THE TRAUMATIC BRAIN INJURY MODEL SYSTEMS CENTER (TBIMSC) program was established by the National Institute on Disability and Rehabilitation Research (NIDRR) (now the National Institute on Disability, Independent Living, and Rehabilitation Research—NIDILRR) in 1987, with the goal of improving healthcare (especially rehabilitation care) and outcomes for patients with moderate-to-severe traumatic brain injury (TBI), providing patient/family and healthcare professional education, and conducting research. In a 2010 article, we described the origin, activities, and accomplishments of the TBIMSCs, with a particular emphasis on the research activities.1 This year's 30th anniversary of the TBIMSC program is an appropriate time to update the history and achievements. We focus on the research activities of the TBIMSC program, that is, the site-specific studies, collaborative studies, and longitudinal National Database (NDB), as well as on the multifaceted knowledge translation initiatives. We also provide examples of the many ways in which the infrastructure of the TBIMSC program has been leveraged in collaborations with other TBI stakeholders.
NIDILRR, the funder of much rehabilitation research in the United States, established the TBIMSC program to demonstrate the value of coordinated medical, social, and vocational services for persons with a moderate-to-severe TBI, a group that had started receiving inpatient rehabilitation facility (IRF) services in the 1970s, rather than being placed in psychiatric hospitals or other long-term care facilities.2 Grants supporting demonstration, education, and research, made for 5-year periods, initially went to 5 academic medical centers or equivalent entities; the number has been expanded over time, and for the 2017-2022 grant cycle, the number of recipients is 16 (see Table 1). The requirement of grantees that research activities are connected to a clinical program with emergency, acute neurosurgical, and inpatient and outpatient rehabilitation services is still key. The research funded involves contributions to the longitudinal NDB and collaboration on analysis of its data; site-specific research; and participation in module projects, which are shorter-term research projects undertaken by 2 or more TBIMSCs. The close cooperation between TBIMSCs on the NDB and module studies and the availability of extensive information on the rehabilitation patients at each center have been instrumental in the development of additional joint research, funded by NIDILRR, Patient-Centered Outcomes Research Institute (PCORI), National Institutes of Health (NIH), Department of Defense (DoD), Centers for Disease Control and Prevention (CDC), and other agencies. These mechanisms are described later, with recent and current projects listed.
Management of joint activities involves the project directors (PDs) (principal investigators) of all TBIMSCs and their staff, collaborating in a number of standing and ad hoc committees, which communicate in twice-yearly, face-to-face meetings in Washington, District of Columbia, through regular conference calls and dozens of listservs. NIDILRR staff participate in most meetings, offering advice or clarifying agency objectives, rules and procedures. The staff of the TBI National Data and Statistical Center (NDSC), which is separately funded by NIDILRR, similarly play key roles, participating in management of and providing support for all multicenter activities. The 5 TBIMSC program's standing committees are Executive, Planning, Research, Data, and Knowledge Translation (KT).
Special Interest Groups (SIGs) were first developed in 2008 to allow the TBIMSCs another avenue for research collaboration, focusing on developing new research efforts between centers as well as with outside entities. SIGs can be formed at any time, focus on mutual topics of interest to TBIMSC investigators, but do not necessarily conduct research, although research may result from their deliberations. SIGs are allotted time at PDs' meetings and must also hold regularly scheduled conference calls to ensure continued progress toward stated goals. Currently active SIGs include the following: (1) Aging with TBI and TBI in the Elderly; (2) Analytic Procedures; (3) Caregiver and Family; (4) Cultural Issues; (5) Disorders of Consciousness; (6) PCORI; (7) Sleep-Wake-Fatigue; (8) Geographic Identifiers for Data Linkages; and (9) Department of Veterans Affairs (DVA) Collaboration.
Move from the Department of Education to the Administration on Community Living
In October of 2014, NIDRR began the process of moving from the Department of Education (DoE) to the Administration on Community Living (ACL) within the Department of Health and Human Services (HHS). With the move to ACL came an increased focus on independent living, and “NIDRR” became “NIDILRR.” ACL combines the efforts of NIDILRR, the HHS Office on Disability, the Administration on Aging, and the Administration on Intellectual and Developmental Disabilities and serves as the agency responsible for increasing access to community supports, focusing on the needs of people with disabilities throughout their life span as well as those of older Americans; it also has oversight of the State Implementation and Protection and Advocacy TBI program grants.
