We conducted descriptive analyses on characteristics of the patients, their initial TBI/polytrauma, and the prevalence and source(s) of subsequent injury. To examine the potential for nonresponse bias, we compared survey respondents with nonrespondents by patient-related variables available in VA administrative data (gender, age, race/ethnicity, and FIM® ratings).
We used bivariate and multivariate logistic regression to examine hypothesized risk and protective factors for subsequent injury in this population. All potential risk and protective factors were analyzed as categorical variables. As indicators of polytrauma complexity, we additionally analyzed the number of initial concomitant injuries and ongoing symptoms/health problems as 2 continuous variables.
Multivariate regression models were specified on the basis of causal modeling and directed acyclic graphing methods developed by Greenland et al17 and demonstrated by Hernán et al.18 Our conceptual model (Fig. 1) was developed a priori and included a general framework depicting how hypothesized constructs may be associated with reinjury. The causal model (not shown), based on the conceptual model, was more complex and detailed hypothesized associations using unidirectional arrows between each variable of interest. The causal model was then used to develop directed acyclic graphs, a method to identify covariates that should be included in each statistical model. This method of covariate selection identifies more parsimonious analytical models than alternative approaches by tailoring covariate sets for each regression model through omission of covariates that may introduce bias or that are not anticipated to theoretically behave as a covariate for a specific analysis. To estimate risk of injury, we calculated odds ratios with 95% confidence intervals in reference to each hypothesized risk factor while controlling for potential confounders identified a priori from the directed acyclic graphs. This procedure was repeated for each separate independent variable, resulting in different sets of covariates being included in different models. Covariates included in each model are identified in the table footnotes.
There were 1045 patients discharged from VA Polytrauma Rehabilitation Centers between September 2001 and February 2009. We were unable to reach 208 next-of-kin by mail or phone. Of the 837 we could contact, 47 refused participation and 18 nominated an alternative contact as the patient's primary caregiver. A total of 564 caregivers completed a survey (67% of those we contacted; 54% of all unique patients' caregivers). We found no significant differences (P < .05) in patient demographics of the survey respondents versus nonrespondents across the administrative data examined.11
Most frequently, survey participants were parents (62%) or spouses/romantic partners (32%) of the patients. Patients were primarily male (95%) and in the Army (55%) or Marines (22%) at the time of their initial TBI/polytrauma. Patients ranged in age from 20 to 55 years (mean = 30.5; SD = 7.1). As is common in VA administrative data,19 race/ethnicity data were missing for a high proportion (∼40%) of the study population. Of those for whom data were available, the majority were identified as white (85%) and non-Hispanic (92%).
Nearly equal proportions of patients had incurred their initial TBI/polytrauma in the Middle East (Iraq, Afghanistan, or other Middle Eastern country; 48%) or the United States (45%); the remaining 7% were injured in various other overseas locations. The mean length of time since the initial TBI/polytrauma was 52 months (SD = 26). Initial injuries were most frequently associated with blasts or bullets and shrapnel (43%) or motor vehicles (43%). On the basis of the combined measure of TBI severity, 30% of patients were considered to have incurred mild TBI, while 70% incurred moderate/severe TBI.
Patients' discharge FIM® ratings tended toward the maximums (means/medians = 28.8/31.0, 80.5/89.0, and 109.3/119.0, respectively, for cognitive, motor, and total FIM® ratings), although scores ranged from the minimum to maximum possible for each. At the time of the survey, a large proportion (43%) of caregivers reported they currently provided fewer than 5 hours of care per week for the patient. However, nearly one-fourth provided 40 or more hours of care per week.
Reported subsequent injuries
Caregivers who opted to respond to the shortened telephone survey (n = 26) were not asked about patients' subsequent injuries and were thus excluded from further analysis. An additional 6% (n = 34) of mailed survey respondents did not complete the subsequent injury question, leaving a total of 504 for analysis. Of these, 161 (32%) caregivers reported postdischarge, medically treated injuries among patients (Table 1). Falls (n = 79) and motor vehicles (n = 59) were the most frequent sources of subsequent injuries, associated with 49% and 37%, respectively, of the reported injuries. Assaults (16%), burns (12%), and poisonings or firearm-related injuries (10%) were also frequently endorsed categories. Multiple injury sources were reported for 35 patients. For those reporting subsequent injuries, 43% reported that VA care was sought for at least one of those injuries. By source-of-injury category, VA treatment was most commonly received for fall- and burn-related injuries (42% and 37%, respectively; data not shown).
