlations were shown for the patient and caregiver ratings of attention, memory, executive function, and social communication domains and analogous measures of functional cognition.
Implications:
While unidimensionality could not be confirmed for the domains, the battery showed acceptable item-level psychometrics. Concurrent validity was most evident with external measures of functional cognition.
Impacts:
OEF/OIF veterans receiving medical care within the VA screening positive for possible TBI are offered further evaluation and treatment by clinicians with expertise in TBI. Self-proxy reports of functional cognition may be useful in determining the impact of blast injuries on the cognitive challenges faced by these individuals in every day life.