7

Gaps, Challenges, and Opportunities

Prior chapters have described the knowledge gaps and clinical care challenges that constrain the ability of providers to meet the needs of TBI patients and their families. This chapter recaps those gaps and challenges and details opportunities for addressing them. The discussion is framed within the bio-psycho-socio-ecological view of TBI (Chapter 3), as leveraging these opportunities will require addressing gaps and overcoming challenges associated with all of these dimensions to advance care and improve outcomes for persons with TBI.

GAPS AND CHALLENGES IN CURRENT TBI CARE AND RESEARCH

People who experience TBI and their families need to receive the best possible medical care, be prepared for the nature of the injury and the likely course of recovery, and receive rehabilitation services that address postinjury symptoms and enable a return to the greatest possible level of preinjury function. To meet these needs, health care providers must be able to identify accurately whether a person has experienced a TBI (diagnosis), and have access to, and impart to patients and their families in an accurate and caring manner, the best possible information on the expected course of recovery, the potential for changes in symptoms, and the patient’s future quality of life (prognosis). Also essential is for providers to deliver across the full care continuum care that is both evidence-based and tailored to the individual’s unique characteristics and needs. Yet, major gaps and challenges exist in all these aspects of TBI care, as well as in the research enterprise that supports it. These gaps and challenges fall into the seven key areas detailed below.

1. The high degree of heterogeneity and variability of symptoms among people who experience TBI; the many types of brain injuries they experience; and the suite of biological, psychological, social, and ecological factors that influence outcomes

Understanding of the natural history of recovery after TBI and the factors that predict outcomes remains relatively rudimentary. While a combination of injury severity and premorbid factors influences outcomes after TBI, a granular understanding of how these factors interact is currently lacking. This knowledge gap is attributable to the vast heterogeneity that characterizes TBI along dimensions that include injury setting (military, civilian, sport), and characteristics (fragment or projectile, blunt trauma, blast overpressure injury), severity (concussion to coma), and neuropathology (neuronal, axonal, vascular damage; focal versus diffuse trauma). Preinjury factors—who is injured—and postinjury factors—such as the individual’s response to the injury and social environment—also influence TBI outcomes. Conventional approaches to data analysis have proven inadequate to overcome the complexities of TBI and its multivariate nature, falling short of fully interrogating and integrating data on injury mechanisms, effects, and recovery. Specific gaps and challenges in this area include those described below.

Lack of sufficiently precise criteria and terminology for classifying TBI. There are no universally accepted criteria for classifying TBI, and the current classification scheme of mild, moderate, and severe, based on Glasgow Coma Scale (GCS) summative scores, is imprecise and can lead to biases in care. Variability also exists around a definition of so-called “mild TBI,” which has at least 38 possible definitions, many based on overlapping criteria (Carroll et al., 2004). Varying significantly as well is the terminology used to refer to this condition among clinicians from different medical specialties (e.g., “mild TBI,” “concussion,” “minor head injury”).

Need for greater understanding of the factors that affect outcomes after TBI. The current health care system is designed around making early predictions of a patient’s trajectory after injury.¹ However, the ability to predict outcomes after TBI is limited, and as discussed in Chapter 6, an individual’s condition often evolves. Clinicians are generally unable to predict who will do well and who will do poorly, regardless of acute injury characteristics and initial GCS score. Better understanding of what factors predict outcomes, along with identification and validation of additional outcome metrics, would help improve both care and management. Unmet needs and open questions include the following:

• Better understanding of the effects of biological sex. Sex-based differences in TBI remain poorly understood (Levin et al., 2021). Research has identified differences in inflammation, neuroplasticity, mitochondrial function, and other characteristics of brain injury and recovery associated with biological sex, but “sex differences in mechanistic measures of neuropathology or physiology are not always associated with differences in functional or neurocognitive outcome” (Gupte et al., 2019, pp. 3079–3080). The biological basis for and implications of observed differences, as well as how they should translate to improved care for patients, remain under investigation. There is also limited knowledge on how certain types of interventions, such as neuromodulation, are affected by biological sex (Phillips et al., 2020).
• Better understanding of TBI in older adults. Relative to younger individuals, older adults are more likely to have preinjury comorbidities that complicate treatment, and tools used to measure TBI severity and predict outcomes may be compromised

___________________

¹ Hammond, F. 2021. TBI Care Gaps and Opportunities: Provider Perspectives on the Post-Acute Continuum of Care. Panel discussion during virtual workshop for the Committee on Accelerating Progress in Traumatic Brain Injury Research and Care, March 18, 2021.

by age-related cognitive decline. Further research is needed to inform case management of this population (Stein et al., 2018). In addition, the aging population includes people who experienced TBI at younger ages, highlighting the need to understand how prior TBI intersects with the experience of aging and with neurodegenerative processes that may manifest in later years (Griesbach et al., 2018). Research conducted over longer periods to examine later effects could help elucidate connections between acute and long-term outcomes (such as mortality, risk of dementia, risk of sleep disorder, or risk of Parkinson’s disease), as well as the underlying mechanisms behind these connections.

• Study of TBI outcomes in subgroups that reflect the diversity of the populations who experience TBI—for example, groups stratified by biological factors (such as age or biological sex) and by social constructs (such as race or ethnicity, location [rural, urban], or military service). Trial exclusion criteria may be exacerbating disparities in TBI outcomes by failing to include key groups.² During its information gathering for this study, the committee heard about the need to reflect social determinants of health in study populations and to include information on such social determinants in electronic medical records, as well as a need to go beyond a reliance on demographic covariates conveniently collected in studies to pinpoint sources of disparities. Rather, to be truly informative on these issues, research needs to incorporate empirically and theoretically based variables, particularly those that might be amenable to change, such as perceived cultural competence of providers, racial or ethnic identity, disability identity, and endorsement of traditional gender-role ideologies.³ Studies need also to identify and connect results to implementation strategies aimed at reducing disparities in TBI care and outcomes.

Lack of tools for precision diagnosis, prognosis, and monitoring. No validated tools are currently available for predicting acute or chronic outcomes of TBI or for identifying which patients will have a favorable recovery and which will remain at risk of poor long-term outcomes. It is generally accepted that patients with seemingly equivalent injuries can experience a wide range of recovery trajectories and functional outcomes. As of this writing, no Food and Drug Administration (FDA)-approved biomarker is available that can accurately predict outcomes after TBI. Even serial computed tomography (CT) scanning has only limited predictive ability unless it demonstrates a nonsurvivable injury. More sophisticated imaging, such as magnetic resonance imaging (MRI), may have a role, but although MRI provides additional information, it, too, is incapable of accurately predicting long-term outcomes. Monitoring the evolution of TBI as a condition over time remains a major challenge, and additional research is needed to understand whether or which biomarkers can be most useful for monitoring a patient’s progress and potential responses to treatments. Unmet needs and open questions include the following:

• Identifying and validating new types of biomarkers and incorporating those shown to add value into clinical care. A number of the biomarkers currently available for use and those in development are summarized in Appendix B.
• Ensuring that validation of candidate biomarkers is conducted across diverse patient populations.

