Abstract
Common methods for evaluating history of traumatic brain injury (TBI) include self-report, electronic medical record review (EMR), and structured interviews such as the Head Trauma Events Characteristics (HTEC). Each has strengths and weaknesses, but little is known regarding how TBI diagnostic rates or the associated symptom profile differ among them. This study examined 200 Veterans recruited within the VA Portland Health Care System, each evaluated for TBI using self-report, EMR, and HTEC. Participants also completed validated questionnaires assessing chronic symptom severity in broad health-related domains (pain, sleep, quality of life, post-concussive symptoms). The HTEC was more sensitive (80% of participants in our cohort) than either self-report or EMR alone (40%). As expected from the high sensitivity, participants screening positive for TBI through the HTEC included many people with mild or no post-concussive symptoms. Participants were grouped according to degree of concordance across these diagnostic methods: no TBI, n = 43; or TBI-positive in any one method (TBI-1dx, n = 53), positive in any two (TBI-2dx, n = 45), or positive in all three (TBI-3dx, n = 59). The symptom profile of the TBI-1dx group was indistinguishable from the no TBI group. The TBI-3dx group had the most severe symptom profile. Our results show that understanding the exact methods used to ascertain TBI is essential when interpreting results from other studies, given that results and conclusions may differ dramatically depending on the method. This issue will become even more critical when interpreting data merged from multiple sources within newer, centralized repositories (e.g., Federal Interagency Traumatic Brain Injury Research [FITBIR]).
Keywords: chronic pain, concussion, post-concussion syndrome, sleep, traumatic brain injury
Introduction
There are an estimated 69,000,000 cases of traumatic brain injury (TBI) globally per year,1 with >2,500,000 in the United States.2 Although post-concussive issues may resolve within weeks or months following TBI,3 patients can also experience persistent symptoms.4,5 Long-term sequela include chronic pain,6,7 sleep–wake disturbances,8–10 mental health decrements (e.g., anxiety, depression),11 and an overall reduction in quality of life.12 Indeed, Veterans with a history of mild TBI demonstrate persistent post-concussive symptomology >20–25 years later.13–16 Chronic post-concussive symptoms have been attributed to a combination of neurogenic factors related to the initial injury and to additional psychosocial factors.17–20
A challenge to understanding the long-term consequences of TBI is the variability in how the presence or absence of TBI in populations is determined, whether by (1) participant self-report, (2) medical record review, or (3) structured clinical interview.21 There is also variability within each evaluative method.21 For example, self-report can be achieved by a single yes/no question or based on a screening approach,22 medical record review can vary by search terms/criteria, and structured clinical interviews employ multiple approaches. Nevertheless, neither the concordance among these three general methods, nor the symptom profile associated with each method, has been previously examined.
Here, we applied each of these approaches to the same individuals to assess TBI history. We focused on a population at high risk for TBI, United States Veterans.1 We hypothesized that TBI diagnostic rates would differ with the method of evaluation. We also hypothesized that participants would differ in symptom severity depending on the method used, and whether the different methods were considered individually or in combination.
Methods
The VA Portland Healthcare System institutional review board approved this study (IRB #3988) and participants were provided written informed consent prior to participation. Participants (n = 214) were enrolled between June 2018 and February 2020 from flyers placed on site at the VA, via clinical referral from the VA Sleep Disorders Clinic (M.M.L.), an outpatient clinic population with high rates of TBI, or via word of mouth. Participants were excluded (n = 14) if they were not Veterans, or if they had eye disease (absence of eye disease is necessary to undergo photosensitivity testing, which is not discussed here). The remaining 200 participants were included in the current analyses.
TBI evaluation and diagnostic groups
Three TBI evaluative methods were performed. (1) Self-report; a single yes/no question asking, “Have you ever had a concussion or traumatic brain injury?” (2) Electronic medical record review (EMR) searching the terms: TBI, head injury, brain injury, head trauma, brain trauma, concussion, concussive, amnesia, loss of consciousness, and blast. Participants were determined to have had a TBI if at least one EMR note with those keywords described a TBI. They were considered negative for TBI if there was a documented instance when TBI assessment was performed and the result was negative, or if there was no evidence of TBI assessment. (3) A structured clinical interview, using the Head Trauma Events Characteristics (HTEC) form, conducted in person or over the phone.23 The HTEC follows a similar format to that of other structured clinical interviews (e.g., Ohio State University – OSU TBI ID24), which begins with a standard screening question followed by branching logic questions addressing injury type, location, intracranial injury/skull fracture, duration of loss of consciousness (LOC), and anterograde or retrograde post-traumatic amnesia (PTA). A diagnosis of no, mild, moderate, or severe TBI was made by a licensed neuropsychologist (M.L.C.).
Participants were grouped according to the outcomes of these three methods. If outcomes were negative across all three measures, participants were determined to be negative for TBI (no TBI). If they were positive for TBI in any one measure, they were determined to have had a history of TBI based on that evaluative method. To consider overlap in positivity across measures, participants were categorized based on the number of positive measures: positive in any one diagnostic method (TBI-1dx), positive in any two methods (TBI-2dx), or positive in all three methods (TBI-3dx).
Medical and military history
Medical and military/service histories were obtained via self-report. Medical history included cancer, cardiovascular health, autonomic management, pain syndromes, trauma history, cognitive complaints, and current medication usage for sleep, pain, or depression. Service history included duration of service, exposure to combat, percent service connection (i.e., disability related to military service), branch of service, and total number of deployments. Specific information pertaining to TBI (recency/age, number, type, location, and estimated LOC/PTA) was collected using the HTEC.
Self-report questionnaires
Post-concussive symptom severity
The Neurobehavioral Symptom Inventory (NSI) (22 items; range = 0–88) was used to assess post-concussive symptoms over the prior 2 weeks.25–27 The Post-Traumatic Stress Disorder (PTSD) Checklist for the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) (PCL-5) (20 items; range = 0–80) was used to assess PTSD symptom severity over the prior 30 days.28 The Visual Light Sensitivity Questionnaire (VSLQ) (7 items; range = 1–33) was used to assess photosensitivity, a common post-concussive symptom.29
Current and chronic pain
The Defense Veterans Pain Rating Scale (DVPRS) (5 items; range = 0–50) evaluated the biopsychosocial impact of pain over the prior 24 h, and the Symptom Impact Questionnaire-Revised (SIQR) (21 items; range = 0–100) addressed the impact of current pain over the previous 7 days on activities of daily living. The Michigan Body Map was used identify the presence of pain over 35 different locations persisting for >3 months (score = number of locations).
