Response to Charles W. Hoge, MD, "Letters to the Editor Pertaining to the Article 'Traumatic Brain Injury Screening: Preliminary Findings in a US Army Brigade Combat Team'"a
In response to the letter submitted by Charles W. Hoge, several important points should be highlighted. First and foremost, we reemphasize that the primary objective of the submission1 was to report the proportion of soldiers in a Brigade Combat Team with a history of traumatic brain injury (TBI) and describe the nature of symptoms reported in this cohort. It was not our intention to imply causality; therefore, we used terms such as "associated," "risk," and "association" to describe observed relationships.1 Moreover, the study was not intended to offer information regarding the accuracy, validity, or effectiveness of postdeployment screening for mild TBI (mTBI). In fact, we suggested that "further research to both determine the validation of this and other screening programs and address the impact of substance abuse, psychiatric history, and predeployment history of TBI [horizontal ellipsis] is indicated."1(p22)
Dr Hoge also raises concerns regarding the methodology used to assess a history of TBI. As discussed by Corrigan and Bogner,2 there are inherent challenges in assessing for mTBI in civilian populations, particularly if medical care was never sought. These challenges are certainly compounded in a combat setting. At present, structured or in-depth interview is considered the diagnostic "gold standard."2-4 As described by Terrio et al,1 the Warrior Administered Retrospective Causality Assessment Tool (WARCAT) was used to begin the process of data collection. Soldiers were asked to respond to symptom questions if any history of injury was present. The WARCAT was then used by trained clinicians to facilitate an in-depth interview. Dr Hoge's comments suggest that clinician-confirmed diagnoses are somehow inferior to responses generated solely by surveys and violate scientific principles. He states that the data gathered on symptoms and clinician-confirmed TBI are not independent and should not have been entered into regression analyses. It is not clear why clinician-confirmed variables are less "independent" from one another than variables derived solely from self-report surveys. Clearly there are unresolved, and thorny, methodological issues in assessing symptoms that may be related to a remote mTBI experienced by service members during their Operation Enduring Freedom/Operation Iraqi Freedom deployment. Unfortunately, there is no easy or obvious solution. Nevertheless, we believe that the combined approach described by Terrio et al1 and currently used at Fort Carson has advantages over using a screening tool, survey, or clinical interview alone. In particular, omitting the interview process could lead to seemingly unimportant, yet clinically relevant data being underreported or underappreciated by either soldiers or clinicians. Recent work by Schwab et al5 seems to support this assertion. The research team used follow-up interviews to confirm TBI history (initial reports were obtained using a short screening tool) and found that information elicited during the interviews assisted in determining whether the participant was likely to have had a TBI.
As was stated in the article (see Table 1, Abstract, Discussion),1 the vast majority of injuries identified were mild in nature. In addition, language used both during the evaluation process and in the article1 was consistent with recommendations provided in the recently released Veterans Affairs/Department of Defense Clinical Practice Guidelines, regarding management of concussion and mTBI.6 That is, "the terms concussion and mTBI can be used interchangeably."6(p8)
The inclusion of the word "preliminary" in the title of the article was intentional and indicative of the evolving state of knowledge, regarding the impact of trauma (physical and/or psychological) on our returning soldiers.1 Continued dialogue and study is indicated. In Terrio et al,1 we suggested that endorsed symptoms recalled may have been wholly or in part related to other deployment-related medical or psychiatric conditions. Data recently collected by Brenner et al (unpublished data, 2009) support this assertion. Findings suggest that in soldiers, with histories of physical injury, mTBI and posttraumatic stress disorder were independently associated with postconcussive symptom reporting. Moreover, those with both conditions were at greater risk for postconcussive symptoms than those with posttraumatic stress disorder alone, mTBI alone, or neither. This conclusion appears consistent with a combined biological and psychological model of postcombat symptoms.7 These findings, along with those presented in Hoge et al8 and Schneiderman et al,9 support the importance of screening for both conditions with the aim of promoting an expectation for recovery.
Heidi Terrio, MD, MPH
Lisa A. Brenner, PhD
Brian J. Ivins, MS
John M. Cho, MD
Katherine Helmick, MS, CRNP
Karen Schwab, PhD
Katherine Scally, MS, ANP-C
Rick Bretthauer, PA-C
Lonnie A. Nelson, PhD
Deborah Warden, MD
Fort Carson
REFERENCES