National Data and Statistical Center
Since 2006, Craig Hospital in Englewood, Colorado, has been funded via 5-year competitive awards from NIDILRR to serve as the TBIMSC NDSC. Major initiatives over the years have included a public and private Web site to describe the NDB and facilitate the work of the TBIMSCs; a Standard Operating Procedures (SOP) Manual with a Web-based template for all TBIMSC policies and procedures; a standardized follow-up interview, built into the data entry system; mechanisms to support defunded TBIMSCs to continue data collection with those participants already enrolled in the NDB; data collector certification processes; dynamic data summary reporting for each center and for the TBIMSCs as a whole; resources for improving cultural competency in TBI research; and the introduction of many advanced statistical methodologies to analyze the wealth of NDB longitudinal data. In the current NDSC funding cycle (2016-2021), major initiatives are the standardization of data curation and data sharing to support reproducible research, as well as collaboration with the NDSCs from the Spinal Cord Injury and the Burn Model Systems Center (MSC) programs to maximize standardization where possible and promote trauma injury research.
TBI Interagency Conferences
NIDILRR and the TBIMSCs have a major role in the hosting of the TBI Federal Interagency Conferences through the efforts of the KT Committee. To date, 4 have been held in the Washington, District of Columbia, area (December 1999, March 2006, June 2011, and June 2018). The most recent one had participation from more than 30 federal agencies and institutions, including NIDILRR, CDC, DoD, DoE, DVA, Health Resources & Services Administration, and NIH. The interdisciplinary conference offers an opportunity for federal policy and research administration staff and the researchers they fund to learn about cutting-edge research and emerging evidence-based practices.
Collaboration with the DVA Polytrauma Rehabilitation Centers
In 2005, the DVA established Polytrauma Rehabilitation Centers (PRCs), which focused on TBI after this became the signature injury of the Middle-East wars. In 2008, NIDILRR and the DVA signed an interagency agreement to create a database for the PRCs, which parallels the TBIMSC NDB. The NDSC created a separate but similar Web-based data management system and provides the same training, technical, and administrative support, SOP development, and data access as is afforded to the TBIMSCs. Four PRCs began enrollment in 2010: James A. Haley Veterans Hospital, Tampa, Florida (482 participants enrolled); Minneapolis VA Medical Center (138); Hunter Holmes McGuire VA Medical Center, Richmond, Virginia (213); and VA Palo Alto Health Care System, California (165). In 2014, the South Texas Veterans Health Care System joined this initiative; it has enrolled 84 participants. The active military and veteran participants are being followed at the same postinjury time points as the TBIMSC NDB (see later).
The collaboration between the DVA PRCs and the TBIMSC program has grown well beyond the parallel databases. The DVA PRCs have representation, including voting privileges, at the biannual TBIMSC meetings, and on TBIMSC committees and SIGs, including the DVA Collaboration SIG specifically designed as a mechanism to bring TBIMSC and DVA researchers together. In this collaboration, the VA PRCs are now referred to as the VA TBIMSCs, although funding for these centers remains separate from that for the (NIDILRR) TBIMSCs. To date, DVA researchers have initiated 27 analyses of the VA PRC database, with 8 already published; 3 analyses comparing the TBIMSC NDB with PRC NDB data have been completed or are under way.3–10
THE TBIMSC NATIONAL DATABASE
Currently, the TBIMSC NDB is housed in a Web-based data management system. This system includes online data entry with a fully integrated data dictionary, an online structured interview for follow-up data collection, several types of data quality checks, and several types of data reports to support best practices in data quality. Data quality is further supported by on-site training for newly funded TBIMSCs, an online training manual, comprehensive quality support visits performed by the NDSC, quarterly data collectors' teleconferences, and in-person data collector training every 2.5 years. The NDB data dictionary (referred to as the Syllabus) is available on the NDSC Web site.11
Inclusion and exclusion criteria for enrollment in the NDB remain largely unchanged from the description provided in Dijkers et al1 (see Table 2). Data continue to be collected using the same 2 forms: Form I includes data on preinjury history, injury, and acute and rehabilitation care collected at study enrollment, whereas Form II is completed at the anniversary of injury at 1, 2, 5, and 10 years, and every 5 years thereafter, covering demographic, health, functional, social, and quality-of-life data. Changes in variables collected are allowed twice in each 5-year grant cycle. Committees, SIGs, and/or module groups can submit proposals for adding or deleting variables; the Planning Committee oversees the process of discussing and voting on proposals. Piloting of new variables is required, and surveys of TBIMSCs are completed to identify questions and concerns, with the group originally proposing the change responding to any issues raised. Ultimately, members of the Planning Committee (ie, the TBIMSC PDs) decide on additions and deletions. Supplementary Digital Content (SDC) Tables 1 and 2 (available at: http://links.lww.com/JHTR/A279 and http://links.lww.com/JHTR/A280, respectively) summarize the history of NDB variables collected over the years, including the recent addition of health-related variables in consideration of the view that TBI is a chronic health condition impacting the central nervous system and other organ systems.12 Notes detailing changes in the definition or codes for variables are included in the online syllabus.11
The NDB has data quality targets for enrollment (number projected for enrollment per each center's grant application; actual enrollment of ≥80% of those admitted to the IRF and eligible for enrollment) and follow-up (≥90% successful follow-ups at years 1 and 2; ≥80% successful follow-ups at all other follow-up time points). The NDSC assesses enrollment and follow up target achievement quarterly, disseminates all-centers data quality reports with a detailed report to each TBIMSC and its NIDILRR Project Officer. When any data quality targets are missed for 2 consecutive quarters, a written improvement plan is developed and monitored. Retention of participants from minority backgrounds is especially monitored via an NDSC-generated quarterly report. During 2017, across all centers, 81% of eligible IRF patients were enrolled and enrollment was 15% over anticipated numbers. Successful follow-ups were the following: year 1: 91%; year 2: 92%; year 5: 89%; year 10: 84%; year 15: 80%; year 20: 80%; year 25: 82%. Enrollment percentages and follow-up percentages were lower before the various standards were developed.