Sleep quality
The Insomnia Severity Index (ISI) (7 items; range = 0–28) was used to assess difficulty initiating/maintaining sleep,30,31 and the Functional Outcomes of Sleep Questionnaire-10 (FOSQ-10) (10 items; range = 5–20) to assess impact of sleep disturbances on daily life.32
Quality of life
Outcomes included (1) the World Health Organization Disability Assessment Schedule 2.0 (WHO-DAS 2.0) (12 items; range = 0–100%), (2) the Neuro-QoL Satisfaction and Participation scales (each 8 items; range = 8–40), and (3) the National Institutes of Health Patient-Reported Outcomes Measurement Information System (NIH PROMIS) Cognitive Function short-form 4a (4 items; range = 4–20). The WHO-DAS 2.0 assesses general health and disability in major life domains over the past 30 days. The Neuro-QoL Satisfaction and Participation questionnaires target their respective areas of quality of life. The NIH PROMIS Cognitive Function survey assesses self-perceived thinking and memory problems.
Symptom burden
Scores for each questionnaire were normalized to a 0–100% scale (based on the total possible score) and if necessary, the directionality corrected such that higher percentages indicate worse outcomes, described here as “symptom burden.” The symptom burden was averaged for each domain: (1) post-concussive symptom severity, (2) current and chronic pain, (3) sleep quality, and (4) quality of life, and these four values were averaged for an overall weighted symptom burden.
Statistical analysis
Analyses were performed using GraphPad Prism version 8.4.3 and R version 3.3.2. Significance was defined a priori as p < 0.05. Data are presented as mean ± standard deviation, or as a percent of the total sample size, unless otherwise noted. Numerical data were assessed using a one-way analysis of variance (ANOVA) with Tukey's post-hoc test or with unpaired t test. Categorical data were assessed using a χ2 test with Bonferroni post-hoc test (Tables 1–5 and Fig. 2). Numerical data in Figure 3 were analyzed via an unpaired Student's t test (e.g., self-report positive vs. self-report negative), and across evaluative methods (e.g., self-report positive, EMR positive, HTEC positive) via Cohen's d with 95% confidence intervals. Differences in effect sizes across evaluative methods were based on non-overlapping confidence intervals.
Table 1.
Demographic Information across all Participants and TBI Diagnostic Groups
| Whole group n = 200 | No TBI n = 43 (21%) | TBI-1dx n = 53 (26%) | TBI-2dx n = 45 (22%) | TBI-3dx n = 59 (29%) | |
|---|---|---|---|---|---|
| Gender | |||||
| Male | 164 (82.0%) | 35 (81.4%) | 43 (81.1%) | 39 (86.7%) | 47 (79.7%) |
| Female | 35 (17.5%) | 8 (18.6%) | 10 (18.9%) | 6 (13.3%) | 11 (18.6%) |
| Non-binary | 1 (0.5%) | 0 | 0 | 0 | 1 (1.7%) |
| Age, years | 54.9 ± 14.5 | 57.7 ± 17.8 | 55.3 ± 13.1 | 55.2 ± 12.5 | 52.4 ± 13.7 |
| Race | |||||
| American Indian/Alaska Native | 6 (3.0%) | 1 (2.3%) | 2 (3.8%) | 1 (2.2%) | 2 (3.4%) |
| Asian | 5 (2.5%) | 1 (2.3%) | 2 (3.8%) | 0 | 2 (3.4%) |
| Black or African American | 9 (4.5%) | 0 | 3 (5.7%) | 3 (6.7%) | 3 (5.1%) |
| White | 159 (79.5%) | 36 (83.7%) | 39 (73.6%) | 37 (82.2%) | 47 (79.7%) |
| Mixed | 16 (8.0%) | 4 (9.3%) | 5 (9.4%) | 3 (6.7%) | 4 (6.8%) |
| Other/Not Reported | 5 (2.5%) | 1 (2.3%) | 2 (3.7%) | 1 (2.2%) | 1 (1.7%) |
| Education, ≥some college | 174 (87.0%) | 36 (83.7%) | 46 (86.8%) | 41 (91.1%) | 51 (86.4%) |
| Exercise, ≥90 min/week | 63 (31.5%) | 9 (20.1%) | 19 (35.8%) | 16 (35.5%) | 19 (32.2%) |
Data are presented as n (% total) or mean ± standard deviation. Continuous variables were analyzed via one-way analysis of variance (ANOVA), and categorical variables were analyzed using χ2. No significant differences were found between diagnostic groups.
TBI, traumatic brain injury; TBI-1dx, TBI-2dx, and TBI-3dx, participants respectively receiving 1, 2, or 3 positive diagnoses using the three diagnostic approaches.
Table 2.
HTEC-Derived TBI Characteristics across all Participants and TBI Diagnostic Groups
| Whole group n = 153 | TBI-1dx n = 50 (33%) | TBI-2dx n = 44 (29%) | TBI-3dx n = 59 (38%) | |
|---|---|---|---|---|
| Severity | ||||
| Mild | 121 (79.1%) | 46 (92%) | 36 (81.8%) | 39 (66.1%)† |
| Moderate | 23 (15.0%) | 3 (6.0%) | 8 (18.2%) | 12 (20.3%)† |
| Severe | 9 (5.9%) | 1 (2%) | 0 | 8 (13.6%) |
| Recency | ||||
| <1 year | 4 (2.6%) | 1 (2.0%) | 2 (4.5%) | 2 (3.4%) |
| 1-5 years | 19 (12.4%) | 4 (8.0%) | 3 (6.8%) | 9 (15.2%) |
| 6-10 years | 9 (5.9%) | 1 (2.0) | 3 (6.8%) | 5 (8.5%) |
| 11-30 years | 51 (32.7%) | 19 (38.0%) | 14 (31.8%) | 18 (30.5%) |
| >30 years | 70 (45.7%) | 25 (50.0%) | 22 (50.0%) | 23 (40.0%) |
| Average, years | 27.8 ± 17.9 | 31.8 ± 17.4 | 27.7 ± 16.3 | 24.4 ± 19.0 |
| Age of injury, years | 26.1 ± 15.8 | 23.2 ± 15.1 | 26.9 ± 16.1 | 28.0 ± 16.0 |