___________________

² Thompson, H. 2021. Research to Fill Identified TBI Care Gaps. Panel discussion during virtual workshop for the Committee on Accelerating Progress in Traumatic Brain Injury Research and Care, March 18, 2021.

³ Perrin, P., and J. Iyasere. 2021. Social Determinants of Health. Presentations and panel discussion during virtual workshop for the Committee on Accelerating Progress in Traumatic Brain Injury Research and Care, April 1, 2021.

• Studying the ability of multimodal biomarkers to improve outcome prediction and incorporating them into enhanced prognostic tools.
• Leveraging large datasets to analyze risk factors and develop tools capable of better predicting outcomes based on patient characteristics.

2. Fragmented care systems with multiple handoffs, during which information can be lost and people can fall out of the system, and the lack of a clear health care lead for addressing TBI over time

A person with TBI can experience multiple handoffs from the site of injury and prehospital care, to care locations within a hospital, to post-acute and rehabilitation care, to community-level services and supports. TBI care and recovery also involve multidisciplinary teams, diverse rehabilitation and community interactions during recovery, and needs that can evolve over long time scales. Handoffs can easily turn into gaps, and coordination is challenging across the many phases of care, specialties, types of providers, and community environments. Unmet needs and open questions include the following:

• Ensuring continuity of care from the point of injury to acute care, post-acute care, rehabilitation, and life-course outcomes. As noted, TBI can have lifelong implications. Yet TBI care is currently episodic, and care transitions represent an opportunity for mistakes to be made. Patients and families often are left to navigate specialized services that are confusing and difficult to find and do not share data with other services. As discussed in Chapter 6, many patients with TBI lack access to the types or amount of rehabilitation care and supportive services they may need over time. Families and caregivers of people with TBI continue to report significant burdens and unmet needs.
• Addressing the need for expanded prehospital research and guidance on the management and triaging of TBI patients, along with an effective handoff process from prehospital environments to hospital emergency departments (EDs). Questions include whether blood biomarkers and/or mobile head CT scanners could be used to improve prehospital triaging; what strategies could enable easier identification of severe TBI that requires surgery, so that patients could be brought directly into surgery from a prehospital setting; and what neurocognitive and neuropsychiatric assessment tools can be developed and used effectively in prehospital settings and EDs.
• Meeting the need for enhanced communication and connections between the care and research communities. The breadth of research issues that touch TBI makes it difficult to communicate about TBI among and between researchers. Researchers also need to understand what new or expanded capabilities are most needed by patients, families, and care providers. However, bidirectional communication between the care and research communities—particularly the basic science and preclinical research communities—remains challenging.
• Enhancing data integration, which remains a key issue. While multiple national, state, and organizational registries, databanks, and databases capture clinical and research data on TBI, there is no national TBI registry. Data cannot be integrated across the care continuum from prehospital care environments to hospital-based care, rehabilitation settings, and community services. And while many care environments use electronic health records (EHRs), neither these systems nor the ways in which patients’ TBI information is entered into them are standardized. This lack of

standardized TBI information in EHRs limits the ability to conduct certain types of research, such as comparative effectiveness trials using causal inference methods.

3. Unwarranted variability in the care patients receive and the failure of available evidence-based guidelines to cover all aspects of TBI care

The quality of care received by a TBI patient is provider- and facility-dependent. Absent established and agreed-upon treatment algorithms and endpoints, variation in care increases. In some areas, the best practices for managing brain and patient health are not fully understood, and more research is needed to fill these knowledge gaps (see below). In other areas, the need is to implement more consistently what is already known and increase adherence to clinical practice guidelines. Variability is seen, for example, in adherence to existing decision rules for identifying patients who warrant head CT evaluation. In addition, the TBI care available to patients is variable based on such factors as geographic location and socioeconomic factors, with limited providers or services available in rural and underserved areas.

4. Insufficient access to care, often because of insurance preapproval or reimbursement issues

Access to rehabilitation care remains a key issue for many people with TBI and their families. The rehabilitation and long-term needs of many TBI patients are not sufficiently provided for by current health and insurance systems. It is important that insurance payment for post-acute and long-term TBI care and rehabilitation align better with the evidence on what care helps and when. In particular, current payment guidelines often appear to be inadequate, too restrictive, and poorly timed. For example, windows for effective rehabilitation may lie well downstream of the end of current benefit designs. Lack of access to care, particularly inequitable access to care, may also impede the ability to conduct clinical research by limiting or skewing who can be recruited into studies aimed at understanding and improving outcomes.

5. Lack of evidence-guided therapies for treating TBI

Despite significant efforts, preclinical and clinical research has been difficult to translate into improvements in clinical care. No FDA-approved therapy yet exists for treating underlying damage resulting from TBI, and treatments for acute TBI have failed to demonstrate effectiveness in promoting recovery in large randomized controlled trials with global outcomes.⁴ Both the development of effective treatments and information on when and to whom to administer them are needed. Therapies aimed at reducing TBI morbidity and mortality are actively being tested in large animal models and in clinical trials. Selected examples of acute-stage interventions assessed in recent studies include use of the anticonvulsant drug valproic acid to reduce neurologic impairment after TBI (Dekker et al., 2021; Wakam et al., 2021), use of prehospital plasma transfusion to improve survival of severely injured TBI patients at risk of hemorrhagic shock (Sperry et al., 2018; Gruen et al., 2020), and prehospital or ED administration of tranexamic acid to improve survival and post-TBI outcomes (Brenner et al., 2020; Rowell et al., 2020). The studies illustrate the need to identify the most effective interventions given the heterogeneity of TBI and the many circumstances under which it occurs, as well as the impor-

___________________

⁴ Brody, D. 2021. Research to Fill Identified TBI Care Gaps. Panel discussion during virtual workshop for the Committee on Accelerating Progress in Traumatic Brain Injury Research and Care, March 18, 2021.

tance of and challenges inherent in conducting prehospital research. Continued efforts to identify those patients at higher risk for various adverse outcomes (such as intracranial bleeding) and those most likely to benefit from various interventions will be needed, along with further studies on these and other TBI treatments. Ongoing advances in assessment tools, such as the expanded incorporation of blood-based biomarker tests, may also help refine study designs and clinical decision making on which patients should receive which treatments.

Unmet needs and open questions in preclinical research include the following:

• Understanding longer-term risks and outcomes after TBI. While difficult, there is a continuing need to explore how preclinical research can represent the clinical reality past acute-stage TBI, including whether or what animal models might be developed to better capture long-term effects. In addition, there is a need to conduct preclinical studies over longer times and to make greater use of longitudinal study designs.
• Enhancing preclinical models to be more representative of the human populations who experience TBI. These questions are difficult and may not be possible to address, since different preclinical models are more or less representative of different human attributes, use of the models (e.g., certain types of animal models) may have ethical and cost implications, and no model can fully represent the broad range of human TBI experiences. Moreover, such factors as social determinants of health and quality-of-life issues are difficult to measure in preclinical models. However, experimental injury models have used predominantly male rodents (Gupte et al., 2019), and ensuring greater inclusion of female animals may be one straightforward improvement. It may be possible as well to consider such issues as whether and how preclinical models can incorporate the effects of comorbidities and preexisting medical conditions or multiple injuries (such as polytrauma), and whether and how preclinical models can incorporate other elements, including biological sex, age, time to treatment, and prior history of TBI, associated with differences in outcomes after TBI.