Data quality is also assessed in terms of missing data. The Missing Data Report shows the prevalence of missing data by variable, and the Case Completeness Report shows the percentage of all required data that is actually obtained for each case. The 2 reports allow TBIMSCs to identify any problematic variables or trends. The most recent Case Completeness Report indicates that at follow-up 96% of variables are collected from the average NDB participant.
While there has been some turnover, the majority of centers were refunded in the latest grant competition, resulting in personnel stability and consistent follow-up. Table 1 indicates the number of cases entered into the TBIMSC NDB by each center through 2017, and Table 3 indicates the number of follow-ups completed by rehabilitation discharge year and follow-up year through 2017.
DATABASE ANALYSIS: INTERNAL AND EXTERNAL NOTIFICATIONS
All investigators who contribute to the NDB have the right to analyze the data; however, to prevent duplication of effort and encourage collaboration, a process was established in which researchers circulate a 1-page proposal to solicit coinvestigators from other centers who have relevant expertise. In June 2018, the cumulative list of notifications contained more than 200 projects. These database analyses are the major vehicles through which NDB analysis takes place, and these efforts have resulted in most of the conference presentations and publications disseminating NDB information. An incomplete list of NDB publications counts more than 180 entries, with on average of more than 10 being published during each of the last dozen years. The notification and publication list using the NDB will soon be available on the public side of the NDSC Web site.
Mitchell Rosenthal, PhD, pushed for the establishment of the TBIMSCs as a board member of the National Head Injury Foundation (the original name of the Brain Injury Association of America). He later was a TBIMSC investigator and PD for the NDSC. After his untimely death, the TBIMSCs created an annual award in his name for the best article that offers an analysis of some component of the NDB. Table 4 lists the winners. As such, this list of titles gives a good indication of the variety of topics that can be addressed with NDB contents.
The NDB is available to qualified researchers who are not part of the TBIMSC program. Details are presented on the NDSC Web site.13 To obtain data, a researcher writes a 2- to 3-page proposal outlining the research plan, which is reviewed by NDSC staff and Research Committee members for the availability of the required data, feasibility of the analyses, and minimal qualifications of the requesters. It is then reviewed by all subscribing to the main TBIMSC listserv for possible duplication of existing, approved internal requests. If nothing stands in the way, the proposer works with NDSC staff to have a de-identified data set created. Non-TBIMSC users have the option to have an experienced TBIMSC investigator assigned as a mentor on understanding the data collection process and resulting data. To date, 17 requests for data from external researchers have been fulfilled.
A second NDB data distribution channel is through the Federal Interagency Traumatic Brain Injury Research (FITBIR) informatics system.14 In October 2017, the NDSC uploaded the TBIMSC NDB and it will continue to do so annually. With FITBIR access privileges, researchers can retrieve the data and, based on the ongoing collection of the Global Unique Identifier (GUID), merge NDB data with those they may have collected on the same individuals in their own studies.
In recent years, both the US government and the research community at large have pushed the adoption of policies for data sharing (“open data”). Besides providing data access via the 2 mechanisms described earlier, the NDSC has plans to adopt the best practices for open access to data recommended by ACL15 and also plans to develop an interactive version of the NDB available on the public NDSC Web site.