| Number of injuries | ||||
| 1 | 28 (18.3%) | 15 (30.0%) | 9 (20.4%) | 4 (6.8%)† |
| 2-4 | 61 (39.9%) | 18 (36.0%) | 18 (40.9%) | 25 (42.4%) |
| 5-10 | 42 (27.4%) | 11 (22.0%) | 12 (27.3%) | 19 (32.2%) |
| 11-25 | 13 (8.5%) | 5 (10.0%) | 3 (6.8%) | 5 (8.5%) |
| >25 | 7 (4.6%) | 1 (2.0%) | 2 (4.5%) | 4 (6.8%) |
| Maximum | 160 | 26 | 46 | 160 |
| Average, number | 7.1 ± 14.8 | 5.1 ± 6.3 | 5.9 ± 8.3 | 9.7 ± 21.9 |
| Type | ||||
| MVC | 30 (19.6%) | 10 (20.0%) | 6 (13.6%) | 14 (23.7%) |
| Sports | 25 (16.3%) | 13 (26.0%) | 8 (18.2%) | 4 (6.8%)† |
| Pedestrian-MVC | 10 (6.5%) | 3 (6.0%) | 1 (2.3%) | 6 (10.2%) |
| Blast | 9 (5.9%) | 2 (4.0%) | 2 (4.6%) | 5 (8.5%) |
| Fall | 34 (22.2%) | 7 (14.0%) | 14 (31.8%) | 13 (22.0%) |
| Assault | 23 (15.0%) | 5 (10.0%) | 8 (18.2%) | 10 (16.9%) |
| Other | 22 (14.4%) | 10 (20.0%) | 5 (11.4%) | 7 (11.9%) |
| Loss of consciousness | ||||
| None | 69 (45.0%) | 30 (60.0%) | 23 (52.3%) | 16 (27.1%)† ‡ |
| <1 min | 17 (11.1% | 6 (12.0%) | 4 (9.1%) | 7 (11.9%) |
| 1-30 min | 41 (26.8%) | 11 (22.0%) | 11 (25.0%) | 19 (32.2%) |
| 30 min - 24 h | 18 (11.8%) | 2 (4.0%) | 6 (13.6%) | 10 (16.9%) |
| 24 h - 7 days | 6 (3.9%) | 1 (2.0%) | 0 | 5 (8.5%) |
| >7 days | 2 (1.3%) | 0 | 0 | 2 (3.3%) |
| Post-traumatic amnesia | ||||
| None | 131 (85.6%) | 45 (90.0%) | 37 (84.0%) | 49 (83.0%) |
| <1 min | 1 (0.6%) | 1 (2.0%) | 0 | 0 |
| 1-30 min | 8 (5.2%) | 3 (6.0%) | 2 (4.6%) | 3 (5.1%) |
| 30 min - 24 h | 7 (4.6%) | 1 (2.0%) | 2 (4.6%) | 4 (6.8%) |
| 24 h - 7 days | 5 (3.3%) | 0 | 3 (6.8%) | 2 (3.3%) |
| >7 days | 1 (0.6%) | 0 | 0 | 1 (1.7%) |
Data are derived from the Head Trauma Events Characteristics (HTEC) interview and are presented as n (% total) or mean ± standard deviation. In participants with multiple traumatic brain injuries (TBIs), these data reflect only the most severe TBI. This includes153/157 total participants with HTEC-derived data and excludes 4 who screened negative for TBI on the HTEC. Continuous variables were analyzed via one-way analysis of variance (ANOVA) with Tukey post-hoc comparison, and categorical variables via χ2 with Bonferroni post-hoc comparison.
p < 0.05 vs. TBI-1dx, ‡p < 0.05 vs TBI-2dx.
MVC, motor vehicle collision.
Table 3.
Medical History across all Participants and TBI Diagnostic Groups
| Whole group n = 200 | No TBI n = 43 (21%) | TBI-1dx n = 53 (26%) | TBI-2dx n = 45 (22%) | TBI-3dx n = 59 (29%) | |
|---|---|---|---|---|---|
| Cancer | 41 (20.5%) | 13 (30.2%) | 9 (17.0%) | 9 (20.0%) | 10 (17.0%) |
| Diabetes | 42 (21.0%) | 4 (9.3%) | 13 (24.5%) | 12 (26.7%) | 13 (22.0%) |
| Dizziness | 73 (35.5%) | 13 (30.2%) | 15 (28.3%) | 16 (35.6%) | 29 (49.2%) |
| Fibromyalgia | 25 (12.5%) | 2 (4.7%) | 7 (13.2%) | 6 (13.3%) | 10 (16.9%) |
| Gulf War Syndrome | 10 (5.0%) | 0 | 3 (5.6%) | 2 (4.4%) | 5 (8.5%) |
| Heart disease/heart attack | 33 (16.5%) | 5 (11.6%) | 9 (17.0%) | 11 (24.4%) | 8 (13.6%) |
| High blood pressure | 94 (47.0%) | 22 (51.7%) | 31 (58.5%) | 17 (37.8%) | 24 (40.7%) |
| Migraine/cluster headache | 66 (33.0%) | 10 (23.3%) | 13 (24.5%) | 10 (22.2%) | 33 (55.9%)* † ‡ |
| PTSD | 88 (44.0%) | 5 (11.6%) | 18 (34.0%) | 20 (44.4%) | 45 (76.3%)* † ‡ |
| Fatigue | 81 (40.5%) | 11 (25.6%) | 21 (39.6%) | 18 (40.0%) | 31 (52.5%)* |
| Memory problems | 59 (29.5%) | 7 (16.3%) | 15 (28.3%) | 13 (28.9%) | 24 (40.7%)* † |
| Medication usage | |||||
| Sleep medication | 89 (44.5%) | 18 (41.9%) | 22 (41.5%) | 20 (44.4%) | 29 (49.2%) |
| Pain medication | 114 (57.0%) | 20 (46.5%) | 30 (56.6%) | 26 (57.8%) | 38 (64.4%) |
| Depression medication | 93 (46.5%) | 18 (41.9%) | 19 (35.8%) | 18 (40.0%) | 38 (64.4%)† |
Data are presented as n (% of total). Fatigue and memory problems reflect only moderate to severe issues. Sleep medication includes use of any of the following: sedative-hypnotics, melatonin, benzodiazepines, γ-hydroxybutyric acid, doxylamine, trazadone, quetiapine, diphenhydramine, mirtazapine, and over-the-counter herbs. Pain medication includes use of any of the following: oxycontin, hydrocodone, morphine, oxycodone, hydrocodone bitartrate, fentanyl patch, methadone, codeine, naltrexone/suboxone, and lidocaine patch. Depression medication includes use of any of the following: citalopram/Celexa, fluoxetine/Prozac, paroxetine/Paxil, sertraline/Zoloft, duloxetine/Cymbalta, venlafaxine/Effexor, amitriptyline or other tricyclic antidepressants (TCA), bupropion/Wellbutrin, buspirone/Buspar, trazodone/Desyrel, quetiapine/Seroquel, aripiprazole/Abilify, St. John's Wort, escitalopram/Lexapro. Data obtained via self-report without medical record corroboration. Analyzed via χ2 with Bonferroni post-hoc comparison.
p < 0.05 vs. no TBI, †p < 0.05 vs. TBI-1dx, ‡p < 0.05 vs. TBI-2dx.
TBI, traumatic brain injury; PTSD, post-traumatic stress disorder.
Table 4.