Unmet needs and open questions in clinical research include the following:

• Undertaking clinical research that is more representative of the diverse populations that experience TBI (see also the above discussion on research outcomes). This might include conducting additional studies involving older adults and those with preexisting conditions. It might also include recruiting more women with TBI into trials to advance the limited knowledge of sex and gender differences in symptoms, treatment responses, rehabilitation outcomes, and recovery.
• Conducting clinical research over longer periods to illuminate clinical trajectories beyond the acute period.
• Using study designs that incorporate multiple outcome measures, including mechanistic and functional outcomes; using outcome measures that better capture the full burden of disability; and/or using measures that can better capture outcomes identified as being most relevant to patients and families.

6. Gaps in the knowledge base informing best practices and evidence-based acute care

As described in Chapter 5, further evidence and clinical guidance are needed on aspects of acute TBI care, especially when a patient has concomitant injuries or confounding comorbidities. Unmet needs and open questions include the following:

• Determination of more exact treatment thresholds, including development of additional clinical decision tools to guide acute care of TBI patients. Areas of identified knowledge gaps include (1) use of hypertonic saline, mannitol, and nutrition provision; (2) indications for and timing of intracranial pressure monitoring; (3) indications for and timing of decompressive craniectomy, despite recent guidance on this issue in patients with severe TBI; (4) which vasoactive agent should be used to augment cerebral perfusion pressure; (5) lack of a clear definition for hypotension in pediatric and geriatric populations; and (6) transfusion thresholds studied in sufficient numbers of patients to allow stratification by sex and TBI severity.
• Roles and indications for emerging therapies, such as superoxide radicals, deep brain stimulators to awaken patients in minimally conscious states, vagal nerve stimulators, or use of stimulating devices to treat post-TBI depression.
• Cost/benefit evaluation and translation to clinical practice of advanced imaging technologies, such as MRI. Evidence-based guidelines for when and in whom to use MRI or other newer imaging modalities are currently lacking, and there is a need for decision aids to identify patients who warrant further evaluation with brain MRI imaging. Improved informatics are also needed for consistent interpretation of brain imaging results.
• Effects of multisystem trauma on TBI and uncertainty regarding optimal management of such patients to improve outcomes. The optimal interventions and their timing for TBI patients experiencing life-threatening hemorrhage is one significant concern, with unknowns including blood pressure target, use of resuscitation fluids, and the priority order of surgical interventions. The timing of non-life-threatening, noncranial surgeries in the setting of TBI also remains unclear (Wang et al., 2007). The contribution of medical comorbidities and preexisting risk factors, such as hypertension, diabetes, hypercholesterolemia, and prior cardiovascular disease, to clinical outcomes after TBI needs further research.
• Understanding of how multimodal data can be integrated to enhance clinical decisions, including optimizing diagnosis, informing transport to the appropriate level of care, and informing appropriate use of neuroimaging. For example:
  • Could blood-based biomarker tests be incorporated into decision guidance on when to conduct brain CT imaging? Could imaging or blood-based biomarkers further aid decision making on surgical timing for intracranial bleeding?
  • Would equipping ambulances with mobile CT scanners so paramedics could perform CT scans at the injury scene or en route to the hospital be useful and cost-effective?
  • Would incorporating point-of-care testing for approved blood biomarkers, such as glial fibrillary acidic protein (GFAP, a structural protein found in astrocytes) and ubiquitin carboxy-terminal hydrolase L1 (UCH-L1, a highly abundant enzyme found in neuronal cell bodies), into prehospital guidelines improve triaging of patients to the appropriate level of care?
  • Would widely implementing neurocognitive and neuropsychiatric TBI assessment tools in hospital EDs be effective at enhancing diagnosis and management of TBI?

7. Gaps in the knowledge base informing rehabilitation interventions

The evidence base for the effectiveness of many rehabilitation interventions is limited, which also complicates obtaining insurance coverage for such therapies. Few well-controlled

intervention trials have been conducted in the post-acute stage, and many of those trials that have been conducted were applied to exclusive samples and settings. Research in this area has been hampered by limited follow-up, particularly for studies employing neuroimaging and biomarkers, and by the inconsistencies in definitions of injury severity noted earlier. Thus the availability of systematic, evidence-based interventions—therapies that are also culturally resonant and patient-centered—is limited. Unmet needs and open questions include the following:

• Development and evaluation of rehabilitation interventions that are tailored to individual and family needs and strengths. As part of such efforts, stakeholder engagement will be needed to identify and refine research questions of greatest interest and impact for patients, families, and caregivers (Winter et al., 2016).
• A general lack of efficacy, effectiveness, pragmatic, causal inference/comparative effectiveness, and implementation studies on comprehensive TBI rehabilitation. As noted above, the evidence base also remains limited on how best to tailor and deliver rehabilitation care for older adults with TBI.
• A need to evaluate the most effective and practical methods for providing follow-up to patients and families once they return to their communities, including through community-based services, interactions with primary care providers, self-management, and/or case management. As one example, a pragmatic trial is currently under way on optimized follow-up for persons with moderate to severe TBI who received inpatient rehabilitation (Fann et al., 2021).

OPPORTUNITIES

Learning from and Connecting Existing Care Systems and Networks

The TBI landscape is not a blank canvas. Connections can be strengthened among existing networks and models of care and research, and best practices further disseminated, to help build an optimized TBI system. As reported in prior chapters, there remains a general disconnect between acute hospital-based care and rehabilitation and community reintegration systems, including a need for greater connections among smaller trauma centers and rehabilitation services.⁵ In addition, many patients who experience “mild TBI” are assessed by and referred from primary care practices rather than being seen in trauma systems.

Continuing efforts are needed to connect prehospital care and research and to connect primary care practices with other TBI care system elements. Better linkages and stronger connections among components of the TBI care system would benefit patients and families. At the rehabilitation stage, periodic reassessment of a person’s progress and needs could better inform decision making about interventions that could help them and their families regain function and recover to the greatest extent possible. Selected examples of existing TBI care and research efforts that can be integrated into such a TBI system are described below. Learning from and leveraging the success of such efforts can provide an important foundation for the development of an integrated TBI system.

___________________

⁵ Wright, D. 2021. TBI Care Gaps and Opportunities: Provider Perspectives on the Acute-Stage Continuum of Care; and Moore, M. 2021. Disparities in TBI Outcomes. Presentations and panel discussion during virtual workshop for the Committee on Accelerating Progress in Traumatic Brain Injury Research and Care, March 16 and March 18, 2021.

