Abstract
We compared mental health outpatient, primary care, and emergency care service utilization among veterans screening TBI positive (S-TBI+) versus those screening TBI negative (S-TBI−) and describe associations between TBI-related symptoms and health service utilization. Our study population consisted of 1746 Iraq and Afghanistan veterans in VA care screened for TBI between April 1, 2007 and June 1, 2010. Rates of mental health outpatient, primary care, and emergency services utilization were greater for S-TBI+ veterans, compared with S-TBI− veterans, even after adjusting for mental health screen results. Irritability on the initial TBI screen was associated with increased mental health outpatient utilization rates [incidence rate ratio (IRR), 1.64; 95% confidence interval (CI), 1.18–2.3; p<0.01]. Reports of dizziness (IRR, 1.24; 95% CI, 1.02–1.51; p<0.05) and headaches (IRR, 1.41; 95% CI, 1.16–1.7; p<0.001) were associated with increased primary care utilization rates. Higher utilization rates among veterans who screened positive for TBI were not better explained by screening positive for comorbid mental health problems. Knowing that certain symptoms are more strongly associated with increased utilization in certain health service domains will help to better plan for the care of returning veterans who screen positive for TBI.
Introduction
M
Although symptoms of mTBI are expected to resolve within 3–12 months postinjury, 4 –6 there is evidence that up to one third of patients may continue to experience persistent cognitive, emotional, behavioral, and physical symptoms. 6,7 Research among athletes suggests that military personnel sustaining multiple mTBIs from repeated blast exposures or falls may experience an even longer recovery. 8 Given that TBI-related problems may include physical, cognitive, and emotional complaints, and recovery time varies among OEF/OIF veterans with mTBI, there is a need to plan for adequate medical treatment.
Few research studies have examined health care service utilization among patients with TBI. Among civilian patients with mTBI to severe TBI (sTBI), approximately 40–81% of patients utilized medical services, followed by allied health services, such as physiotherapy, occupational therapy, and psychological services. 9,10 Further, health care services for civilian patients with TBI continued well beyond the early stages of recovery. 10 For example, within 6 months to 4 years postinjury, health services utilization was more related to rehabilitation and restoration of function, whereas services utilized 6–17 years postinjury were related to ongoing management. 10 Other studies found that approximately half of the civilian patients with moderate to sTBI had annual rehospitalization for surgery and/or general medical problems within the first 3 years after injury as well as rehospitalization for behavioral or psychiatric problems 5 years postinjury. 11,12
To date, only one study has examined VA outpatient health service utilization among veterans with TBI. Veterans with mTBI had more visits to VA primary care than veterans with moderate or sTBI. 13 All visits increased over time for veterans with mTBI, whereas the pattern for veterans with moderate and sTBI varied. 13 Although these results demonstrated that veterans with TBI engage in long-term use of health care services and that patterns of utilization may change over time, what remains to be examined are the specific postconcussive symptoms that may be driving particular patterns of health care use.
Recognizing the importance of early evaluation and treatment of OEF/OIF veterans with mTBI to prevent chronic disability, the Veterans Health Administration (VHA) and Department of Veterans Affairs (VA) mandated TBI screening of all OEF/OIF Veterans who were not previously diagnosed with TBI upon initiation of care at a VA medical facility. 14,15 The aim of this retrospective study was to delineate health services utilization patterns among OEF/OIF veterans who underwent VA TBI screening. Using administrative data from one VA medical facility and its affiliated community-based clinics, we (1) compared mental health outpatient, primary care, and emergency care service utilization among veterans screening TBI positive (S-TBI+) versus those screening TBI negative (S-TBI−) and (2) described associations between particular TBI-related symptoms and health service utilization.
Methods
Data source
We used VA administrative data for OEF/OIF veterans who were screened for TBI at one VA medical center and five affiliated VA community-based outpatient clinics. Responses to the VA TBI screen were obtained from the VA Corporate Data Warehouse (CDW). The Iraq and Afghanistan Post-Deployment Screen for post-traumatic stress disorder (PTSD), depression, and alcohol misuse, as well as service utilization data, were extracted from the Veterans Health Information Systems and Technology Architecture (VistA). Demographic information and military service information were derived from the VA OEF/OIF Roster database. 16 The study was approved by the Committee on Human Research, University of California, San Francisco (San Francisco, CA) and the Human Research Protection Program at the San Francisco VA Medical Center.
