The purpose of this study is to evaluate the reliability and validity of the existing Traumatic Brain Injury (TBI) Clinical Reminder Screen for OEF/OIF Veterans.
1. Operationalize a gold standard for TBI identification.
2. Identify VHA system factors and patient characteristics predicting delay in or failure to complete the TBI Clinical Reminder screen (e.g., patient characteristics and VA System levels of Polytrauma care).
3. Using the gold standard, evaluate the validity (sensitivity and specificity) and reliability of the current TBI Clinical Reminder Screen, as well examine the effects of psychiatric comorbidities, veteran characteristics (age, gender, race) and facility (level of Polytrauma System of care, urban versus rural) on the validity of TBI identification.
4. Identify approaches to improve the TBI Clinical Reminder screening protocol, including screening instrument and process.
This proposed two year study will employ a mixed methods research design addressing four objectives. Both retrospective, population-based cohort analyses of existing databases, as well as prospective data collection will complement each other to assess the reliability and validity of the TBI Screen.
Objective 1: A panel of experts in TBI was convened to establish a gold standard semi-structured clinical interview for assessing whether a TBI has occurred. This gold standard will then be used in the subsequent reliability and validity of the TBI Screen.
Objective 2: Identify provider, patient, and system characteristics predicting failure to successfully complete the TBI screening process: The database available from Patient Care Services Office of Strategic Planning and Measurement will be used to identify veterans who are eligible to be screened and enrolled in the VHA healthcare system. Additional information for this cohort of veterans will be accessed from databases available from Austin (e.g. National Patient Care Database). Databases will be accessed and used in a retrospective analysis allowing comparisons of patients who successfully completed the screening process in a timely manner versus those in whom there were delays. As the database includes information from all OEF/OIF veterans who have been seen within the VA Healthcare System, all records will be used. Threats to Validity.
Objective 3: Using the Gold Standard, evaluate the validity (sensitivity and specificity) and reliability of the current TBI Clinical Reminder: A prospective study will be completed on a sample of veterans at three levels of Polytrauma care (PRC, PNS, PSCT) to assess the psychometric characteristics of the TBI screen.
Objective 4: Improve the current TBI Clinical reminder: We will examine each of the questions and response options within the screen to determine which are most related to correct gold standard identification versus result in false positive responses. In addition, provider interview versus paper questionnaire administration methods of completing the TBI Clinical Reminder will be compared. For this objective the prospective data will be used.
Based on an expert panel of TBI specialists, we developed a "VA TBI Identification Clinical Interview" tool (interview form and accompanying manual) for the purposes of this study. This tool had good inter-rater reliability (kappa = .854) with agreement on 28 / 29 (97%) of the cases examined by two research physicians. This tool is now available for use by VA clinicians.
VA's TBI Clinical Reminder Screen, when evaluated against this criterion TBI Indentification interview standard, demonstrated fair sensitivity (.63) and good specificity (.89), and good negative predictive power (.89 - .91). No modifications of individual items or response choices in the VA's TBI Clinical Reminder Screen resulted in improved sensitivity or specificity.
VA's TBI Clinical Reminder Screen, when evaluated against the follow-up Comprehensive TBI Evaluation, has good sensitivity (.87-.90) but poor specificity(.13-.18). In addition, the TBI Clinical Reminder Screen has generally good negative predictive power with an estimated Veterans Health Administration system-wide TBI prevalence rate of 15% (.89). Positive predictive power was acceptable (.77). The screen performs comparably across patient demographic and symptom severity characteristics, as well as across level of polytrauma care.
The VA screen is sensitive but not specific. Findings from this study suggested that modifications of individual items or responses within the existing TBI Clinical Reminder fail to improve the screening process. A semi-structured interview tool has been developed and desseminated for use by VA clinicians completing the Comprehensive TBI Evaluation.
- Vanderploeg RD, Groer S, Belanger HG. Initial developmental process of a VA semistructured clinical interview for TBI identification. Journal of rehabilitation research and development. 2012 Nov 15; 49(4):545-56.
- Belanger HG, Vanderploeg RD, Soble JR, Richardson M, Groer S. Validity of the Veterans Health Administration's traumatic brain injury screen. Archives of physical medicine and rehabilitation. 2012 Jul 1; 93(7):1234-9.
- Vanderploeg RD, Groer S, Belanger HG, Soble JR, Richardson M. Validity of a Population-Based Screen for TBI: Comparing the VA’s TBI Screen to Follow-Up Evaluation. Presented at: VA HSR&D / QUERI National Meeting; 2012 Jul 16; National Harbor, MD.