Department of Defense (DoD) and Department of Veterans Affairs (VA) care systems. DoD’s military health system is a $50 billion enterprise that comprises 51 hospitals, 424 clinics, and 248 dental clinics. It supports 9.5 million beneficiaries and encounters between 12,000 and 14,000 TBI-related cases each month.⁶ The DoD Intrepid Network for TBI, including the National Intrepid Center of Excellence and a network of Intrepid Spirit Centers, provides comprehensive outpatient services for military service members with TBI, particularly concussion (Lee et al., 2019). In addition, eligible service members and Veterans with TBI may receive care through the VA’s Polytrauma System of Care (PSC), which was established in coordination with DoD to provide a cadre of specialized clinicians to address the rehabilitation needs of this TBI/polytrauma cohort. The PSC operates as a hub-and-spoke model, with five regional Polytrauma Rehabilitation Centers. The centers provide specialized inpatient rehabilitation and coordinate with numerous network sites to offer comprehensive and sustained care from the stage of acute rehabilitation to transition back to the person’s community and ongoing outpatient care (Chung et al., 2015).⁷ Additional programs, including programs focused on assistive technology, amputation needs, case management, and vision rehabilitation, work in concert to support care.⁸ The DoD/VA system casts a wide net to identify TBI among service members and Veterans, and because it is an encompassing system, it aids in reducing some of the variability in and barriers to care that may be encountered in current civilian health systems.

Civilian trauma care systems. Trauma care in the United States involves state, local, and regional systems for injury response, from emergency medical services (EMS) to designated trauma centers. The American College of Surgeons (ACS) provides a voluntary process for verification of Level I, II, and III trauma centers as having the resources and capabilities outlined in Resources for Optimal Care of the Injured Patient:

Movement toward a national trauma care system. Many trauma health systems are strong but disjointed, and there are a reported “50 million Americans unable to reach a level I and II center within 60 min” (Soto et al., 2018, p. 78). Efforts are under way to foster the creation of a better-integrated, national trauma care system, building on recommendations made in the National Academies report A National Trauma System: Integrating Military and Civilian Trauma Systems to Achieve Zero Preventable Deaths After Injury (NASEM, 2016). The Mission Zero Act of 2019⁹ incorporated military trauma care providers into civilian trauma systems. Significant efforts to advance this concept are also under way through such

___________________

⁶ Lee, K. 2021. System Challenges for TBI Care. Panel discussion during virtual workshop for the Committee on Accelerating Progress in Traumatic Brain Injury Research and Care, March 30, 2021.

⁷ See also https://www.polytrauma.va.gov/system-of-care/index.asp (accessed August 31, 2021).

⁸ See, for example VA Assistive Technology: https://www.prosthetics.va.gov/AssistiveTechnology/index.asp; Regional Amputation Centers: https://www.prosthetics.va.gov/asoc/Regional_Amputation_Centers.asp; Care Management for Post 9/11 Veterans: https://www.oefoif.va.gov/; and Blind Rehabilitation Centers: https://www.rehab.va.gov/PROSTHETICS/blindrehab/locations.asp (all accessed August 31, 2021).

⁹ Part of Public Law No. 116-22: Pandemic and All Hazards Preparedness and Advancing Innovation Act.

organizations as the ACS. Fostering relationships and feedback between larger and smaller trauma centers is one area of opportunity for system development.

Civilian and VA Longitudinal Research on TBI Outcomes. The National Institute on Disability, Independent Living, and Rehabilitation Research (NIDILRR) within the Administration for Community Living has established a network of TBI Model Systems (TBIMS) sites that focus on rehabilitation research and knowledge translation. Currently, 16 sites distributed throughout the United States are funded. To be part of the network, centers must provide comprehensive multidisciplinary rehabilitation, conduct clinical research studies, share results though the Model Systems Knowledge Translation Center (MSKTC), and submit longitudinal data for inclusion in the TBIMS National Database. The TBIMS National Database is currently the largest multisite longitudinal TBI resource with information from preinjury through acute care and long-term outcomes among patients who received inpatient rehabilitation (Tso et al., 2021). As of this writing, the TBIMS centers were collecting data on outcomes up to 30 years postinjury. In 2008, the VA and NIDILRR collaborated to establish a VA TBIMS among the five Polytrauma Rehabilitation Centers that address TBI rehabilitation in Veterans. The VA TBIMS includes data on Veterans and service members who experienced TBI of any severity. The TBIMS network enables essential research that informs best practices in care for people with TBI to optimize functional outcomes, particularly in rehabilitation.

Care organization accreditors.¹⁰ Many trauma centers participate in the ACS verification program, while the Commission on Accreditation of Rehabilitation Facilities (CARF) accredits a number of programs providing rehabilitation services. The Joint Commission is the largest accreditor of health care organizations in the United States. Other organizations, such as the National Commission on Correctional Health Care, play roles in accrediting components of the U.S. health care system. The organizations have quality standards and metrics on which accreditation is based. Studies have generally found that adopting quality standards and achieving accreditation contribute to the delivery of quality care for patients and families, including providing recommended care, applying evidence-based practices, and enhancing efficiency, safety, effectiveness, timeliness, and patient-centeredness (Araujo et al., 2020; Bogh et al., 2016; Devkaran et al., 2019; Falstie-Jensen et al., 2017; Hussein et al., 2021). For example, a study of 3,891 U.S. hospitals found that those that were Joint Commission-accredited in 2004 and 2008 performed better than nonaccredited hospitals on most of 16 performance measures (Schmaltz et al., 2011), while a study of 246 CARF-accredited nursing homes showed better outcomes on short-stay quality measures relative to 15,393 non-CARF-accredited nursing homes (Wagner et al., 2013). Although not all studies report consistent impacts from accreditation, the process represents an opportunity to encourage diverse care settings to incorporate current clinical guidance and best practices in care delivery to support positive outcomes after TBI. Engaging relevant accreditation organizations thus provides an additional opportunity to help create and establish an improved TBI system across the full care continuum.

Approaches to conducting out-of-hospital research. The ability to conduct research outside of acute inpatient settings remains challenging, but is essential to advancing TBI care and research. Relevant settings include prehospital environments, EDs, and outpatient rehabilitation and home and community services locations. The Resuscitation Outcomes Consortium (ROC) is an example of an effort to improve out-of-hospital cardiac arrest and trauma

___________________

¹⁰ See https://www.facs.org/quality-programs/trauma/tqp/center-programs/vrc; http://www.carf.org/Accreditation (both accessed March 2, 2022).

outcomes. The ROC, funded by the National Institutes of Health (NIH), DoD, the American Heart Association, and organizations in Canada, encompassed 10 regions in North America. Research was initiated in the field with a waiver from informed consent. The effort continued for more than 10 years and demonstrated that out-of-hospital research could be performed on a large scale and over diverse geographic regions. The ROC completed two major TBI studies—one on the efficacy of a bolus of hypertonic saline and the other on the efficacy of tranexamic dosing regimens (Bulger et al., 2010; Rowell et al., 2020). Although the studies did not find significant differences in outcomes, they proved that research and 6-month follow-up could be assessed in 85 percent of patients (Rowell et al., 2020).

Box 7-1 describes features within the VA system that can serve the specialized needs of patients with TBI. A nationally integrated health care system for the provision of coordinated rehabilitation services does not exist in the private sector. In addition, the demographics of the patient population served by the VA are skewed toward a greater percentage of male versus female patients, and possibly toward an older patient population, compared with private-sector rehabilitation providers. Although the rehabilitation care provided by the VA is unique relative to what is available in many civilian care settings, these practices can serve as a benchmark for rehabilitation care and provides examples of strategies that may be adaptable to other settings.