STUDIES TO ASSESS AND EXPAND NDB INFORMATION
We describe here recent efforts to enrich the NDB and endeavors to assess the quality and generalizability of the data collected.
As described earlier, for the past 2 decades, new variables were not added to the NDB unless they were successfully piloted by 1 or more TBIMSCs. NDB legacy variables (see SDC Tables 1 and 2, available at: http://links.lww.com/JHTR/A279 and http://links.lww.com/JHTR/A280, respectively) never received that attention, and even variables scrutinized on a limited basis might be less than reliable in broader use. Research by 6 TBIMSCs investigated test-retest reliability of all variables that were part of the Form II interview as of October 2013.26 A total of 224 English-speaking persons with TBI (proxies were excluded) were telephone-interviewed again, between 14 and 28 days after their routine follow-up interview, with all 66 measures (either single-item or index) readministered. The authors found that except for 4, all variables had good to excellent test-retest reliability using accepted benchmarks, with values equivalent to or sometimes even exceeding those of studies that used non-TBI general population samples. This research shows that reliable self-report data can be obtained from persons who incurred a moderate-severe TBI 1 to 20 years previously.
Because there are limits as to how frequently Form II follow-up interviews can be done, and how much information can be obtained in a single interview session, TBIMSC investigators have looked for alternative ways to augment the NDB. Kesinger et al27 developed an approach to probability match de-identified NBD data from one TBIMSC to de-identified data in the hospital system's trauma registry and demonstrated that this method of record linkage is feasible. This shows that the rich trauma registry data on acute treatments and complications can be joined to TBIMSC data on rehabilitation treatments and postrehabilitation life course, as confirmed in a second study.28
In the same vein, Corrigan and Bogner29 used patient address information for one TBIMSC to link NDB information to the rich data on patients' environment that is contained in US Census Bureau block group descriptors of local area economic and social characteristics. They showed that economic and social characteristics of a person's neighborhood explained variance in outcomes such as satisfaction with life and overall disability, beyond what could be accounted for by individual characteristics represented in the database.29 A module project with 12 TBIMSCs later collected and linked address data with census data.30 Ultimately, current address was added to the NDB so that the rich variety of information at the neighborhood, city, and state level is available for analyses.
Under an interagency agreement between CDC and NIDILRR and with CDC funding, several studies were launched to investigate the representativeness of the TBIMSC NDB. Corrigan et al31 obtained data from eRehabData and Uniform Data System for Medical Rehabilitation (UDSMR) on all US IRF patients 16 years or older admitted for TBI between October 1, 2001, and December 31, 2007, and compared their characteristics with those of TBIMSC NDB admissions in the same period. The major difference noted for about 20 variables compared was in age: the NDB cohort did not include as many patients older than 65 years. Differences in age-related characteristics (employment status etc) were also observed; however, if patients were divided into those younger than 65 years versus 65 years and older, the subgroups had minimal differences, especially within the younger than 65 years subset. The authors concluded that the NDB sample is largely representative of all US TBI rehabilitation inpatients, provided the age division is applied.
A follow-on study by Cuthbert et al32 that extended the admission window to December 31, 2010, found the same results. The authors recommended that when using the NDB data for population-based research, a statistical adjustment be made to account for the lower percentage of patients older than 65 years, inpatient rehabilitation stays less than 10 days, and preinjury vocational status.
Using an advanced statistical method called iterative proportional fitting (raking), the combined UDSMR and eRehabData population admission characteristics have been used to weight the NDB to represent the larger population of those who receive IRF rehabilitation for TBI in the United States. This weighted version of the NDB has been used to perform analyses to estimate outcomes in the larger population, with several projects initiated and 5 articles, an outcomes book, and a policy fact sheet published to date.33–39
TBIMSC grant recipients are required to conduct a site-specific study, which is peer reviewed as part of their grant applications. During the 2012-2017 cycle, each center could conduct 1 or 2 such studies; for the 2017-2022 cycle, only one was allowed. SDC Table 3 (available at: http://links.lww.com/JHTR/A281) provides the title and nature of the 40 projects funded. Clearly, there is a great variety of topics, with rehabilitation intervention studies for various TBI sequelae the most popular. While randomized controlled trials are the most often used design, other designs are used for both intervention and other research.
TBIMSC grant recipients must also participate in at least one module research project, a short-term (5 years or less) investigation of a TBI-related topic conducted jointly by at least 2 TBIMSCs. For the current and previous cycle, SDC Table 4 (available at: http://links.lww.com/JHTR/A282) provides the titles of the various projects, as well as their design and the participating centers. As is shown, all centers chose to take part in at least 2 modules and some in more than 5 modules. While a few randomized controlled trials are listed, most of these projects are prospective observational studies, with NDB participants as subjects and in some cases their caregivers, as well as individuals with TBI who are not part of the NDB sample.