Military Service History across all Participants and TBI Diagnostic Groups
| Whole group n = 200 | No TBI n = 43 (21%) | TBI-1dx n = 53 (26%) | TBI-2dx n = 45 (22%) | TBI-3dx n = 59 (29%) | |
|---|---|---|---|---|---|
| Service duration, years | 6.9 ± 6.9 | 6.8 ± 6.8 | 7.3 ± 7.7 | 5.9 ± 5.1 | 7.5 ± 7.6 |
| Exposure to combat | 57 (28.5%) | 8 (18.6%) | 12 (22.6%) | 13 (28.9%) | 25 (42.4%) |
| Service connection, % | 55.3 ± 30.8 | 44.4 ± 29.4 | 50.7 ± 32.6 | 60.0 ± 26.5 | 60.6 ± 31.7 |
| Branch | |||||
| Army | 82 (41.0%) | 17 (39.5%) | 19 (35.8%) | 19 (42.2%) | 27 (45.8%) |
| Navy | 33 (16.5%) | 10 (23.3%) | 10 (18.9%) | 6 (13.3%) | 7 (11.9%) |
| Air Force | 31 (15.5%) | 9 (20.9%) | 8 (15.1%) | 7 (15.6%) | 7 (11.9%) |
| Marines | 27 (13.5%) | 3 (7.0%) | 7 (13.2%) | 9 (20.0%) | 8 (13.6%) |
| Coast Guard | 3 (1.5%) | 1 (2.3%) | 1 (1.9%) | 0 | 1 (1.7%) |
| Reserve | 11 (5.5%) | 1 (2.3%) | 4 (7.5%) | 2 (4.4%) | 4 (6.8%) |
| National Guard | 3 (1.5%) | 0 | 1 (1.9%) | 1 (2.2%) | 1 (1.7%) |
| Multiple Branches | 7 (3.5%) | 1 (2.3%) | 3 (5.7%) | 1 (2.2%) | 2 (3.4%) |
| Unknown | 3 (1.5%) | 1 (2.3%) | 0 | 0 | 2 (3.4%) |
| Number of deployments | |||||
| Not deployed | 68 (34.0%) | 16 (37.2%) | 21 (39.6%) | 16 (35.6%) | 13 (23.7%) |
| 1 | 39 (19.5%) | 11 (25.6%) | 18 (33.9%) | 16 (35.6%) | 24 (10.2%) |
| 2-3 | 75 (37.5%) | 13 (30.2%) | 13 (24.5%) | 13 (28.8%) | 11 (18.6%) |
| 4-5 | 18 (9.0%) | 3 (7.0%) | 1 (2.3%) | 0 | 11 (18.6%)† ‡ |
Data are presented as n (% total) or mean ± standard deviation. Data obtained via self-report without medical record corroboration. Continuous variables were analyzed via one-way analysis of variance (ANOVA). Categorical variables were analyzed via χ2 with Bonferroni post-hoc comparison.
p < 0.05 vs. TBI-1dx, ‡p < 0.05 vs. TBI-2dx.
TBI, traumatic brain injury.
Table 5.
Raw Survey Scores across all Participants and TBI Diagnostic Groups
| Whole group n = 200 | No TBI n = 43 (21%) | TBI-1dx n = 53 (26%) | TBI-2dx n = 45 (22%) | TBI-3dx n = 59 (29%) | |
|---|---|---|---|---|---|
| Symptom severity | |||||
| NSI | 27.1 ± 15.9 | 16.6 ± 12.4 | 24.2 ± 17.8 | 28.4 ± 17.4* | 39.2 ± 16.2*†‡ |
| PCL-5 | 26.9 ± 17.9 | 13.3 ± 14.7 | 26.2 ± 19.7* | 27.8 ± 18.4* | 40.1 ± 19.0*†‡ |
| VSLQ | 11.6 ± 5.8 | 8.1 ± 5.4 | 8.7 ± 5.2 | 12.1 ± 5.9*† | 17.7 ± 6.5*†‡ |
| Current and chronic pain | |||||
| DVPRS | 14.4 ± 10.8 | 9.3 ± 9.0 | 11.1 ± 11.0 | 17.5 ± 12.1*† | 19.6 ± 11.0*† |
| SIQR | 31.2 ± 19.5 | 22.9 ± 18.4 | 26.3 ± 20.3 | 34.2 ± 18.9* | 41.6 ± 20.4*† |
| Michigan Body Map | 6.7 ± 5.5 | 3.5 ± 4.1 | 6.2 ± 5.4 | 7.3 ± 5.9* | 9.6 ± 6.5*† |
| Sleep quality | |||||
| ISI | 13.2 ± 6.2 | 10.9 ± 6.6 | 12.6 ± 6.6 | 13.3 ± 6.0 | 15.8 ± 5.8*† |
| FOSQ-10 | 15.4 ± 3.1 | 17.6 ± 2.4 | 15.7 ± 3.7* | 14.3 ± 3.4* | 13.9 ± 3.0*† |
| Quality of life | |||||
| WHO-DAS 2.0 | 28.0 ± 17.8 | 20.3 ± 18.8 | 24.8 ± 15.9 | 32.4 ± 18.8* | 34.5 ± 17.8*† |
| Participation | 29.5 ± 7.0 | 32.6 ± 7.5 | 30.5 ± 6.8 | 28.3 ± 7.4* | 26.4 ± 6.5*† |
| Satisfaction | 26.0 ± 8.0 | 30.8 ± 7.6 | 26.1 ± 8.2 | 24.4 ± 8.6* | 22.8 ± 7.8*† |
| Cognitive Function | 14.1 ± 4.6 | 16.6 ± 3.2 | 15.2 ± 5.2 | 13.0 ± 5.3* | 11.7 ± 4.7*† |
Data are presented as mean ± standard deviation. Data were analyzed via one-way analysis of variance ANOVA with Tukey post-hoc test.
p < 0.05 vs. no TBI, †p < 0.05 vs. TBI-1dx, ‡p < 0.05 vs. TBI-2dx
TBI, traumatic brain injury; NSI, Neurobehavioral Symptom Inventory; PCL-5, Post-Traumatic Stress Disorder Checklist DSM-V; Pain Interference, National Institutes of Health Patient-Reported Outcomes Measurement Information System (NIH PROMIS) short-form 4a; Pain Intensity, NIH PROMIS short-form 3a; DVPRS, Defense Veterans Pain Rating Scale; SIQR, Symptom Impact Questionnaire-Revised; ISI, Insomnia Severity Index; FOSQ-10, Functional Outcomes Of Sleep Questionnaire-10; WHO-DAS 2.0, World Health Organization Disability Assessment Schedule; Participation, Neuro-QoL; Satisfaction, Neuro-QoL; Cognitive Function, NIH PROMIS short-form 4a.