Enhancing and Connecting Existing Data Systems

Multiple registries, databanks, and information systems capture aspects of TBI care and research. Two issues—data integration to improve the capability to collate and analyze patient TBI information, and the ability to tie patient outcomes back to clinical care—remain, given the varied settings in which TBI occurs, the multiple phases of care and recovery, the heterogeneity of injury and diversity of people affected by TBI, and the active base of preclinical and clinical research. This issue starts with the need to link prehospital data to in-hospital or posthospital registries, connecting data across the continuum of care and services. The need for trauma information systems and access to patient-level data was also noted in the Zero Preventable Deaths report (NASEM, 2016); efforts to advance and integrate data systems as part of developing a national trauma system may provide valuable building blocks applicable to TBI as well.

Selected examples of databanks and registries relevant to TBI include the National EMS database, NEMSIS, supported by the National Highway Traffic Safety Administration (NHTSA); the ACS’s National Trauma Databank (NTDB), a voluntary aggregation of U.S. trauma registry data; and multiple state trauma registries. DoD’s Trauma Registry (DoDTR) was established in 2005, modeled after the NTDB; it incorporates subregistries, including one for TBI. The VA’s Traumatic Brain Injury Veterans Health Registry collects information on Veterans with a TBI-related diagnostic code who served in Iraq or Afghanistan, applied for VA disability benefits, and/or screened positive for TBI on the screening assessment administered when a Veteran seeks VA health care. In addition, the NIDILRR and VA TBI Model Systems databases contain information on long-term outcomes and rehabilitation.

Although there is no U.S. national TBI registry, efforts to connect or crosslink data sources provide an opportunity to better harness the wealth of collected TBI information. Potential areas of study that might be aided by analyzing registry data include optimization and timing

of patient transport, indications for intracranial pressure monitoring and novel monitoring methods, optimal ventilation strategies, and development of point-of-care biomarkers.

Care providers and hospitals have also widely implemented EHRs. Large EHR providers such as EPIC, Cerner, and Meditech are estimated to provide software to almost three-quarters of the hospital market.¹¹ In addition, the VA is in the process of a large-scale EHR modernization effort,¹² due to be completed by 2028, whose purpose is to create a connected EHR across DoD and VA medical centers and facilitate connections with community care providers. These efforts provide valuable opportunities to establish within EHR systems the collection of standardized patient information needed to support research on TBI care and outcomes. For example, data on social determinants of health are missing from many systems, but greater collection of such information could be encouraged by including relevant EHR fields. The VA, for instance, includes primary care screening questions on housing instability and refers at-risk Veterans for services (Montgomery et al., 2013). And a project to implement the Centers for Disease Control and Prevention’s (CDC’s) Stopping Elderly Accidents, Deaths, and Injuries (STEADI) program for older adults at risk of falls identified the incorporation of screening tools into the clinic’s workflows and EHR system as key components of the effort (Casey et al., 2017). This issue may represent an opportunity for lessons learned and further public–private dialogue and partnership efforts.

In the research community, investigators are required to enter data from federally funded TBI research into the Federal Informatics System for Research Data (FITBIR), used by both NIH and DoD. Entered information includes Common Data Elements (CDEs) encompassing information from clinical studies, and now being extended to include preclinical research. Although entering the information into the system can reportedly be time-consuming, the system enables valuable access to DoD- and NIH-funded data in a common place.

Collectively, crosslinking or finding improved ways to share and connect information from these valuable resources provides an opportunity to advance an integrated TBI system.

Analyzing Data from Recent Large-Scale and Long-Term Research Efforts

Large-scale and longitudinal TBI studies, such as those highlighted throughout this report and in Appendix A, are providing a wealth of data to enable better understanding of TBI. It has been a challenge to apply a precision medicine approach to TBI treatment, in which management strategies are targeted to the right patient at the right time and in the right setting. The lack of predictive tools also inhibits rational initiation and timing of interventions to reduce overall disability associated with TBI. There is an opportunity to leverage the information being obtained from the past decade of TBI efforts to better understand how multiple individual biological, psychological, social, and ecological factors affect outcomes after TBI, and to identify and validate new imaging, blood-based, and other types of biomarkers that can contribute now and in the future, as research advances.

Significant efforts have been devoted to discovery, validation, and optimization of objective TBI biomarkers that can supplement such injury information as physiological measures and vital signs. A biomarker is an indicator of the presence of a condition, such as TBI, and

___________________

¹¹ See https://klasresearch.com/blog/top-trends-in-emr-market-share/856 and https://www.beckershospitalreview.com/ehrs/ehr-market-share-2021-10-things-to-know-about-major-players-epic-cerner-meditech-allscripts.html (both accessed March 2, 2022).

¹² See https://www.ehrm.va.gov (accessed March 2, 2022).

can also be measured or monitored for changes in the condition over time. The primary TBI biomarker types are

• neuroimaging, such as CT and MRI;
• biofluid-based, such as those from cerebrospinal fluid (CSF), saliva, urine, or blood; and
• neurophysiological, such as electroencephalography (EEG) and eye tracking.

These biomarkers are being studied to identify such features as the nature, location, and extent of injuries—such as bleeding and swelling in the brain; disruption of the blood–brain barrier; inflammation; changes in cellular metabolism; and damage to brain cells, such as changes in brain cell signaling and markers of neurodegeneration—to identify and track the effects of TBI. During the past decade, candidate biomarkers have emerged that may provide insight into the mechanisms and dynamic course of TBI (Agoston et al., 2017; Di Battista et al., 2015; Diaz-Arrastia et al., 2014; McCrea et al., 2020; Zetterberg et al., 2013). Appendix B reviews progress in TBI biomarkers in greater detail.

The lack of demonstrated effectiveness for TBI therapies in randomized controlled trials supports the need for more precise recruitment of participants into trials designed to test the efficacy of these therapies. Lumping heterogeneous patient populations together for recruitment into clinical trials and evaluation of trial results has made it challenging to demonstrate effectiveness. It is difficult, for example, to target a drug to the ill-defined category of “mild” TBI. It is also challenging to target a drug to “patients who have in common a ‘severe’ injury phenotype who may vary widely in other injury classification schemes, such as those based on pathoanatomic or pathophysiological features, which may be more relevant to the neuroprotectant action of a particular intervention” (Saatman et al., 2008, p. 722). In today’s world of precision medicine, no one would imagine conducting a drug trial for mild, moderate, or severe cancer.

Analyzing information from recent studies has provided a basis for identifying and refining TBI “phenotypes” that can aid in more standardized and precise stratification. Ongoing efforts to identify and validate multiple biomarkers are also contributing to these efforts. For example, Gravesteijn and colleagues (2020) conducted a cluster analysis of the European CENTER-TBI database and identified four patient clusters “defined by injury mechanism, presence of major extracranial injury and GCS” (p. 1002). A recent paper from Pugh and colleagues (2021) describes a machine learning–based tool and data repository for identifying TBI patient phenotypes and enhancing precision care. In addition to aiding in patient stratification for trials, predictive and pharmacodynamics biomarkers may be able to serve as secondary endpoints in trials evaluating new therapies, providing more objective means of assessing response to treatment at the neurobiological or mechanistic level. Ideally, advances in biomarkers and other tools for phenotyping TBI will increase the likelihood of success in developing novel treatments to improve outcomes.