COLLABORATIVE RESEARCH OUTSIDE THE TBIMSC FUNDING MECHANISM
With their similar clinical settings, access to large numbers of participants with TBI, and close cooperation in managing the NDB and module studies, TBIMSCs have successfully applied for funding of a number of other joint studies. Since 1998, NIDILRR has funded collaborative research projects in TBI that explicitly leverage the capacity of the TBIMSCs. These large-scale 5-year projects require participation from at least 3 TBIMSCs with the goal of contributing to evidence-based rehabilitation interventions. Projects funded since 2010 are listed in Table 5.
Another way the TBIMSCs leverage their capacity is through research proposal applications to PCORI, NIH, and the Congressionally Directed Medical Research Programs. Table 5 lists 5 such projects. The TBIMSCs continue to collaborate, with grant applications under review in 2018 by DoD, NIH, and NIDILRR.
DISSEMINATION AND KNOWLEDGE TRANSLATION
For dissemination and KT, the TBIMSCs individually and collectively work closely with the Model Systems Knowledge Translation Center (MSKTC), which was first separately funded by NIDILRR in 2006. The MSKTC advances a KT paradigm among MSC grantees (TBI, Spinal Cord Injury, and Burn) to ensure their research is relevant and accessible to people with disabilities and their families, researchers, clinicians, policy makers, and advocates. The MSKTC's aims include:
- Enhancing understanding of rehabilitation research;
- Increasing awareness and use of MSC research findings by appropriate stakeholders;
- Centralizing MSC resources for effective and uniform provision of training, technical assistance, and dissemination; and
- Increasing capacity of MSC grantees to engage in KT activities.
The MSKTC also:
- Conducts research on effective KT methods to increase awareness and use of MSC research;
- Develops research-based, user-friendly products grounded in KT science;
- Conducts KT training and technical assistance activities to increase KT capacities of TBIMSCs; and
- Disseminates information to stakeholders through its Web site,40 social media channels, conference exhibits, and promotional materials.
The resources distributed include slide shows, plain language summaries of model systems research, resource packages, press releases, and a policy factsheet. Other factsheets are written by model systems experts but user-tested by means of cognitive interviewing and given graphic design by the MSKTC. Some of these factsheets have received an extended life as patient information handouts printed in the Archives of Physical Medicine and Rehabilitation. The latest TBIMSCs foray into consumer-friendly information dissemination are infocomics,41 expert-vetted information packaged in a comic book format.
The public side of the NDSC Web site42 contains a wealth of information on the NDB, including a PowerPoint presentation with a descriptive summary of data. The Researchers page includes information on how to access the NDB and webinars on statistical strategies for analyzing data, including longitudinal analyses such as Individual Growth Curve modeling, which can be used to develop interactive tools to estimate the long-term outcomes in the years after TBI.
Since 1987, the TBIMSCs have worked to collect data relevant to all stakeholders, using up-to-date, user-friendly, and cost-efficient technology. Major changes in content (see SDC Tables 1 and 2, available at: http://links.lww.com/JHTR/A279 and http://links.lww.com/JHTR/A280, respectively), data collection, processing, and storage have taken place over time. Following is a discussion of current issues likely to result in major changes regarding what and how data are collected for the NDB in the near future.
Outcomes after moderate-severe TBI are determined by the exact nature of the injury, quality and quantity of acute medical and rehabilitative services received, patients' support system and environment, and their preinjury physical, cognitive, and emotional makeup—the phenome. The TBIMSC researchers have not been very successful in explaining who has particular outcomes and why. Genetic differences in genome and exome between subjects may explain many interindividual variances in outcomes in the physical/motor, cognitive, emotional, and other realms, as well as community reintegration and quality of life.43 A group of TBIMSC investigators is interested in exploring the link between genetic factors and outcomes, possibly leading to future genetic risk assessments of individual patients as well as personalized interventions.
