Reliability Assessment of the Defense Automated Neurobehavioral Assessment (DANA) in Extreme Environments

Abstract

Abstract : Background: The purpose of this effort was to report the internal consistency of Defense Automated Neurobehavioral Assessment (DANA) in baseline condition and to estimate the reliability of the DANA within four simulated environmental conditions. Methods: On four non-consecutive days, 16 healthy U.S. Navy personnel were exposed to a baseline condition and three operationally relevant simulated environmental conditions. After 25 minutes in the simulated environment participants began the neurocognitive assessment consisting of two consecutive administrations of the DANA Brief Test Battery. Throughput was calculated as the number of correct responses per mean reaction time. Internal consistency was estimated from the ambient environmental condition using Cronbachs alpha. Reliability was established across the environmental conditions using intra-class correlational coefficients (ICC) two-way, average measures, random effects models. Absolute reliability was established across the environmental conditions with the standard error of the measure (SEM). Results: The mean internal consistency within the baseline condition was 0.79. The mean (i.e., across subtests) ICC was 0.86 and the mean SEM was 8.71. Conclusions: The overall DANA subtests demonstrated adequate internal consistency and acceptable-to-excellent relative reliability across the four simulated environmental conditions with four subtests surpassing the desired reliability threshold for clinical decision-making (ICC 0.90). These findings add to the growing body of evidence regarding the psychometric properties of the DANA subtests.