Developing and Disseminating an Improved Classification System for TBI Diagnosis, Prognosis, Monitoring, and Research

As noted earlier, current methods for classification of acute TBI are based primarily on the GCS score (GCS 3–8: severe, GCS 9–12: moderate, GCS 13–15: mild). Some operational definitions incorporate acute injury characteristics (e.g., loss of consciousness or posttraumatic amnesia) into the classification of mild, moderate, or severe TBI. The construct of “complicated mild TBI” is often applied in reference to patients with GCS 13–15 and

abnormal radiologic findings (e.g., positive head CT or MRI). Still, in most clinical settings, the universal TBI taxonomy is broadly reduced to “mild, moderate, or severe.” Knowledge gained from the past decade of TBI studies has yielded clinical, imaging, and blood-based biomarker elements with diagnostic and prognostic value that can be incorporated to improve on traditional TBI classification. As new biomarkers are developed and validated, it will be important to reevaluate periodically the TBI classification system and how it can be refined and further improved.

Establishing a more accurate framework for characterizing and classifying TBI would help achieve the goals of precision medicine models for individualized treatment, improved prognostic models for predicting a patient’s trajectory of recovery and outcome, and more precise selection of subjects for clinical trials (e.g., phenotyping).

A multidimensional “basic TBI classification” system could be adopted now. It would include

• GCS sum, or total, score (not mild, moderate, or severe);
• clinical neuroimaging results (head CT scan and brain MRI, when available and clinically indicated); and
• blood biomarker results (acute GFAP and UCH-L1 levels, when available and clinically indicated).

This basic system would provide a more precise and objective description of the injury, and while more complex than the current system, would be feasible in the acute care arena.

An “advanced TBI classification” system could be developed to extend the basic system by including additional information that could be obtained in the acute and sub-acute settings, fully characterizing the patient based on factors now known to affect recovery and outcomes after TBI. This information would include

• mechanism of injury;
• GCS score with each of the three components described (e.g., eye, verbal, and motor responses);
• acute injury characteristics, including duration of unconsciousness, duration of posttraumatic amnesia, and duration of altered mental status;
• clinical neuroimaging results;
• blood biomarker results;
• premorbid factors, including history of prior TBI or neurologic disorder, psychiatric or substance use disorder, and major health conditions;
• social factors, such as family support, living situation, and employment status; and
• health care disparities, such as inability to access follow-up care.

Because a single point assessment is insufficient for a dynamic and complex condition such as TBI, which is affected by many factors, repeated measures will be needed as the person’s condition evolves. Implementing improved, multimodal approaches for classifying patients could provide the opportunity for more precise targeting and monitoring of care, as well as more successful translation of therapies from the bench to the bedside.

Making Greater Use of All Types of Study Designs

The TBI field will need to use multiple research design tools, beyond randomized controlled trials, to improve the evidence base in a manner that also allows for implementation

in the real world. Using the full range of clinical trial designs and engaging the voices of those living with TBI would provide important opportunities to advance the evidence base across the continuum of care and recovery.

A variety of clinical trial methods are needed to address the complexities of TBI. Measuring longitudinal outcomes with simple trials is unsustainably inefficient.¹³ Evaluation needs to focus not on time points but on trajectories of recovery from TBI (e.g., using such approaches as recovery progression models), and to use such methods as Bayesian adaptive design. A “one bite at a time” approach has been proposed for tackling TBI as well, in which studies would address specific symptoms and deficits in particular TBI subpopulations (Brody et al., 2019).

Establishing a platform approach to clinical trials would also be valuable for enhancing the efficiency of discovery and tackling the complexity of TBI. A first step in such an approach is often to build the initial structure of the platform so that the cost of conducting trials using the platform is lower than that of not using it.¹⁴ An example of such a platform is the Strategies to Innovate Emergency Care Clinical Trial Network (SIREN),¹⁵ funded by the National Institute of Neurological Disorders and Stroke (NINDS) and the National Heart, Lung, and Blood Institute (NHLBI), which consists of centers with neurological and cardiac emergency and critical care expertise. Investigator-initiated, funded grants are executed through the trial network and have included Brain Oxygen Optimization in Severe TBI Phase 3 (BOOST-3: NCT03754114), Hyperbaric Oxygen Brain Injury Treatment Trial (HOBIT: NCT02407028), and ancillary studies. An aim of establishing trial platforms is to reduce the overall time and cost investment for individual trials by using a network of partner clinical trial sites, flexible Bayesian adaptive designs, and streamlined regulatory processes.

The simultaneous application of multiple treatments and interventions, a common aspect of TBI rehabilitation, further raises the question of whether it is meaningful to try to define the effects of individual rehabilitation components rather than the services and delivery system more holistically. Conducting trials in which interventions are tailored to the needs of individual patients and families rather than being standardized is a challenge, but certain research design methods can aid in achieving the goals of both tailoring and flexibility (e.g., pragmatic trials can be designed to allow for some tailoring to the individual and flexibility in application while yielding evidence that can be applied to a larger proportion of the population who receive care outside the controlled environments often required by randomized controlled trials). There is also an ongoing need for “evidence in TBI rehabilitation that delineates the extent that differences in outcomes are attributable to patients’ characteristics such as age, severity, time since injury, and pre-injury factors, and how much outcomes can be attributed to the timing and dose of specific rehabilitation interventions” (Horn et al., 2015, p. S189). Further research will be needed to compare outcomes of persons who receive different types of interventions and different systems of care. The use of ecological momentary assessments may be one feasible method for measuring outcomes, such as mental health, that can change over time (Juengst et al., 2021).

In addition, using participatory research approaches strengthens the voice of people and families living with TBI in the research process and can help ensure that issues of greatest interest to these communities are being addressed. The use of such approaches also promotes translation of findings to the community (Frank et al., 2014; Winter et al., 2016).

___________________

¹³ Lewis, R. 2021. Research to Fill Identified TBI Care Gaps. Panel discussion during virtual workshop for the Committee on Accelerating Progress in Traumatic Brain Injury Research and Care, March 18, 2021.

¹⁴ Ibid.

¹⁵ Yaffe, K. 2021. Research to Fill Identified TBI Care Gaps. Panel discussion during virtual workshop for the Committee on Accelerating Progress in Traumatic Brain Injury Research and Care, March 18, 2021.

Expanding Consensus Processes for Guidance on TBI Care

As discussed in Chapters 5 and 6, multiple guidelines inform aspects of TBI care. Because there are so many types of TBI in so many populations and involving so many clinical specialties and care providers, there is a need to systematically review and coordinate the content of clinical guidance to promote consistent access to high-quality care and reduce unwarranted variability. Guideline development processes consider the strength of the evidence available to support recommendations, often ranking evidence on a scale in which Level 1 evidence has been obtained through randomized controlled trials, evidence obtained through other types of studies is considered Level 2, and practices based on expert opinion are considered Level 3 (VA, 2019). Guidelines differ in how they treat different types of evidence and which types they include in the guidance development process.