Sample selection
Our study population consisted of 1746 OEF/OIF veterans who were screened for TBI between April 1, 2007 and June 1, 2010. Follow-up appointment data were extracted from the date of the TBI screen through February 17, 2011. For the purposes of this study, we used results from PTSD, depression, and alcohol screens administered within 1 year of the TBI screen. The sample was 10% female, had a mean age of 30.2 years [standard deviation (SD), 8.3] and was 43.1% white, 3.6% black, 13.5% Hispanic, and 39.8% other/unknown race.
Measures
Service utilization
VA clinic codes were used to create three categories of outpatient services: mental health outpatient; primary care; and emergency care. The number of visits made in each service category was counted over the duration of the study period. When multiple appointments were noted at the same clinic on the same day, only one appointment was counted. Initiation of care was a dichotomous measure defined as having utilized a particular service at least once.
TBI screen
The VA TBI screen is a five-section tool modified from the Brief TBI Screen to evaluate possible TBI in OEF/OIF veterans. 17 The first section of the TBI screen identifies those previously diagnosed with TBI. Veterans with previous TBI diagnoses are not administered the remainder of the TBI screen. Veterans without a previous TBI diagnosis are subsequently asked four sequential questions. 15 A positive TBI screen consists of endorsing one or more responses in each of the four sections: (1) a qualifying TBI event (e.g., blast or explosion, vehicular accident or crash, fragment wound or bullet wound above the shoulders, or fall); (2) immediate symptoms subsequent to the event (i.e., losing consciousness/“knocked out,” being dazed, confused, or “seeing stars,” not remembering the event, concussion, or head injury); (3) new or worsening TBI-related symptoms subsequent to the event (memory problems or lapses, balance problems or dizziness, light sensitivity, irritability, headaches, or sleep problems); and (4) current symptoms (same symptoms as in item 3 above). A negative screen for TBI results when a veteran fails to endorse at least one item in any of the four sections. Once a negative response is provided, the screen terminates. The VA TBI screen was found to have high internal consistency (0.77), high test-retest reliability (0.80), high sensitivity (0.94), and moderate specificity (0.59). 18
PTSD screen
The Primary Care Posttraumatic Stress Disorder Screen 19 is a four-item screen designed to detect a PTSD diagnosis in primary care and other medical settings. Veterans are asked to provide a binary response (“yes/no”) for each of four PTSD symptom clusters: re-experiencing; avoidance; emotional numbing; and arousal. Endorsing three or more symptoms constitutes a positive screen for PTSD.
Depression screen
The Patient Health Questionnaire-2 (PHQ-2) 20 is a two-item depression screen derived from the nine-item depression module of the Patient Health Questionnaire. 21 This shorter screen assesses symptoms of depressed mood and anhedonia on a four-point Likert scale ranging from 0 (“not at all”) to 3 (“nearly every day”). Total scores range from 0 to 6. Veterans obtaining a total score of 3 or higher screen positive for depression.
Alcohol screen
The Alcohol Use Disorders Identification Test Consumption (AUDIT-C) 22 is a brief screen for alcohol misuse, including questions about frequency of alcohol use within the past year, quantity of alcohol use on a typical day, and number of heavy drinking episodes within the past year on a five-point Likert scale. Total scores range from 0 to 12. Alcohol misuse was defined on the AUDIT-C as a total score ≥4 for men and ≥3 for women. 23
Statistical analysis
Using chi-square tests and t-tests, we compared demographic and military service characteristics, as well as VA-specific factors (i.e., closest VA facility type and time in VA system since initial TBI screen) in S-TBI+ and S-TBI− OEF/OIF veterans. We compared the initiation of care (utilized services at least once) and numbers of mental health, primary care, and emergency room visits between S-TBI+ and S-TBI− veterans. The proportion that initiated care during the course of the study period was compared using Pearson's chi-square test, the mean number of visits (per person per year) was compared using a t-test, and the median number of visits (per person per year) was compared using Mann-Whitney's test. Next, negative binomial regression of the total number of visits was used to compare rates of utilization between S-TBI+ and S-TBI− veterans. To adjust for the varying length of observation time per veteran, the log of the time from the initial TBI screen to the end of the study was specified as an offset in negative binomial regression. A separate analysis using negative binomial regression examined the association between service utilization rates and postconcussive symptoms among the subset of S-TBI+ veterans. The negative binomial models were adjusted for demographic and military characteristics, VA-specific factors, and results on the PTSD, depression, and alcohol misuse screens. All analyses were conducted with SAS software (version 9.2; SAS Institute, Cary, NC).