FIG. 2.
Self-report questionnaires, grouped into domains related to post-concussive symptom severity (Neurobehavioral Symptom Inventory [NSI], Post-Traumatic Stress Disorder Checklist for the Diagnostic and Statistical Manual for Mental Disorders, Fifth Edition [DSM-V] [PCL-5], and Visual Light Sensitivity Questionnaire [VSLQ]); current and chronic pain (Defense Veteran Pain Rating Scale [DVPRS], Symptom Impact Questionnaire – Revised [SIQR], and Michigan Body Map); sleep quality (Insomnia Severity Index [ISI] and Functional Outcomes of Sleep Questionnaire-10 [FOSQ-10]); and quality of life (World Health Organization Disability Assessment Schedule 2.0 [WHODAS 2.0], Neuro-QoL Participation and Satisfaction, and the Patient-Reported Outcomes Measurement Information System [PROMIS] Cognitive Function short-form 4a) plotted against traumatic brain injury (TBI) diagnostic groups (no TBI, TBI-1dx, TBI-2dx, and TBI-3dx). Self-report questionnaires were normalized to their total scores (and scales reversed, if needed) to be on a common scale of 0–100% describing “symptom burden.” Colored dashed lines represent each diagnostic groups overall (weighted) symptom burden (described further in the Methods section). Data are presented as box and whisker plots with the box bound at the bottom by the 25th percentile, at the top by the 75th percentile, and the median indicated by the interior line. Data were analyzed via one-way analysis of variance (ANOVA) with Tukey honestly significant post-hoc comparisons. *p < 0.05 vs. no TBI, †p < 0.05 vs. TBI-1dx, ‡p < 0.05 vs. TBI-2dx.
FIG. 3.
Each domain (post-concussive symptom severity, pain, sleep, and quality of life) represent the overall average for their respective self-report questionnaires (see Fig. 2). Self-report questionnaires were normalized to their total scores (and scales reversed, if needed) to be on a common scale of “0–100% impairment.” Open and shaded bars represent participants who were negative or positive for traumatic brain injury (TBI) based on the single TBI evaluative method displayed (self-report, electronic medical record [EMR], and Head Trauma Events Characteristics [HTEC]). Differences in each domain within TBI evaluative methods were analyzed via an unpaired two-tailed Student's t test, *p < 0.05 vs. negative. Domains within TBI evaluative methods were analyzed via Cohen's d effect sizes with 95% confidence intervals (the confidence intervals in all comparisons overlap and therefore do not statistically differ).
Results
TBI evaluation and diagnostic groups
The sample (n = 200) was predominantly male (82%), middle-aged (54.9 ± 14.5 years of age), white (79%), with some degree of college education (87.0%), and exercised on average <90 minutes/week (68.5%). Most, n = 157 (78.5%), were positive for TBI based on at least one method (Table 1 and Fig. 1). Only 53 participants were considered TBI-positive based on only a single method (TBI-1dx). Over half of the participants were thus considered TBI-positive based on more than one method (TBI-2dx, TBI-3dx). The TBI-1dx group was driven almost entirely by HTEC diagnoses (94.3%). The TBI-2dx group was evenly split between self-report+HTEC (57.8%), and EMR+HTEC (40.0%). There were no differences in demographics among diagnostic groups (Table 1).
FIG. 1.
Venn diagram illustrating the proportions of participants determined to be traumatic brain injury (TBI)-positive based on self-report, electronic medical record (EMR), and Head Trauma Events Characteristics (HTEC). Overlap of TBI evaluative methods illustrates participants who were TBI-positive with two of three methods, or all three methods.
TBI characteristics and medical history
Mild, moderate, and severe TBI was diagnosed in 121, 23, and 9 participants, respectively (Table 2). The TBI-1dx group had predominantly mild injuries, whereas the TBI-2dx and TBI-3dx groups showed increasing proportions of more severe TBI. LOC and PTA also had longer durations in the TBI-2dx and TBI-3dx groups. Average TBI recency was ∼28 years, with no group differences. There were no notable differences across groups in type of TBI, other than that there were fewer sports-related injuries in the TBI-3dx group. There were also no notable differences in location of injury, with the majority of participants (∼50% of each group) reporting a frontal or occipital lobe injury. Finally, those in the TBI-2dx and TBI-3dx groups reported a greater number of prior brain injuries.
There were increased rates of headache/migraine, PTSD, and use of depression-related medications in the TBI-3dx group (Table 3).
Military service history
On average, Veterans reported ∼7 years of service, ∼30% combat exposure, and ∼55% service connection (Table 4). The TBI-3dx group reported the longest service history (∼7.5 years) and greatest rate of combat exposure (∼41%), although group differences did not reach significance. Interestingly, the TBI-3dx group showed higher rates of deployment than the TBI-1dx and TBI-2dx groups (∼20% were deployed four or five times).
TBI-related symptomology
Symptom profiles were assessed through questionnaires targeting post-concussive symptom severity, current and chronic pain, sleep quality, and quality of life. Raw scores for each questionnaire are presented in Table 5. Symptom burden profiles for each questionnaire are illustrated in Figure 2.
Post-concussive symptom severity
NSI, PCL-5, and VSLQ scores were all highest in the TBI-3dx group, compared with the TBI-2dx group, the TBI-1dx group, and the no TBI group. The TBI-2dx group also scored higher on all three survey measures than the no TBI group, but only their VSLQ scores were significantly higher than the TBI-1dx group.
Current and chronic pain
All pain-related measures were higher in the TBI-3dx group; however, this was only significant compared with the TBI-1dx and no TBI groups. The TBI-2dx group scored only significantly higher than the TBI-1dx group on the DVPRS, but scored higher than the no TBI group on all pain-related measures.
Sleep quality
ISI and FOSQ scores, assessing insomnia severity, and the extent that sleep impairments disrupt daily activities were worst in the TBI-3dx group compared with the TBI-1dx and No TBI groups.
Quality of life
All measures of quality of life were worst in the TBI-3dx group. Although none of these measures differed between the TBI-3dx and TBI-2dx groups, all except the NeuroQoL Satisfaction scale differed between the TBI-3dx and TBI-1dx groups, and all differed between the TBI-3dx and no TBI group. The TBI-2dx groups significantly differed compared with the no TBI group only, again on all measures.
Overall symptom burden
In addition to differences in each domain, there was a progressive increase in overall symptom burden when comparing diagnostic groups (Fig. 2). The overall weighted symptom burden across all outcome domains increased from 19% to 29%, 35%, and 46% in the no TBI, TBI-1dx, TBI-2dx, and TBI-3dx groups, respectively. These overall weighted symptom burden values are signified by colored dashed lines within each panel.