It has been a significant challenge to obtain evidence from randomized controlled trials to inform all relevant aspects of TBI care and recovery. In some cases, such a trial may not yet have been conducted on a specific question. In other cases, such as for many rehabilitation interventions, evidence has been difficult to obtain through randomized controlled trials, and using an expanded toolkit of study designs could better address salient research questions. As a result, guidelines to inform TBI care, and reimbursement for care, have gaps if they rely only on evidence from such trials.

Consensus-based processes represent an important mechanism for collating and disseminating best-practice guidance. One example in which this approach has been successful is the work of the Concussion in Sport group. Over six cycles since 2001, the group has developed and iterated an influential Consensus Statement on Concussion and has successfully disseminated standardized recommendations for the identification and management of this type of TBI in sports. The process for developing the statement incorporates both the latest high-quality evidence and best clinical practices, and can serve as an exemplar for the development of clinical guidelines for TBI more broadly.

The committee believes that multiple types of evidence remain important for informing TBI care. Thus there are opportunities for further rigorously conducted, consensus-based processes to guide care across the phases of TBI, including by providing practical information on current best practices in areas in which evidence is still being gathered.

Harnessing Insights from Implementation Science

Implementation science is the study of methods for promoting the “systematic uptake of proven clinical treatments, practices, organizational, management interventions into routine practice, and hence to improve health.”¹⁶ Clinical research is needed to identify “what works,” while implementation research is needed to identify “how to make it work” (Forrest et al., 2014, p. 1172). To maximize practical impact, specific, concrete aims and implementation strategies must be married to the generation of new knowledge. Both considering in advance what can go wrong through initial planning and “premortem analysis” and learning afterward from study limitations or failures are valuable.

Incorporating insights from implementation science into TBI study design and planning for more effective translation of research results to practice represents an important opportunity to accelerate progress in TBI care. Taking advantage of this opportunity would likely require greater awareness of the role of implementation science among researchers and

___________________

¹⁶ Beidas, R. 2021. “Lessons from Implementation Science.” Presentations and panel discussion during virtual workshop for the Committee on Accelerating Progress in Traumatic Brain Injury Research and Care, April 1, 2021, and quoting from Eccles et al., 2012, p. 2.

clinicians and closer engagement with experts in this field. In particular, widespread adoption of improvements in TBI care will depend on close and ongoing relationships among researchers, active clinicians, and other community stakeholders, and between centers of excellence in TBI care and community-based hospitals and clinicians. Given the extent of the TBI burden and the opportunities for improvement, TBI may indeed be an ideal topic for expanded implementation science research.

CONCLUSIONS

Despite ongoing gaps and challenges, there are a number of opportunities for advancing TBI care and research. These opportunities build on prior efforts and investments made to advance the understanding of TBI and to develop needed care and research infrastructures. Collectively, these opportunities can help the field strengthen and create an optimized system for TBI.

REFERENCES

Eccles, M. P., R. Foy, A. Sales, M. Wensing, and B. Mittman. 2012. Implementation science six years on—Our evolving scope and common reasons for rejection without review. Implementation Science 7:71 https://doi.org/10.1186/1748-5908-7-71.

Falstie-Jensen, A. M., S. B. Bogh, E. Hollnagel, and S. P. Johnsen. 2017. Compliance with accreditation and recommended hospital care-a Danish nationwide population-based study. International Journal for Quality in Health Care 29(5):625-633.

Fann, J. R., T. Hart, M. A. Ciol, M. Moore, J. Bogner, J. D. Corrigan, K. Dams-O’Connor, S. Driver, R. Dubiel, F. M. Hammond, M. Kajankova, T. K. Watanabe, and J. M. Hoffman. 2021. Improving transition from inpatient rehabilitation following traumatic brain injury: Protocol for the BRITE pragmatic comparative effectiveness trial. Contemporary Clinical Trials 104:106332. https://doi.org/10.1016/j.cct.2021.106332.

Forrest, C. B., P. Margolis, M. Seid, and R. B. Colletti. 2014. PEDSnet: How a prototype pediatric learning health system is being expanded into a national network. Health Affairs (Project Hope) 33(7):1171-1177.

Frank, L., E. Basch, J. V. Selby; and Patient-Centered Outcomes Research Institute. 2014. The PCORI perspective on patient-centered outcomes research. Journal of the American Medical Association 312:1513-1514.

Gravesteijn, B., C. Sewalt, A. Ercole, C. Akerlund, D. Nelson, A. Maas, D. Menon, H. Lingsma, E. Stseyerberg, and the Collaborative European NeuroTrauma Effectiveness Research for Traumatic Brain Injury Collaborators. 2020. Toward a new multi-dimensional classification of traumatic brain injury: A collaborative European NeuroTrauma Effectiveness Research for Traumatic Brain Injury Study. Journal of Neurotrauma 37(7):1002-1010.

Griesbach, G. S., B. Masel, R. E. Helvie, and M. J. Ashley. 2018. The impact of traumatic brain injury on later life: Effects on normal aging and neurodegenerative diseases. Journal of Neurotrauma 35(1):17-24.

Gruen, D. S., F. X. Guyette, J. B. Brown, D. O. Okonkwo, A. M. Puccio, I. K. Campwala, M. T. Tessmer, B. J. Daley, R. S. Miller, B. G. Harbrecht, J. A. Claridge, H. A. Phelan, M. D. Neal, B. S. Zuckerbraun, M. H. Yazer, T. R. Billiar, and J. L Sperry. 2020. Association of prehospital plasma with survival in patients with traumatic brain injury: A secondary analysis of the PAMPer cluster randomized clinical trial. JAMA Network Open 3(10):e2016869.

Gupte, R., W. Brooks, R. Vukas, J. Pierce, and J. Harris. 2019. Sex differences in traumatic brain injury: What we know and what we should know. Journal of Neurotrauma 36(22):3063-3091.

Horn, S., J. Corrigan, J. Bogner, F. Hammond, R. Seel, R. Smout, R. Barrett, M. Dijkers, and G. Whiteneck. 2015. Traumatic brain injury practice-based evidence study: Design and patients, centers, treatments, and outcomes. Archives of Physical Medicine and Rehabilitation 96(8):S178-S196.

Hussein, M., M. Pavlova, M. Ghalwash, and W. Groot. 2021. The impact of hospital accreditation on the quality of healthcare: A systematic literature review. BMC Health Services Research 21(1):1057.

Juengst, S. B., L. Terhorst, A. Nabasny, T. Wallace, J. A. Weaver, C. L. Osborne, S. P. Burns, B. Wright, P. S. Wen, C. N. Kew, and J. Morris. 2021. Use of mHealth technology for patient-reported outcomes in community-dwelling adults with acquired brain injuries: A scoping review. International Journal of Environmental Research and Public Health 18(4):2173.

Lee, K. M., W. Greenhalgh, P. Sargent, H. Chae, S. Klimp, S. Engel, B. Merritt, T. Kretzmer, L. Bajor, S. Scott, and S. Pyne. 2019. Unique features of the US Department of Defense Multidisciplinary Concussion Clinics. Journal of Head Trauma Rehabilitation 34(6):402-408.

