2012 National Meeting

3020 — Validity of a Population-Based Screen for TBI: Comparing the VA’s TBI Screen to Follow-Up Evaluation

Belanger HG, HSR&D's Maximizing Rehabilitation Outcomes Center of Excellence, Tampa VAMC; Vanderploeg RD, COE Tampa; Soble JR, Tampa VAMC; Richardson M, and Groer S, COE Tampa;

Objectives:
The objective was to compare the results of the VA’s initial TBI Clinical Reminder screen with the more extensive second level Comprehensive TBI Evaluation in a national sample. The intent was to investigate the extent to which population-based screening, based only on four questions, provides a valid means of establishing a history of mild TBI. Secondary objectives were to examine the extent to which the concurrence between the initial screen and the Comprehensive TBI Evaluation are related to other factors such as demographic variables, symptom-related distress, and level of care.

Methods:
The results from the TBI Screen for 48,175 individuals were pulled from the centralized database. In addition, the corresponding results of Comprehensive follow-up TBI Evaluations were collected. Accuracy statistics were calculated (kappa, positive and negative predictive values, etc).

Results:
In the overall sample, 28,337 individuals screened positive for TBI. Of those that screened positive, 52% were deemed to have sustained a TBI following the Comprehensive TBI Evaluation. If those cases that should have been deemed to be consistent with TBI are included (i.e., the clinician indicated some alteration of consciousness, but did not endorse ‘TBI’), then the prevalence is higher at 77%. The VA’s TBI screen has good sensitivity at the expense of specificity. Overall there was poor agreement between the screen and the subsequent evaluation, regardless of patient characteristics. There was slightly greater agreement at Polytrauma Rehabilitation Centers, as opposed to other levels of Polytrauma care, greater agreement when just the 2nd question from the screen was used, and greater agreement when inconsistencies were eliminated. The symptom complex of the majority of positive screens were ultimately deemed to be due to either TBI (36%) or behavioral/mental health issues (45%). Interestingly, only about 9% of positive screens were thought to be symptomatic due to both TBI and behavioral/mental health issues.

Implications:
The VA may wish to consider just using the first two questions of the current screen. In addition, auto-fill mechanisms might be utilized on the computer screen to improve consistency of reporting within the comprehensive evaluation process. Mental health was perceived as the most typical cause of ongoing symptoms in positive screens.

Impacts:
The VA screen is sensitive but not specific. Findings from this study suggest specific ways to improve the screening process.