Results
Demographic and military service comparisons
The sample included 1746 OEF/OIF veterans who had an initial TBI screen between April 1, 2007 and June 1, 2010. Of those screened, 27% (n=465) were S-TBI+. Veterans who screened positive for TBI were more likely to be younger than age 25, male, of active duty service, of enlisted rank, in the army, and to have served multiple deployments (Table 1).
Numbers vary because of missing data.
p<0.05.
p<0.0001.
OEF, Operation Enduring Freedom; OIF, Operation Iraqi Freedom; TBI, traumatic brain injury; S-TBI+, screening positive for TBI; S-TBI−, screening negative for TBI.
Health service utilization comparisons
Overall, S-TBI+ veterans were more likely to have had at least one VA outpatient visit, compared with S-TBI− veterans (mental health outpatient, 84 vs. 62%; primary care, 87 vs. 83%; and emergency care clinics, 32 vs. 20%; all p<0.05; Table 2). The number of visits per person per year was greater among S-TBI+ veterans, compared with S-TBI− veterans, for all three service categories (mental health outpatient, median of 1.9 vs. 0.5 visits; primary care, median of 2 vs. 1.5 visits; and emergency care, median of 0.3 vs. 0.2 visits; all p<0.0001). Both unadjusted and adjusted rates of mental health outpatient, primary care, and emergency care service utilization were greater for S-TBI+ veterans, compared with S-TBI− veterans. After adjusting for demographic and military characteristics, VA-specific factors, and mental health and alcohol misuse screen results, S-TBI+ veterans were 1.42 times more likely to have utilized services in mental health [95% confidence interval (CI), 1.17–1.72; p=0.0004], 1.31 times more likely to have utilized services in primary care (95% CI, 1.17–1.46; p<0.0001), and 1.46 times more likely to have utilized emergency care services (95% CI, 1.09–1.94; p=0.0099), compared with S-TBI− veterans.
Rate ratios for the negative binomial regression model adjusted for gender, age, race, marital status, component type, rank, service branch, multiple deployment, time between last deployment and TBI screen, closest VA facility type, results on the post-traumatic stress disorder, depression, and alcohol misuse screens, and time in VA system since initial TBI screen. Because of missing values for some covariates, the n in multivariate analysis was reduced to 1492.
p<0.05.
p<0.001.
p<0.0001.
OEF, Operation Enduring Freedom; OIF, Operation Iraqi Freedom; TBI, traumatic brain injury; S-TBI+, screening positive for TBI; S-TBI−, screening negative for TBI; Y/N, yes/no; IRR, incidence rate ratio; CI, confidence interval; SD, standard deviation.
TBI symptom associations with health service utilization
Among S-TBI+ veterans, a report of irritability on the initial TBI screen was independently associated with increased rates of mental health outpatient utilization after adjusting for potential confounders [incidence rate ratio (IRR), 1.64; 95% CI, 1.18–2.30; p=0.0031; see Table 3). Reports of dizziness (IRR, 1.24; 95% CI, 1.02–1.51; p=0.03) and headaches (IRR, 1.41; 95% CI, 1.16–1.7; p=0.0005) were independently associated with increased primary care utilization. TBI-related symptoms were not independently associated with increased emergency services utilization.
Incidence rate ratios for the negative binomial regression model adjusted for gender, age, race, marital status, component type, rank, service branch, multiple deployment, closest VA facility type, time between last deployment and TBI screen, results on the post-traumatic stress disorder, depression, and alcohol misuse screens, and time in VA system since initial TBI screen. Because of missing values for some covariates, the n in multivariate analysis was reduced to 373.
p<0.05.
p<0.001.
p<0.0001.
S-TBI+, screening positive for TBI; TBI, traumatic brain injury; IRR, incidence rate ratio; CI, confidence interval.