When the TBIMSC program began, almost all IRFs in the United States submitted patient data to the UDSMR, which then created program evaluation reports including regional and national benchmarks. A key component was the 13 Motor and 5 Cognitive-Communicative items of the Functional Independence Measure (FIM) at admission and discharge. Because the FIM offered a strong measure of functional status, and collection of data was performed by clinicians at no cost to the TBIMSC grant, the FIM was quickly incorporated into the NDB; only interviews to capture self-report FIM at follow-up were performed with NIDILRR grant funds. With the 2001 introduction of the Prospective Payment System for IRFs, the Centers for Medicare & Medicaid services (CMS) required submission of data that included admission and discharge FIM, further solidifying its universal use. However, in an effort to collect similar data for all post–acute rehabilitation providers (IRFs, nursing homes, home care agencies), the CMS is phasing out the FIM and will instead be using functional assessment items from the Continuity Assessment Report and Evaluation (CARE) Item Set. Since October 2016, both FIM and CARE have been collected by IRFs on all patients. Effective October 2019, the FIM will be dropped by the CMS, creating a dilemma for the TBIMSCs. If TBIMSC investigators desire continuity within the NDB, grant funds must support the expensive collection of inpatient FIM data. One possible alternative, at least for the Motor FIM component, is the creation of a crosswalk between CARE and FIM items—an algorithm that would allow CARE information to be translated into FIM information. A module project for 2017-2022 is assessing this as a potential solution.
The data collection load, on participants interviewed (for Form II) for an hour or more and on TBIMSCs (the longest-existing centers now have cumulated cases over 25 years), has been a major concern of TBIMSCs. Extensive discussions precede changes to the NDB, as the PDs weigh the value of existing variables and of proposed new ones, in light of new areas of research that become feasible and the possible loss of information addressing old questions. The loss of data continuity when centers lose funding is another issue, addressed to some degree by having them serve as follow-up centers—which may not be very cost-effective. A task force of TBIMSC researchers is exploring options that could reduce the burden on participants and centers, including:
- Sampling patients for NDB inclusion, rather than requiring centers with a large caseload to enroll all their eligible patients;
- Reducing variables that provide duplicate information;
- Reducing frequency of collection of variables that are stable within individuals over time;
- Using online surveys to collect Form II information for subgroups of participants; and
- Centralized data collection by NDSC staff for participants of centers that lose funding.
Although TBIMSC researchers played a prominent role in development of the Common Data Elements (CDEs),44–46 some of the core or other CDEs are not part of the NDB. A number of NDB variables could be recoded to have the CDE format. For others, changes in definitions, categories, or other aspects would be required. As noted elsewhere, issues of continuity are important. The NDB is unique in that longitudinal data have been collected for nearly 30 years using a fairly stable set of core variables, with minimal changes in their coding over time. Even minor changes to these variables can complicate longitudinal analyses and may make some impossible. The TBIMSC investigators realize that in the era of open access to research findings and to data, 2 concepts endorsed by NIDILRR, using formats that are commonly accepted in the TBI research community is important. Discussions are ongoing regarding how many of the CDEs to introduce and when, while maintaining the longitudinal integrity of the TBI NDB.
The TBIMSC program was first funded in 1987, and NDB data collection began in 1989. Over the years, 23 centers have contributed, entering more than 16 000 participants into the NDB, the first of whom have completed their 25th anniversary follow-up. The NDB is the largest research database for TBI in the world. Single-site and module research performed by grantees has expanded our knowledge and constituted the basis for hundreds of professional papers and an even larger number of research-based presentations for individuals with TBI and their families, clinicians, researchers, and policy makers. The TBIMSCs have made significant research contributions to advance the field of TBI care and outcomes, and the TBIMSC researchers are working to keep the NDB and other research relevant to the needs of all TBI stakeholders.
Detailing all accomplishments of the TBIMSC program is beyond the scope of this article. Instead, we briefly mentioned themes of research to illustrate the breadth of the program in advancing the science. As discussed throughout this article, the scope of research topics is broad, including, but not limited to, the development of new measures for assessment of long-term outcomes in health and function, participation, employment, and quality of life; the development of manualized treatments for comorbidities such as fatigue/sleep, emotion dysregulation, memory, and other cognitive functions; and the study of trajectories of change in function. Research over at least the last 2 cycles of TBIMSC funding has aided in the establishment of moderate and severe TBIs as a chronic health condition.
The quality and impact of TBIMSC research have been recognized by Congress, for example, through its mandate that the DVA collaborate with NIDILRR TBIMSC grantees in the creation of a DVA TBI registry.47 Similarly, because of the unique role of the TBIMSCs in the larger field of federally funded TBI research, NIDILRR (representing DoE) became a mandated partner with DoD, DVA, and HHS in the creation and continuing implementation of the National Research Action Plan for TBI.48 , 49
1. Dijkers MP, Harrison-Felix C, Marwitz JH. The Traumatic Brain Injury Model Systems: history and contributions to clinical service and research. J Head Trauma Rehabil. 2010;25(2):81–91.