In both bivariate and multivariate modeling, increased odds of caregiver-reported subsequent injury were associated with select patient demographics, injury characteristics, and current health and functioning variables (Table 2). Compared to male patients, the small proportion of female patients had approximately 4 and a half times the odds of sustaining a subsequent injury. Higher age (≥35 vs <25 years) appeared to be associated with odds of subsequent injury but did not reach statistical significance in either bivariate or multivariate models. We found no associations between caregivers' reports of subsequent injury and source of initial injury, our rough indicator of TBI severity, or discharge FIM® ratings.
In univariate and multivariate analyses, patients whose reported current general health was poor or fair were more likely to have incurred injuries following their discharge from inpatient rehabilitation than those with excellent, very good, or good health. Patients requiring assistance performing ADLs or IADLs had significantly greater odds of injury than those who did not require assistance. Compared to patients who were married or living with a romantic partner, those who were divorced, separated, or widowed had elevated odds of subsequent injury in bivariate models; however, this association was not significant in the multivariate models. While the overall number of patients was low (n = 15), those who lived in assisted living or nursing homes had more than 4 times greater odds of injury than those living in a private residence.
Select initial concomitant injury types and ongoing symptoms and health problems, as endorsed by caregivers, were associated with odds of subsequent injury. Of the list of 7 injury types associated with the initial TBI/polytrauma, only spinal cord injuries were associated with caregivers' reports of subsequent injury. When analyzed as a continuous variable, the number of initial concomitant injuries endorsed by caregivers was not significantly associated with patients' odds of subsequent injury. Patients' ongoing symptoms/health problems were associated with odds of subsequent injury in both bivariate and multivariate models. At the bivariate level, patients for whom vision loss, chronic pain (other than headaches), balance or vestibular changes, depression, anxiety, anger problems, and sleep problems were endorsed had greater odds of subsequent injury. After controlling for potential confounders, vision loss, hearing loss, problems with speech, chronic pain (other than headaches), and depression appeared to be important risk factors for injury. When analyzed as a continuous variable, the number of ongoing symptoms/health problems also contributed to injury odds. After controlling for potential confounders, each additional item endorsed was associated with a 14% increase in the patients' odds of subsequent injury (odds ratio = 1.14; confidence interval = 1.06–1.22).
Some characteristics of the patients' caregivers were also associated with patients' odds of subsequent injury. In both bivariate and multivariate models, caregivers who reported their own health as poor or fair were more likely to report subsequent injuries for patients than caregivers who reported their own health as excellent, very good, or good. Similarly, caregivers with higher than average or average depressive symptoms or higher than average or average anxiety symptoms were more likely to report subsequent injuries among patients for whom they provided care than caregivers with lower than average scores on each scale. In bivariate models, caregivers with average or lower than average physical functioning scores had higher odds of reporting subsequent injuries among patients than those with higher than average physical functioning. In multivariate models, only those with lower than average physical functioning scores were more likely to report subsequent injuries among patients.
Our study is the first to examine subsequent nonfatal injury as an outcome among Veteran patients with a history of TBI and polytrauma who have been discharged from specialized inpatient rehabilitation programs. Similar to past mortality studies,3 our results suggest a need for enhanced postdischarge injury prevention programs for this population.
Caregivers reported that nearly one-third of discharged patients had incurred subsequent, medically treated injuries, with falls being the most frequent source of injury. These findings were similar to those reported by Krause7 involving patients with SCI. In Krause's longitudinal study, more than one-fifth of patients had sought medical treatment for subsequent injuries in the year prior to the survey, the majority of which were associated with falls. These findings were not unexpected, given that fall risk factors, such as balance disorders, can be common among survivors of TBI and other trauma20–22 and were reportedly prevalent in our study population. Motor vehicles were also associated with a large proportion of subsequent injuries. Recent research has shown that, compared with normative population data, individuals with a history of TBI have higher traffic crash rates,23 and that this increased risk may be associated with slower responses to traffic hazards.24 While the details of motor vehicle-related injuries were not collected in our study, our findings suggest room for improvement in current driving assessment and rehabilitation among patients with TBI/polytrauma.