Differences among individual diagnostic methods
Overall domain scores were compared between participants screening positive or negative based on each individual diagnostic method (Fig. 3). Regardless of diagnostic method, TBI-positive participants scored significantly worse on each domain score (post-concussion symptoms, pain, sleep quality, and quality of life). The lack of difference may in part be because of the large variation in symptom burden (expressed as a %) within the groups of TBI-positive participants for each diagnosis method. HTEC was far more sensitive for TBI than the other methods, capturing almost twice the number of people than EMR or self-report. Not surprisingly, this sensitivity comes with reduced specificity for symptomatic TBI. There is great variability in overall symptom burden across all the HTEC+ participants; with each additional diagnosis; however, the symptom burden increased (Fig. 4).
FIG. 4.
Overall weighted symptom burden (%) for all participants who were HTEC+ (A), and separately for “no TBI” participants and those who were Head Trauma Events Characteristics (HTEC)+ with different combinations of self-report (SR) and electronic medical record (EMR) diagnostic concordance (B). Data were analyzed via one-way analysis of variance (ANOVA) with Tukey honestly significant post-hoc comparisons. *p < 0.05 vs. no TBI, †p < 0.05 vs. HTEC+ (SR- and EMR-).
Discussion
This study investigated the rate and characteristics of TBI in Veterans, utilizing three different evaluative methods (self-report, EMR review, and HTEC interview) for each participant. Each of these three methods has strengths and weaknesses. Self-report is quick, and carries a low burden for both the participant and investigators, but relies heavily on an individual's awareness and understanding of TBI. EMR review can be automated and standardized, but clinical documentation relies on providers' awareness, recognition, and coding of TBI. Structured clinical interviews are considered the most accurate for TBI diagnosis, but they are time and labor intensive for both the participant and investigator.
We found that the rate of TBI diagnosis differed by evaluative method, with the HTEC interview being most sensitive, with self-report and EMR review being less so. When participants were grouped according to concordance among the methods (TBI-1dx, TBI-2dx, and TBI-3dx), there were significant differences in symptom severity such that those with TBI-3dx reported the greatest symptom burden. However, individual methods did not significantly differ from each other with regard to symptom profile. Our findings highlight differences between assessment methods in “sensitivity” (i.e., identifying TBI-positive vs. TBI-negative participants) and “specificity” (i.e., identifying individuals with chronic sequelae of TBI), depending on the method of evaluation and the extent of concordance across methods. Any one method to assess TBI may therefore lead to different conclusions about symptoms attributable to TBI. Unfortunately, current practices in TBI research typically rely on only one method for assessment. Combining all three methods improves the specificity for the symptomatic TBI population. Therefore, our results show that understanding the exact methods used to ascertain TBI is important when interpreting results from other studies, given that results and conclusions may differ dramatically depending on the method. This issue becomes even more critical when interpreting data merged from multiple sources within newer, centralized repositories (e.g., Federal Interagency Traumatic Brain Injury Research [FITBIR]).
In the present study, the HTEC structured clinical interview had the highest sensitivity for identifying TBI, and identified 80% of all subjects as positive for TBI. However, the rate of TBI diagnosis when using only self-report or medical record review was ∼40%. The wide distribution of symptom burden (Fig. 4) across all of the HTEC+ participants emerges from the overlap of the HTEC+ only (mildly symptomatic), HTEC+ and either EMR or self-report (moderately symptomatic), and the 3-dx group (most symptomatic). Indeed, the symptom burden of the HTEC+ (1dx) group was indistinguishable from the that of the noTBI group. These findings highlight both the poor sensitivity of the self-report or medical record review to detect “any TBI,” and the poor specificity of the HTEC alone in identifying “symptomatic TBI.” Using the HTEC alone may result in asymptomatic individuals “diluting out” the symptom severity of individuals with clinically relevant TBI, arguably the group that scientists should be studying and clinicians should be treating. We suggest that it is not the “any TBI” group that warrants further study, but rather the “symptomatic TBI” group that is of greatest scientific and clinical need. Only when combining all three methods (e.g., TBI-3dx) is the “symptomatic TBI” group most accurately identified. Development of a composite measure that includes self-report, medical status, and neuropsychological assessment elements would be an important next step for the field.
Chronic TBI symptoms were clearly stratified when considering concordance of TBI diagnosis across the three methods. Participants in the TBI-3dx group demonstrated greater symptom burden across all domains than those diagnosed with TBI in only one or two methods. Participants in the TBI-3dx group reported more headaches, had higher PTSD scores, suffered from greater memory impairment, and used more depression-related medications. Therefore, chronic TBI symptoms reflect additional factors beyond those identified in a structured clinical interview, such as salience of the injury (as reflected in self-report), and care seeking (as reflected in the EMR). Nevertheless, when each assessment method was considered individually as if no additional information were available, significant impairments were identified in each of the four health-related domains for each of the three methods (Fig. 3). This latter finding argues that the likelihood of long-term sequelae following TBI is sufficiently great that population effects will be identified irrespective of the method used to assess TBI history.
It is possible that injury severity contributed to the association between TBI-3dx and symptom burden, as eight of the nine participants with severe TBI fell into the TBI-3dx group. However, the number of participants with severe TBI was small relative to the whole sub-group, and injury severity did not differ between the TBI-3dx and TBI-2dx groups. Sensitivity analyses demonstrated no difference across outcomes with the removal of these nine severe TBI participants. Accordingly, although injury severity may play a role, perhaps by increasing the chances of being captured across all diagnostic methods, it does not likely explain the entire relationship between TBI-3dx and symptom burden.
Limitations
The current study utilized the HTEC as the structured interview, as it is recommended by the Department of Defense and the Department of Veteran Affairs.23 Direct comparisons with other commonly used structured interviews such as the Boston Assessment of Traumatic Brain Injury-Lifetime and the Ohio State University Traumatic Brain Injury Identification would be needed to identify potential differences.24 Additionally, HTEC interviews were not conducted immediately after TBI, and the majority of our population were recalling injuries sustained 25–30 years ago. Similarly, the present study did not employ multiple self-report measures, and alternate search terms could have been used. The approach taken in the present study was designed to be as generalizable as possible to previous work.
Participants were determined to be negative for TBI on EMR review if they either had a documented history of screening negative for TBI, or no TBI documentation at all. Therefore, it is possible that participants with no documentation may have been miscategorized as no TBI. However, these participants were likely captured as TBI-positive with the HTEC (and potentially via self-report), and thus would be ultimately categorized into either the TBI-1dx or TBI-2dx group. Indeed, only three participants were determined to be TBI-positive by way of EMR review alone, and conversely, 76 participants who were TBI-positive based on self-report or HTEC were TBI-negative on EMR review. The EMR review was limited to the VA Computerized Patient Record System (CPRS) and did not include active military service members, which could also contribute to the lower rates of TBI in the EMR group.