2. Thomas JP. The evolution of model systems of care in traumatic brain injury. J Head Trauma Rehabil. 1988;3(4):1–5.
3. Farrell-Carnahan L, Barnett S, Lamberty G, et al Insomnia symptoms and behavioural health symptoms in veterans 1 year after traumatic brain injury. Brain Inj. 2015;29(12):1400–1408.
4. Lamberty GJ, Nakase-Richardson R, Farrell-Carnahan L, et al Development of a Traumatic Brain Injury Model System within the Department of Veterans Affairs Polytrauma System of Care. J Head Trauma Rehabil. 2014;29(3):E1–E7.
5. Nakase-Richardson R, Stevens LF, Tang X, et al Comparison of the VA and NIDILRR TBI Model System cohorts. J Head Trauma Rehabil. 2017;32(4):221–233.
6. Dillahunt-Aspillaga C, Nakase-Richardson R, Hart T, et al Predictors of employment outcomes in veterans with traumatic brain injury: a VA Traumatic Brain Injury Model Systems study. J Head Trauma Rehabil. 2017;32(4):271–282.
7. McGarity S, Barnett SD, Lamberty G, et al Community reintegration problems among veterans and active duty service members with traumatic brain injury. J Head Trauma Rehabil. 2017;32(1):34–45.
8. Dreer LE, Tang X, Nakase-Richardson R, et al Suicide and traumatic brain injury: a review by clinical researchers from the National Institute for Disability and Independent Living Rehabilitation
Research (NIDILRR) and Veterans Health Administration Traumatic Brain Injury Model Systems. Curr Opin Psychol. 2018;22:73–78.
9. Stevens LF, Lapis Y, Tang X, et al Relationship stability after traumatic brain injury among veterans and service members: a VA TBI Model Systems study. J Head Trauma Rehabil. 2017;32(4):234–244.
10. Tran J, Hammond F, Dams-O'Connor K, et al Rehospitalization in the first year following veteran and service member TBI: a VA TBI Model Systems study. J Head Trauma Rehabil. 2017;32(4):264–270.
11. Traumatic Brain Injury National Data and Statistical Center. Traumatic Brain Injury Model Systems National Database syllabus. https://www.tbindsc.org/Syllabus.aspx
. Accessed June 15, 2018.
12. Masel BE, DeWitt DS. Traumatic brain injury: a disease process, not an event. J Neurotrauma. 2010;27(8):1529–1540.
13. Traumatic Brain Injury National Data and Statistical Center. Traumatic Brain Injury Model Systems National Database information for researchers. https://www.tbindsc.org/Researchers.aspx
. Accessed June 15, 2018.
14. Federal Interagency Traumatic Brain Injury Research (FITBIR). Federal Interagency Traumatic Brain Injury Research (FITBIR) informatics system. https://fitbir.nih.gov
. Accessed June 15, 2018.
16. Walker WC, Marwitz JH, Kreutzer JS, Hart T, Novack TA. Occupational categories and return to work after traumatic brain injury: a multicenter study. Arch Phys Med Rehabil. 2006;87(12):1576–1582.
17. Kreutzer JS, Rapport LJ, Marwitz JH, et al Caregivers' well-being after traumatic brain injury: a multicenter prospective investigation. Arch Phys Med Rehabil. 2009;90(6):939–946.
18. Hart T, Sherer M, Temkin N, et al Participant-proxy agreement on objective and subjective aspects of societal participation following traumatic brain injury. J Head Trauma Rehabil. 2010;25(5):339–348.
19. Whiteneck GG, Dijkers MP, Heinemann AW, et al Development of the participation assessment with recombined tools-objective for use after traumatic brain injury. Arch Phys Med Rehabil. 2011;92(4):542–551.
20. Harrison-Felix C, Kreider SE, Arango-Lasprilla JC, et al Life expectancy following rehabilitation
: a NIDRR Traumatic Brain Injury Model Systems study. J Head Trauma Rehabil. 2012;27(6):E69–E80.
21. Whyte J, Nakase-Richardson R, Hammond FM, et al Functional outcomes in traumatic disorders of consciousness: 5-year outcomes from the National Institute on Disability and Rehabilitation
Research Traumatic Brain Injury Model Systems. Arch Phys Med Rehabil. 2013;94(10):1855–1860.
22. Corrigan JD, Cuthbert JP, Harrison-Felix C, et al US population estimates of health and social outcomes 5 years after rehabilitation
for traumatic brain injury. J Head Trauma Rehabil. 2014;29(6):E1–E9.