Some of our findings suggest that patients with poorer outcomes (eg, those requiring assistance with ADLs or IADLs; those living in institutional settings; those in poor/fair health; those with certain ongoing symptoms/health problems) have the greatest odds of subsequent injury. In fact, we found that the odds of subsequent injury increased as the number of reported ongoing symptoms/health problems increased. This finding is consistent with population-based research showing odds of residential injury to be directly related to the number of disabilities reported by study participants.25 We also expected to find that individuals with more severe injuries would have the greatest risk of subsequent injury. This appeared to be true in the case of patients who sustained SCI as part of their TBI/polytrauma, a finding that is consistent with past research.5–8 However, we found no association between injury odds and patients' FIM® ratings or TBI severity. The former null finding may be due to the fact that most patients' FIM® ratings were relatively close to the maximum values; the latter may be due to our inability to differentiate and compare moderate and severe cases of TBI to mild cases of TBI. We hope that future research will continue to examine these potential risk factors. Such injury characteristics, and ongoing symptoms/health problems, may be useful for identifying patients at greatest risk of postdischarge injury and tailoring interventions to address their needs. For example, active long-term follow-up, assessment, and treatment (eg, of depression or chronic pain symptoms) of patients after discharge from inpatient rehabilitation programs may result in decreased incidence or severity of subsequent injuries.
Our multivariate regression analyses identified vision loss, hearing loss, speech problems, chronic pain, and depression as potential risk factors for subsequent injury. Other ongoing symptoms/health problems, such as anxiety, balance problems, anger problems, sleep problems, mobility problems, and seizures, also appeared to be important but did not reach statistical significance after controlling for potential confounders. It is possible that these problems also increase patients' risk of subsequent injuries but were not identified as risk factors in the current study because of our relatively low number of participants. Overall, our findings are consistent with a variety of past studies linking sensory impairments, pain, depression, and disability to higher rates of nonfatal injury.25–29 It is therefore likely that injury-related impairments at least partially drive vulnerability to subsequent injuries in this population.
While ongoing impairments (eg, vision or hearing problems) may be important in the identification of those at greatest risk, it is also critical to direct attention toward potentially modifiable factors. Hospital- and clinic-based injury prevention counseling can be effective for some high-risk populations such as emergency department patients30 and children and adolescents.31 Similar to caregivers of pediatric patients, caregivers of adult patients with TBI/polytrauma tend to be integrally involved in the well-being of the patient11 and may be highly receptive to evidence-based injury prevention counseling and educational materials. Further research may be needed to identify best practices for the timing and delivery of information about injury risk to patients and caregivers.32 Environmental modifications are often considered the most effective injury prevention and control strategies33 and should be the focus of such communication about risk and prevention. It is possible that enhanced home safety assessments prior to and following discharge from acute rehabilitation could also reduce the incidence of subsequent injuries among community-dwelling patients with TBI/polytrauma. Further empirical research examining specific elements of patients' environments that put them at risk may be needed.
A unique finding in our study was that the self-reported health of the patients' caregivers (eg, physical health, depression symptoms, and anxiety symptoms) was associated with patients' odds of subsequent injury. There have been studies examining the effects of TBI on the well-being of patients' informal caregivers.34,35 However, with the exception of one recently published study,36 there has been little research examining effects of caregiver health on patient outcomes. Because our study was cross-sectional, it is not possible to determine the direction of causality between caregiver health and patients' subsequent injuries. However, our results suggest that caregivers in better health may impart some protective effect on patient health. This, in turn, suggests that family-centered care programs, such as the one recently implemented by the VA Polytrauma System of Care,37 and caregiver support programs initiated as part of the new US legislation to support Veterans' family caregivers (Public Law 111-163; the Caregivers and Veterans Omnibus Health Services Act of 2010) could indirectly improve patient outcomes through enhanced support and services for patients' caregivers.
Another unique finding in this study was the greatly increased odds of subsequent injury among female patients. This was a surprising result given that, across epidemiologic studies of injury, males are almost always found to have increased risk of injury compared with females.38–40 The small number of female patients in our study makes this finding highly preliminary; however, this result may indicate unique gender-related injury risk factors among patients with TBI/polytrauma. Future research is needed to replicate this finding and explore potential explanations.