Levin, H., N. Temkin, J. Barber, L. Nelson, C. Robertson, J. Brennan, M. Stein, et al. 2021. Association of sex and age with mild traumatic brain injury—Related symptoms: A TRACK-TBI study. JAMA Network Open 4(4):e213046. https://doi.org/10.1001/jamanetworkopen.2021.3046.

McCrea, M., S. Broglio, T. McAllister, J. Gill, C. Giza, D. Huber, J. Harezlak, K. Cameron, M. Houston, G. McGinty, J. Jackson, K. Guskiewicz, J. Mihalik, M. A. Brooks, S. Duma, S. Rowson, L. Nelson, P. Pasquina, T. Meier, and the CARE Consortium Investigators. 2020. Association of blood biomarkers with acute sport-related concussion in collegiate athletes: Findings from the NCAA and Department of Defense CARE Consortium. JAMA Network Open 3(1):e1919771. https://doi.org/10.1001/jamanetworkopen.2019.19771.

Montgomery, A. E., J. D. Fargo, T. H. Byrne, V. R. Kane, and D. P. Culhane. 2013. Universal screening for homelessness and risk for homelessness in the Veterans Health Administration. American Journal of Public Health 103(Suppl 2):S210-S211.

NASEM (National Academies of Sciences, Engineering, and Medicine). 2016. A national trauma care system: Integrating military and civilian trauma systems to achieve zero preventable deaths after injury. Washington, DC: The National Academies Press.

Phillips, A., S. Sami, and M. Adamson. 2020. Sex differences in neuromodulation treatment approaches for traumatic brain injury: A scoping review. Journal of Head Trauma Rehabilitation 35(6):412-429.

Pugh, M. J., E. Kennedy, E. Prager, J. Humpherys, K. Dams-O’Connor, D. Hack, M. K. McCafferty, J. Wolfe, K. Yaffe, M. McCrea, A. Ferguson, L. Lancashire, J. Ghajar, and A. Lumba-Brown. 2021. Phenotyping the spectrum of traumatic brain injury: A review and pathway to standardization. Journal of Neurotrauma.https://doi.org/10.1089/neu2021.0059.

Rowell, S. E., E. Meier, B. McKnight, D. Kannas, S. May, K. Sheehan, E. Bulger, A. Idris, J. Christenson, L. Morrison, R. Frascone, P. Bosarge, M. R. Colella, J. Johannigman, B. Cotton, J. Callum, J. McMullan, D. Dries, B. Tibbs, N. Richmond, M Weisfeldt, J. Tallon, J. Garrett, M. Zielinski, T. Aufderheide, R. Gandhi, R. Schlamp, B. Robinson, J. Jui, L. Klein, S. Rizoli, M. Gamber, M. Fleming, J. Hwang, L. Vincent, C. Williams, A. Hendrickson, R. Simonson, P. Klotz, G. Sopko, W. Witham, M. Ferrara, and M. Schreiber. 2020. Effect of out-of-hospital tranexamic acid vs placebo on 6-month functional neurologic outcomes in patients with moderate or severe traumatic brain injury. Journal of the American Medical Association 324(10):961-974.

Saatman, K., A-C. Duhaime, R. Bullock, A. Maas, A. Valadka, G. Manley, and Workshop Scientific Team and Advisory Panel Members. 2008. Classification of traumatic brain injury for targeted therapies. Journal of Neurotrauma 25:719-738.

Schmaltz, S. P., S. C. Williams, M. R. Chassin, J. M. Loeb, and R. M. Wachter. 2011. Hospital performance trends on national quality measures and the association with Joint Commission accreditation. Journal of Hospital Medicine 6(8):454-461.

Soto, J. M., Y. Zhang, J. H. Huang, and D. X. Feng. 2018. An overview of the American trauma system. Chinese Journal of Traumatology 21(2):77-79.

Sperry, J. L., F. X. Guyette, J. B. Brown, M. H. Yazer, D. J. Triulzi, B. J. Early-Young, P. W. Adams, B. J. Daley, R. S. Miller, B. G. Harbrecht, J. A. Claridge, H. A. Phelan, W. R. Witham, A. T. Putnam, T. M. Duane, L. H. Alarcon, C. W. Callaway, B. S. Zuckerbraun, M. D. Neal, M. R. Rosengart, R. M. Forsythe, T. R. Billiar, D. M. Yealy, A. B. Peitzman, M. S. Zenati, and PAMPer Study Group. 2018. Prehospital plasma during air medical transport in trauma patients at risk for hemorrhagic shock. New England Journal of Medicine 379(4):315-326.

Stein, D. M., R. A. Kozar, H. Livingston, F. Luchette, S. D. Adams, V. Agrawal, S. Arbabi, J. Ballou, R. Barraco, A. Bernard, W. Biffl, P. Bosarge, K. Brasel, Z. Cooper, P. Efron, S. Fakhry, C. Hartline, F. Hwang, B. Joseph, S. Kurek, F. Moore, A. Mosenthal, A. Pathak, M. Truitt, J. Yelon, and AAST Geriatric Trauma/ACS Committee. 2018. Geriatric traumatic brain injury—what we know and what we don’t. Journal of Trauma and Acute Care Surgery 85(4):788-798.

Tso, S., A. Saha, and M. D. Cusimano. 2021. The Traumatic Brain Injury Model Systems National Database: A review of published research. Neurotrauma Reports 2(1):149-164.

VA (Department of Veteran Affairs). 2019. Guideline for guidelines. Department of Defense. Revised January 29, 2019. https://www.healthquality.va.gov/documents/GuidelinesForGuidelinesRevised013019.pdf (accessed August 3, 2021).

Wagner, L. M., S. M. McDonald, and N. G. Castle. 2013. Impact of voluntary accreditation on short-stay rehabilitative measures in U.S. nursing homes. Rehabilitation Nursing 38(4):167-177.

Wakam, G. K., B. E. Biesterveld, M. P. Pai, M. T. Kemp, R. L. O’Connell, A. M. Williams, A. Srinivasan, K. Chtraklin, A. Z. Siddiqui, U. F. Bhatti, C. A. Vercruysse, and H. B. Alam. 2021. Administration of valproic acid in clinically approved dose improves neurologic recovery and decreases brain lesion size in swine subjected to hemorrhagic shock and traumatic brain injury. Journal of Trauma and Acute Care Surgery 90(2):346-352.

Wang, M. C., N. R. Temkin, R. A. Deyo, G. J. Jurkovich, J. Barber, and S. Dikmen. 2007. Timing of surgery after multisystem injury with traumatic brain injury: Effect on neuropsychological and functional outcome. Journal of Trauma: Injury, Infection, and Critical Care 62(5):1250-1258.

Winter, L., H. J. Moriarty, C. V. Piersol, T. Vause-Earland, K. Robinson, and B. Newhart. 2016. Patient- and family-identified problems of traumatic brain injury: Value and utility of a target outcome approach to identifying the worst problems. Journal of Patient-Centered Research and Reviews 3:30-39.

Zetterberg, H., D. H. Smith, and K. Blennow. 2013. Biomarkers of mild traumatic brain injury in cerebrospinal fluid and blood. Nature Reviews Neurology 9(4):201-210.