23. Dams-O'Connor K, Pretz C, Billah T, Hammond FM, Harrison-Felix C. Global outcome trajectories after TBI among survivors and nonsurvivors: a National Institute on Disability and Rehabilitation
Research Traumatic Brain Injury Model Systems Study. J Head Trauma Rehabil. 2015;30(4):E1–E10.
24. Ritter AC, Wagner AK, Szaflarski JP, et al Prognostic models for predicting posttraumatic seizures during acute hospitalization, and at 1 and 2 years following traumatic brain injury. Epilepsia. 2016;57(9):1503–1514.
25. Dams-O'Connor K, Mellick D, Dreer LE, et al Rehospitalization over 10 years among survivors of TBI: A National Institute on Disability, Independent Living, and Rehabilitation
Research Traumatic Brain Injury Model Systems Study. J Head Trauma Rehabil. 2017;32(3):147–157.
26. Bogner JA, Whiteneck GG, MacDonald J, et al Test-retest reliability of traumatic brain injury outcome measures: a Traumatic Brain Injury Model Systems study. J Head Trauma Rehabil. 2017;32(5):E1–E16.
27. Kesinger MR, Kumar RG, Ritter AC, Sperry JL, Wagner AK. Probabilistic matching approach to link deidentified data from a trauma registry and a Traumatic Brain Injury Model System Center. Am J Phys Med Rehabil. 2017;96(1):17–24.
28. Kesinger MR, Juengst SB, Bertisch H, et al Acute trauma factor associations with suicidality across the first 5 years after traumatic brain injury. Arch Phys Med Rehabil. 2016;97(8):1301–1308.
29. Corrigan JD, Bogner JA. Neighborhood characteristics and outcomes after traumatic brain injury. Arch Phys Med Rehabil. 2008;89(5):912–921.
30. Corrigan JD, Bogner J, Pretz C, et al Use of neighborhood characteristics to improve prediction of psychosocial outcomes: a Traumatic Brain Injury Model Systems investigation. Arch Phys Med Rehabil. 2012;93(8):1350.e2–1358.e2.
31. Corrigan JD, Cuthbert JP, Whiteneck GG, et al Representativeness of the Traumatic Brain Injury Model Systems National Database. J Head Trauma Rehabil. 2012;27(6):391–403.
32. Cuthbert JP, Corrigan JD, Whiteneck GG, et al Extension of the representativeness of the Traumatic Brain Injury Model Systems National Database: 2001 to 2010. J Head Trauma Rehabil. 2012;27(6):E15–E27.
33. Cuthbert JP, Harrison-Felix C, Corrigan JD, Bell JM, Haarbauer-Krupa JK, Miller AC. Unemployment in the United States after traumatic brain injury for working-age individuals: prevalence and associated factors 2 years postinjury. J Head Trauma Rehabil. 2015;30(3):160–174.
34. Cuthbert JP, Harrison-Felix C, Corrigan JD, et al Epidemiology of adults receiving acute inpatient rehabilitation
for a primary diagnosis of traumatic brain injury in the United States. J Head Trauma Rehabil. 2015;30(2):122–135.
35. Harrison-Felix C, Pretz C, Hammond FM, et al Life expectancy after inpatient rehabilitation
for traumatic brain injury in the United States. J Neurotrauma. 2015;32(23):1893–1901.
36. Corrigan JD, Cuthbert JP, Harrison-Felix C, et al US population estimates of health and social outcomes 5 years after rehabilitation
for traumatic brain injury. J Head Trauma Rehabil. 2014;29(6):E1–E9.
37. Kowalski RG, Haarbauer-Krupa JK, Bell JM, et al Acute ischemic stroke after moderate to severe traumatic brain injury: incidence and impact on outcome. Stroke. 2017;48(7):1802–1809.
40. Model Systems Knowledge Translation Center. Living with traumatic brain injury. http://www.msktc.org/tbi
. Accessed June 15, 2018.
42. Traumatic Brain Injury National Data and Statistical Center. National Data and Statistical Center Traumatic Brain Injury Model Systems. https://www.tbindsc.org
. Accessed June 15, 2018.
43. McAllister TW. Genetic factors in traumatic brain injury. J Head Trauma Rehabil. 2015;128:723–739.
44. Hicks R, Giacino J, Harrison-Felix C, Manley G, Valadka A, Wilde EA. Progress in developing Common Data Elements for traumatic brain injury research: version two—the end of the beginning. J Neurotrauma. 2013;30(22):1852–1861.
45. Maas AI, Harrison-Felix CL, Menon D, et al Standardizing data collection in traumatic brain injury. J Neurotrauma. 2011;28(2):177–187.