In the current study, we did not collect information on patients' high-risk health behaviors such as alcohol or drug use. Krause7 identified both binge drinking and psychotropic prescription drug use as risk factors for subsequent injury among patients with SCI. Other studies have also identified alcohol and drug abuse as risk factors for subsequent trauma among hospitalized trauma patients.41 Future research should examine how these and other behavioral factors contribute to increased risk of injury among patients with TBI/polytrauma. Clinical trials have shown that alcohol counseling at the time of trauma hospitalization is highly effective in reducing subsequent trauma-related hospitalizations42; it is possible that similarly designed health behavior counseling could reduce injury risk among discharged patients with TBI/polytrauma.
The current study also did not ask caregivers to report the intentionality of patients' injuries. Some of the identified injuries, particularly those associated with firearms or poisonings, may have been acts of self-harm. Our study focused on nonfatal rather than fatal injuries; however, the mechanisms of nonfatal and fatal injuries are often similar. In previous mortality studies, patients with TBI have been shown to have increased risks of suicide compared with the general population.43–46 While this trend has not been consistent across studies,3,47 suicidal ideation is understood to be a common psychological sequelae of TBI.48 Thus, attention toward intentionality of subsequent injuries, whether fatal or nonfatal, in this population is warranted.
This study has several important limitations, including the lack of data on environmental or behavioral factors and injury intentionality as mentioned earlier. The cross-sectional nature of our study inhibits our ability to infer causality. For example, patients' reported poor or fair current health may have been a result of, rather than a risk factor for, subsequent injuries. Future longitudinal studies would help delineate injury risk factors from outcomes. The rate of participation in our study was 67%, which is consistent with response rates in other studies involving caregivers.49,50 While our nonrespondent analysis showed no significant differences between patients of responding versus nonresponding caregivers, suggesting less bias in our estimates, it is possible that response bias may have affected our results. We were also limited by our reliance on proxy (caregiver) responses for information pertinent to the patients' experiences, and the relatively long window of recall on which they were asked to report. Data on patients' initial injuries, such as the length of time patients had lost consciousness, and the incidence and details of subsequent injuries, may have been affected by these factors. In the context of patients who have suffered TBI, caregivers' proxy reports of injury details may be more reliable than patient self-reports if patients suffer ongoing cognitive limitations. In our data, we observed increased likelihood of caregiver reports of subsequent injury as the patients' time since discharge (ie, time at risk) increased, which we took as evidence of accuracy of these proxy responses. However, validation of injury reports against medical records would have further strengthened this evidence. We did not conduct a validity substudy with VA data given that a substantial proportion of injuries were treated in non-VA settings. In future studies, proxy and self-report responses should be validated against medical records when possible.
Our study is the first to our knowledge to highlight postdischarge nonfatal injury as a problem among patients who have been treated for TBI/polytrauma in inpatient rehabilitation settings. Our results build the case for future longitudinal research efforts to examine how patient health problems (ie, ongoing symptoms/health problems) as well as caregiver health problems (ie, depression, anxiety) may interact with environmental factors to place patients with TBI/polytrauma at risk for subsequent injury. Future longitudinal work should validate reports of medically treated injury, collect further details about injury events, and further assess potentially risky environmental and behavioral exposures among patients with TBI/polytrauma.
1. Probst C, Zelle BA, Sittaro NA, Lohse R, Krettek C, Pape HC. Late death after multiple severe trauma: when does it occur and what are the causes? J Trauma. 2009;66(4):1212–1217.
2. Sims DW, Bivins BA, Obeid FN, Horst HM, Sorensen VJ, Fath JJ. Urban trauma: a chronic recurrent disease. J Trauma. 1989;29(7):940–947.
3. Harrison-Felix C, Whiteneck G, DeVivo MJ, Hammond FM, Jha A. Causes of death following 1 year postinjury among individuals with traumatic brain injury. J Head Trauma Rehabil. 2006;21(1):22–33.
4. Waxweiler RJ, Thurman D, Sniezek J, Sosin D, O'Neil J. Monitoring the impact of traumatic brain injury: a review and update. J Neurotrauma. 1995;12(4):509–516.
5. Krause JS, Coker JL, Charlifue S, Whiteneck GG. Health outcomes among American Indians with spinal cord injury. Arch Phys Med Rehabil. 2000;81:924–934.
6. Krause JS. Factors associated with risk for subsequent injuries after the onset of traumatic spinal cord injury. Arch Phys Med Rehabil. 2004;85:1503–1508.