Medication and military service are two demographic domains that were captured using only self-report without additional medical record corroboration. These variables may therefore be prone to the standard inaccuracies associated with patient self-report processes. The concordance of self-report versus medical record corroboration is not presently well described in the literature, and future work addressing this issue would be of interest.
Finally, our Veteran population is primed for exposure to TBI. Most participants were recruited either from the VA Sleep Clinic or from posted flyers at the VA. Veterans referred to the sleep clinic have higher rates of TBI than Veterans without sleep disturbances, as sleep disturbances are a common feature of post-concussive syndrome. Further, Veterans who seek care at the VA also have higher rates of TBI than Veterans who get their health care from the community, as a result of service connection and degree of disability. In addition, although our cohort from a single site is representative of the Pacific Northwest region of the United States, the racial/ethnic/gender representation is likely not generalizable to United States Veterans elsewhere.
Conclusions
The three commonly used TBI evaluation methods of self-report, EMR review, and clinical interview each have their own strengths and weaknesses, but when used alone, may be insufficient to accurately capture the full clinical picture of TBI and associated chronic symptoms. The structured interview demonstrated high sensitivity for TBI but captures people across the spectrum from low to high chronic TBI symptomatology. Combining all three methods yielded a subset of TBI-positive participants with the most impaired symptom severity, and arguably represents the population of greatest scientific and clinical relevance. Our data suggest that differences in diagnostic methods may explain the variability of reported rates of TBI and the controversial association of chronic symptoms with TBI in previous literature. Future studies should consider the pros and cons of using each method, or of using all three methods simultaneously, in order to capture the highest sensitivity and specificity for TBI and TBI-related symptoms.
Acknowledgments
The authors express their sincere appreciation and gratitude for the participation of all participants and to the staff at the VA Portland Health Care System Sleep Disorders Clinic.
Dr. Lim takes full responsibility for the data, the analyses and interpretation, and the conduct of the research; has full access to all of the data; and has the right to publish any and all data separate and apart from any sponsor.
This work was conducted at the VA Portland Health Care System.
Authors' Contributions
N.M.B., J.E.E., M.P.B., M.M.H., and M.M.L. were responsible for the conception and design of the study. N.M.B., J.E.E., A.A.M., M.L.C., and K.T.S. were responsible for the acquisition and analysis of data. J.E.E., N.M.B., M.M.H., and M.M.L. were responsible for drafting a significant portion of the manuscript or figures; that is, making a substantial contribution beyond copy editing and approval of the final draft, which was expected of all authors.
Funding Information
This work was supported by the Office of the Assistant Secretary of Defense for Health Affairs under Award #W81XWH-17-1-0423 to M.M.H.; M.M.L. was supported by VA Career Development Award #1K2 BX002712 and the Portland VA Research Foundation; N.M.B. was supported by NIH grant TL1TR002371; and J.E.E. was supported by NIH grant T32 AT002688 and VA Career Development Award #1K2 RX002947.
Author Disclosure Statement
No competing financial interests exist. The interpretations and conclusions expressed in this article are those of the authors and do not necessarily reflect the position or policy of the Department of Defense or Department of Veterans Affairs.
References
- 1. Dewan, M.C., Rattani, A., Gupta, S., Baticulon, R.E., Hung, Y.C., Punchak, M., Agrawal, A., Adeleye, A.O., Shrime, M.G., Rubiano, A.M., Rosenfeld, J. V., and Park, K.B. (2019). Estimating the global incidence of traumatic brain injury. J. Neurosurg. 130, 1080–1097. [DOI] [PubMed] [Google Scholar]
- 2. Centers for Disease Control and Prevention (2016). Injury Prevention and Control: Traumatic Brain Injury and Concussion.
- 3. Rutherford, W.H., Merrett, J.D., and Mcdonali, J.R. (1977). Sequelæ of concussion caused by minor head injuries. Lancet 309, 1–4. [DOI] [PubMed] [Google Scholar]
- 4. Haarbauer-Krupa, J., Pugh, M.J., Prager, E.M., Harmon, N., Wolfe, J., and Yaffe, K.C. (2021). Epidemiology of chronic effects of traumatic brain injury. J. Neurotrauma [Epub ahead of print]. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Mac Donald, C.L., Johnson, A.M., Wierzechowski, L., Kassner, E., Stewart, T., Nelson, E.C., Werner, N.J., Adam, O.R., Rivet, D.J., Flaherty, S.F., Oh, J.S., Zonies, D., Fang, R., and Brody, D.L. (2017). Outcome trends after US military concussive traumatic brain injury. J. Neurotrauma 34, 2206–2219. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Ofek, H., and Defrin, R. (2007). The characteristics of chronic central pain after traumatic brain injury. Pain 131, 330–340. [DOI] [PubMed] [Google Scholar]
- 7. Nampiapampil, D.E. (2008). Prevalence of chronic pain after traumatic brain injury. JAMA 300, 711–719. [DOI] [PubMed] [Google Scholar]
- 8. Ouellet, M.-C., Savard, J., and Morin, C.M. (2004). Insomnia following traumatic brain injury: a review. Neurorehabil. Neural Repair 18, 187–198. [DOI] [PubMed] [Google Scholar]
- 9. Castriotta, R.J., Wilde, M.C., Lai, J.M., Atanasov, S., Masel, B.E., and Kuna, S.T. (2007). Prevalence and consequences of sleep disorders in traumatic brain injury. J. Clin. Sleep Med. 3, 349–356. [PMC free article] [PubMed] [Google Scholar]
- 10. Duclos, C., Dumont, M., Wiseman-Hakes, C., Arbour, C., Mongrain, V., Gaudreault, P.-O., Khoury, S., Lavigne, G.J., Desautels, A., and Gosselin, N. (2014). Sleep and wake disturbances following traumatic brain injury. Pathol. Biol. 62, 252–261. [DOI] [PubMed] [Google Scholar]
