A Rasch Analysis of the Neurobehavioral Symptom Inventory in a National Cohort of Operation Enduring and Iraqi Freedom Veterans With Mild Traumatic Brain Injury

This study investigated the dimensionality and measurement properties of the Neurobehavioral Symptom Inventory (NSI), a 22-item questionnaire of postconcussive symptoms, in a national sample of 9,679 Operation Enduring Freedom and Operation Iraqi Freedom (OEF/OIF) Veterans with mild traumatic brain injury. Dimensionality of the NSI was examined using exploratory factor analysis (EFA) and Rasch analysis. Rasch measurement properties, including overall fit, unidimensionality, item level statistics, and category functioning were examined for individual NSI symptom domains identified through EFA. Differential item functioning (DIF) analyses of subgroups based on gender and Posttraumatic Stress Disorder (PTSD) diagnosis were also conducted. Results showed that the NSI items are multidimensional and responses can be differentiated into 3 unidimensional domains reflecting Cognitive, Mood-Behavioral, and Vestibular-Sensory symptoms. All items in each of the 3 subscales demonstrated adequate fit to the Rasch model. In each domain, the item hierarchy was consistent with expectations regarding the theoretical ordering of symptoms. Some problems were observed regarding test targeting for all 3 subscales, such that items were generally concentrated around the mean ability level of the population. As such there were fewer items that differentiated between those at the upper and lower extremes of the scale. Differential item functioning (DIF) based on gender was noted for hearing difficulty. This item on average reflected a higher degree of severity for women than for men. There was no DIF based on PTSD status. Implications for using the NSI as an outcome measure are discussed. Public Significance Statement This study suggests that the Neurobehavioral Symptom Inventory (NSI) is a multidimensional measure, and that the total score should not be used as a global measure of postconcussive symptoms. Instead items should be differentiated into Cognitive, Mood-Behavioral, and Vestibular-Sensory symptom domains. In addition, using Rasch interval scores instead of raw scores may allow for the assessment of interventions at more specific levels, and enhance measurement precision for individuals with different levels of symptom severity.