7. Krause JS. Risk for subsequent injuries after spinal cord injury: a 10-year longitudinal analysis. Arch Phys Med Rehabil. 2010;91:1741–1746.
8. Brotherton S, Krause JS, Nietert P. Falls in individuals with incomplete spinal cord injury. Spinal Cord. 2007;45:37–40.
9. Sigford BJ. “To care for him who shall have borne the battle and for his widow and his orphan” (Abraham Lincoln): the Department of Veterans Affairs polytrauma system of care.” Arch Phys Med Rehabil. 2008;89(1):160–162.
10. Sayer NA, Chiros CE, Sigford B, et al. Characteristics and rehabilitation outcomes among patients with blast and other injuries sustained during the Global War on Terror. Arch Phys Med Rehabil. 2008;89(1):163–170.
11. Griffin JM, Friedemann-Sánchez G, Jensen A, et al. The invisible side of war: families caring for US service members with traumatic brain injuries and polytrauma. [published online ahead of print August 25, 2011] J Head Trauma Rehabil.
12. Wright J. Introduction to the FIM(TM). The Center for Outcome Measurement in Brain Injury. http://www.tbims.org/combi/FIM
. Published 2000. Accessed January 30, 2011.
14. Katz S, Ford AB, Moskowitz RW, Jackson BA, Jaffe MW. Studies of illness in the aged. The index of ADL: a standardized measure of biological and psychosocial function. JAMA. 1963;185:914–919.
15. Lawton MP, Brody EM. Assessment of older people: Self-maintaining and instrumental activities of daily living. Gerontologist. 1969;9(3):179–186.
16. Centers for Disease Control and Prevention (CDC). Behavioral Risk Factor Surveillance System Survey Questionnaire. Atlanta, Georgia: US Department of Health and Human Services, Centers for Disease Control and Prevention; 2008.
17. Greenland S, Pearl J, Robins JM. Causal diagrams for epidemiologic research. Epidemiology. 1999;10(1):37–48.
18. Hernán MA, Hernández-Díaz S, Werler MM, Mitchell AA. Causal knowledge as a prerequisite for confounding evaluation: an application to birth defects epidemiology. Am J Epidemiol. 2002;155(2):176–184.
19. Sohn M-W, Zhang H, Arnold N, et al. Transition to the new race/ethnicity data collection standards in the Department of Veterans Affairs. Popul Health Metr. 2006;4(7):1–10.
20. Weightman MM, Bolgla R, McCulloch KL, Peterson MD. Physical therapy recommendations for service members with mild traumatic brain injury. J Head Trauma Rehabil. 2010;25(3):206–218.
21. Williams GP, Schache AG. Evaluation of a conceptual framework for retraining high-level mobility following traumatic brain injury: two case reports. J Head Trauma Rehabil. 2010;25(3):164–172.
22. Hoffer ME, Balough BJ, Gottshall KR. Posttraumatic balance disorders. Int Tinnitus J. 2007;13(1):69–72.
23. Schanke A-K, Rike P-O, Molmen A, Osten PE. Driving behaviour after brain injury: a follow-up of accident rate and driving patterns 6–9 years post-injury. J Rehabil Med. 2008;40:733–736.
24. Preece MHW, Horswill MS, Geffen GM. Assessment of drivers' ability to anticipate traffic hazards after traumatic brain injury. J Neurol Neurosurg Psychiatry. 2011;82:447–451.
25. Vladutiu CJ, Casteel C, Runyan C. Disability and risk of non-fatal residential injuries among adults. Inj Prev. 2008;14:302–305.
26. Schwebel DC, Brezausek CM. Brief report: unintentional injury risk among children with sensory impairments. J Pediatr Psychol. 2010;35(1):45–50.
27. Carruth AK, Skarke L, Moffett B, Prestholdt C. Women in agriculture: risk and injury experiences on family farms. J Am Med Womens Assoc. 2001;56(1):15–18.
28. Carlson KF, Gravely AA, Noorbaloochi S, Simon AB, Bangerter AK, Sayer NA. Post-deployment injury among new combat veterans enrolled in Veterans Affairs (VA) healthcare. Inj Prev. 2011;17(5):343–347.
29. Brophy M, Zhang X, Xiang H. Injuries among US adults with disabilities. Epidemiology. 2008;19(3):465–471.
30. Johnston BD, Rivara FP, Droesch RM, Dunn C, Copass MK. Behavior change counseling in the emergency department to reduce injury risk: a randomized, controlled trial. Pediatrics. 2002;110(2):267–274.