- 11. Moldover, J.E., Goldberg, K.B., and Prout, M.F. (2004). Depression after traumatic brain injury: A review of evidence for clinical heterogeneity. Neuropsychol. Rev. 14, 143–154. [DOI] [PubMed] [Google Scholar]
- 12. Ponsford, J.L., Draper, K., and Schönberger, M. (2008). Functional outcome 10 years after TBI: Its relationship with injury severity, cognitive and emotional status. J. Int. Neuropsychol. Soc. 14, 233–242. [DOI] [PubMed] [Google Scholar]
- 13. Balba, N.M., Elliott, J.E., Weymann, K.B., Opel, R.A., Duke, J.W., Oken, B.S., Morasco, B.J., Heinricher, M.M., and Lim, M.M. (2018). Increased sleep disturbances and pain in Veterans with co-morbid TBI and PTSD. J. Clin. Sleep Med. 14, 1865–1878. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Elliott, J.E., Opel, R.A., Weymann, K.B., Chau, A.Q., Papesh, M.A., Callahan, M.L., Storzbach, D., and Lim, M.M. (2018). Sleep disturbances in traumatic brain injury: associations with sensory sensitivity. J. Clin. Sleep Med. 14, 1177–1186. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Elliott, J.E., Opel, R.A., Pleshakov, D., Rachakonda, T., Chau, A.Q., Weymann, K.B., and Lim, M.M. (2020). Posttraumatic stress disorder increases the odds of REM sleep behavior disorder and other parasomnias in Veterans with and without comorbid traumatic brain injury. Sleep 43, zsz237. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Boyle, E., Cancelliere, C., Hartvigsen, J., Carroll, L.J., Holm, L.W., and Cassidy, J.D. (2014). Systematic review of prognosis after mild traumatic brain injury in the military: results of the international collaboration on mild traumatic brain injury prognosis. Arch. Phys. Med. Rehabil. 95, S230–S237. [DOI] [PubMed] [Google Scholar]
- 17. Iverson, G.L. (2005). Outcome from mild traumatic brain injury. Curr. Opin. Psychiatry 18, 301–317. [DOI] [PubMed] [Google Scholar]
- 18. Porter, K.E., Stein, M.B., Martis, B., Avallone, K.M., McSweeney, L.B., Smith, E.R., Simon, N.M., Gargan, S., Liberzon, I., Hoge, C.W., and Rauch, S.A.M. (2018). Postconcussive symptoms (PCS) following combat-related traumatic brain injury (TBI) in Veterans with posttraumatic stress disorder (PTSD): influence of TBI, PTSD, and depression on symptoms measured by the Neurobehavioral Symptom Inventory (NSI). J. Psychiatr. Res. 102, 8–13. [DOI] [PubMed] [Google Scholar]
- 19. Wood, R.L. (2004). Understanding the “miserable minority”: a diasthesis-stress paradigm for post-concussional syndrome. Brain Inj. 18, 1135–1153. [DOI] [PubMed] [Google Scholar]
- 20. Ruff, R. (2005). Two decades of advances in understanding of mild traumatic brain injury. J. Head Trauma Rehabil. 20, 5–18. [DOI] [PubMed] [Google Scholar]
- 21. Carlson, K.F., Kehle, S.M., Meis, L.A., Greer, N., MacDonald, R., Rutks, I., Sayer, N.A., Dobscha, S.K., and Wilt, T.J. (2011). Prevalence, assessment, and treatment of mild traumatic brain injury and posttraumatic stress disorder: a systematic review of the evidence. J. Head Trauma Rehabil. 26, 103–115. [DOI] [PubMed] [Google Scholar]
- 22. Schwab, K.A., Ivins, B., Cramer, G., Johnson, W., Sluss-Tiller, M., Kiley, K., Lux, W., and Warden, D. (2007). Screening for traumatic brain injury in troops returning from deployment in Afghanistan and Iraq: initial investigation of the usefulness of a short screening tool for traumatic brain injury. J. Head Trauma Rehabil. 22, 377–389. [DOI] [PubMed] [Google Scholar]
- 23. Cifu, D., Hurley, R., Peterson, M., Cornis-Pop, M., Rikli, P.A., Ruff, R.L., Scott, S.G., Sigford, B.J., Silva, K.A., Tortorice, K., Vanderploeg, R.D., Withlock, W., Bowles, A., Cooper, D., Drake, A., Engel, C., Geckle, L.S., Helmick, K., Hoge, C.W., Jaffee, M., Labutta, R., Ling, G., Lowe, L., Mims, S., Orman, D.T., Solomon, B., Stone, J.M., Thomas, K., Tolbert, M., Williams, C.S., Francis, J., Cassidy, C., Klar, A., Degenhardt, E., Susskind, O., Cohen, J., Kasten, J.J., Radcliff, S., E, S.W., D'Erasmo, M., Fishman, R., and Marko, J. (2009). VA/DoD clinical practice guideline: management of concussion/mild traumatic brain injury. J. Rehabil. Res. Dev. 46, CP1-68. [PubMed] [Google Scholar]
- 24. Corrigan, J.D., and Bogner, J. (2007). Initial reliability and validity of the Ohio State University TBI identification method. J. Head Trauma Rehabil. 22, 318–329. [DOI] [PubMed] [Google Scholar]
- 25. Belanger, H.G., Kretzmer, T., Vanderploeg, R.D., and French, L.M. (2009). Symptom complaints following combat-related traumatic brain injury: relationship to traumatic brain injury severity and posttraumatic stress disorder. J. Int. Neuropsychol. Soc. 16, 194–199. [DOI] [PubMed] [Google Scholar]
- 26. Bulson, R., Jun, W., and Hayes, J. (2012). Visual symptomatology and referral patterns for Operation Iraqi Freedom and Operation Enduring Freedom veterans with traumatic brain injury. J. Rehabil. Res. Dev. 49, 1075–1082. [DOI] [PubMed] [Google Scholar]
- 27. Vanderploeg, R.D., Cooper, D.B., Belanger, H.G., Donnell, A.J., Kennedy, J.E., Hopewell, C.A., and Scott, S.G. (2014). Screening for postdeployment conditions: development and cross-validation of an embedded validity scale in the Neurobehavioral Symptom Inventory. J. Head Trauma Rehabil. 29, 1–10. [DOI] [PubMed] [Google Scholar]
- 28. Blevins, C.A., Weather, F.W., Davis, M.T., Witte, T.K., and Domino, J.L. (2015). The Posttraumatic Stress Disorder Checklist for DSM-5 (PCL-5): development and initial psychometric evaluation. J. Trauma. Stress 28, 489–498. [DOI] [PubMed] [Google Scholar]
- 29. Papesh, M.A., Elliott, J.E., Callahan, M.L., Storzbach, D., Lim, M.M., and Gallun, F.J. (2019). Blast exposure impairs sensory gating: evidence from measures of acoustic startle and auditory event-related potentials. J. Neurotrauma 36, 702–712. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Morin, C.M. (1993). Insomnia. Psychological Assessment and Management. Guilford Press: New York. [Google Scholar]
- 31. Bastien, C.H., Vallières, A., and Morin, C.M. (2001). Validation of the insomnia severity index as an outcome measure for insomnia research. Sleep Med. 2, 297–307. [DOI] [PubMed] [Google Scholar]
- 32. Mastin, D.F., Bryson, J., and Corwyn, R. (2006). Assessment of sleep hygiene using the Sleep Hygiene Index. J. Behav. Med. 29, 223–227. [DOI] [PubMed] [Google Scholar]