Discussion
Given that traumatic brain injury has been called the signature injury of the current conflicts in Iraq and Afghanistan 2 and evidence that this new generation of veterans delays seeking care, 24 it is important to understand patterns of health services use among S-TBI+ veterans in VA health care to best prepare for their future care. We found that after controlling for a number of variables, including positive mental health screens, S-TBI+ veterans utilized health care services to a greater extent than their S-TBI− counterparts. Consequently, preparation for higher rates of utilization among S-TBI+ veterans on a national level is important, as is a better understanding about particular clinical presentations and corresponding higher utilization rates of certain VA health services.
We found that particular rates of utilization were driven by certain TBI-related symptoms. Though irritability was associated with increased mental health utilization, dizziness and headaches were associated with increased primary care use. This is particularly important, given that dizziness and headaches are the most frequently reported postinjury symptoms, with irritability persisting over time. 25 Further, we found, in a previous investigation, that dizziness and headaches were among the distinguishing symptoms that did not overlap with other mental health problems, and that irritability was a shared symptom with PTSD among OEF/OIF veterans screened for postdeployment problems. 26 Consequently, one possibility is that by better managing irritability within a integrated care setting (e.g., through emotion regulation and anger management groups in primary care), mental health service utilization will decrease for those experiencing TBI-related irritability as a primary complaint. For example, a recent study demonstrated preliminary evidence for the utility of a brief, anger self-management treatment for individuals with TBI. 27
An important area for future investigation is to better understand whether symptoms of TBI, such as irritability, dizziness, and headaches, can be ameliorated by implementing evidence-based treatments (EBTs) for TBI 28 within an integrated primary care setting. If these existing evidence-based treatments can be used to reduce TBI-related symptoms within an integrated primary care setting, theoretically there would be less of a need for more expensive specialty services in the management of these postconcussive symptoms. Positive recovery expectations are an important cornerstone of diverse postconcussive symptoms, and these also can be accomplished through primary care. 29
Given that, within the VA system, most veterans who screen positive on the initial TBI screen will receive a second-level comprehensive TBI evaluation, with the goal of gathering more detailed TBI symptom assessment, information gathered during this evaluation can be used to determine whether a veteran would be appropriate for TBI symptom management within an integrated care setting. For example, some veterans who endorse TBI-related symptoms and either few mental health symptoms or refuse to engage in mental health care can be managed within an integrated care setting; however, if more intensive mental health symptoms emerge (e.g., chronic PTSD symptoms or violent tendencies, rather than more basic irritability), referral to mental health will be a priority.
There are several limitations of this study that should be noted. First, in this study, we used TBI screen results, which are different than confirmed TBI diagnoses. Related to this, we used mental health screen results, which are not confirmed mental health diagnoses. Second, although all screens were done within a year of each other, not all were necessarily done at the same time, and in some cases, veterans may have been screened several years after returning from their deployment(s). Third, it should be noted that these results are from one VA and its associated community clinics and, consequently, should not be generalized to all veterans. Future studies should replicate these findings among national samples. Fourth, it is important to note that a positive TBI screen does not trigger referral to primary care, mental health, or emergency services, which were outcomes in this study, but rather further evaluation in specialty services (e.g., neurology), which was not an outcome in this study. Despite these limitations, VA screen results represent the level of information that clinicians use to make triage and clinical decisions about care for their patients.
Higher utilization rates in primary care, mental health outpatient, and emergency care services among veterans who screen positive for TBI were not better explained by positive mental health screens. Though irritability was associated with mental health utilization, dizziness and headaches were associated with primary care use. Knowing that certain symptoms are more strongly associated with increased utilization in certain service domains will help to better plan for the care of returning veterans who screen positive for TBI.
Footnotes
Acknowledgments
This research was supported by the Department of Defense Psychological Health and Traumatic Brain Injury (PH/TBI) Research Program (to S.M.) and a VA Health Sciences Research and Development (HSR&D) Career Development Award (to S.M.). The authors thank Gary Abrams, Tatjana Agopian-Novokovic, Daniel Bertenthal, Jeane Bosch, Julie Dinh, Don Donati, Thomas Neylan, and Richard Pham for their contributions.
Author Disclosure Statement
No competing financial interests exist.