31. Chen J, Kersnow M, Simon TR, Dellinger A. Injury-prevention counseling and behavior among US children: results from the second injury control and risk survey. Pediatrics. 2007;119(4):e958–e965.
32. Friedemann-Sanchez G, Griffin JM, Rettman N, Rittman M, Partin MR. Communicating information to families of polytrauma patients: narrative literature review. Rehabil Nurs. 2008;33(5):206–213.
33. Christoffel T, Gallagher SS. Injury prevention: Environmental modification. In:Christoffel T, Gallagher SS, eds. Injury Prevention and Public Health: Practical Knowledge, Skills, and Strategies. 2nd ed. Sudbury, MA: Jones and Bartlett Publishers; 2006.
34. Anderson MI, Simpson GK, Morey PJ, Cok MMC, Gosling TJ, Gillett LE. Differential pathways of psychological distress in spouse vs. parents of people with severe traumatic brain injury (TBI): multi-group analysis. Brain Inj. 2009;23(12):931–943.
35. Marsh NV, Kersel DA, Havill JH, Sleigh JW. Caregiver burden 6 months following severe traumatic brain injury. Brain Inj. 1998;12(3):225–238.
36. Vangel SJ, Rapport LJ, Hanks RA. Effects of family and caregiver psychosocial functioning on outcomes in persons with traumatic brain injury. J Head Trauma Rehabil. 2011;26(1):20–29.
37. Hall C, Sigford B, Sayer N. Practice changes associated with the department of veterans affairs' family care collaborative. J Gen Intern Med. 2009;25(suppl 1):18–26.
38. Adams PF, Heyman KM, Vickerie JL. Summary health statistics for the U.S. population: national health interview survey, 2008. Vital Health Stat 10. 2009;243:1–104.
39. Coronado VG, Xu L, Basavaraju SV, et al. Surveillance for traumatic brain injury-related deaths—United States, 1997–2007. MMWR Surveill Summ. 2011;60(5):1–32.
40. Carlson KF, Gerberich SG, Church TR, et al. Tractor-related injuries: a population-based study of a five-state region in the Midwest. Am J Ind Med. 2005;47(3):254–264.
41. Dischinger PC, Mitchell KA, Kufera JA, Soderstrom CA, Lowenfels AB. A longitudinal study of former trauma center patients: the association between toxicology status and subsequent injury mortality. J Trauma. 2001;51:877–874.
42. Gentilello LM, Rivara FP, Donovan DM, et al. Alcohol interventions in a trauma center as a means of reducing the risk of injury recurrence. Ann Surg. 1999;230(4):473–483.
43. Harrison-Felix CL, Whiteneck GG, Jha A, DeVivo MJ, Hammond FM, Hart DM. Mortality over four decades after traumatic brain injury rehabilitation: a retrospective cohort study. Arch Phys Med Rehabil. 2009;90:1506–1513.
44. Pentland B, Hutton LS, Jones PA. Late mortality after head injury. J Neurol Neurosurg Psychiatry. 2005;76:395–400.
45. Teasdale TW, Engberg AW. Suicide after traumatic brain injury: a population study. J Neurol Neurosurg Psychiatry. 2001;71:436–440.
46. Ventura T, Harrison-Felix C, Carlson N, et al. Mortality after discharge from acute care hospitalization with traumatic brain injury: a population-based study. Arch Phys Med Rehabil. 2010;91:20–29.
47. Shavelle RM, Strauss D, Whyte J, Day SM, Yu YL. Long-term causes of death after traumatic brain injury. Am J Phys Med Rehabil. 2001;80:510–516.
48. Simpson G, Tate R. Suicidality after traumatic brain injury: demographic, injury and clinical correlates. Psychol Med. 2002;32:687–697.
49. van Ryn M, Sanders S, Kahn K, et al. Objective burden, resources and other stressors among informal cancer caregivers: a hidden quality issue? Psychooncology. 2011;20(1):44–52.
50. National Alliance for Caregiving, AARP. Caregiving in the U.S.—2009. Bethesda, MD: National Alliance for Caregiving and AARP; 2010.
caregivers; health services; rehabilitation; traumatic brain injury; trauma; injury© 2012 Lippincott Williams & Wilkins, Inc.