In cases where an individual has sustained multiple cognitive or behavioral impairments, as is often the case with traumatic brain injury (TBI), a comprehensive treatment program may be ideal. In comprehensive cognitive rehabilitation therapy (CRT) programs (also called multi-modal or holistic), a team of therapists and other rehabilitation providers work together to ensure the most appropriate timing, delivery, and content of therapy for an individual. These treatment programs may occur during inpatient stays, or extend through outpatient programs. In this chapter, the committee reviews the studies on multi-modal/comprehensive CRT, divided by phase of recovery. Controlled studies are divided by comparator arm within these sections, and the committee’s conclusions are included at the end of each section.
The committee identified and reviewed six randomized controlled trials (RCTs) of multi-modal or comprehensive (holistic) CRT (Cicerone et al. 2008; Ruff and Niemann 1990; Salazar et al. 2000; Tiersky et al. 2005; Vanderploeg et al. 2008; Zhu et al. 2007). These trials were heterogeneous. Only one trial targeted mild TBI; three focused on the subacute phase while the other three focused on the chronic phase of recovery. Four of the six RCTs had some form of CRT in both trial arms.
Eight additional studies wtraining with traumatically brain-injured clients. Brainere identified as nonrandomized parallel group controlled studies. Three of the eight included CRT in the comparator group. One study was in the subacute phase, seven were in the chronic
phase, two included both subacute and chronic patients, and one did not report the time since injury. None of the studies was identified as exclusively or predominantly enrolling mild TBI patients. Studies ranged in sample size from 36 to 205 and were equally split between inpatient and outpatient settings. Seven studies were pre-post, single group design without any comparison or control group. However, there was a broad range in the quality of the design, execution, and reporting of the studies. Table 11-1 (at the end of the chapter) presents a summary of all included studies in this review.
The committee reviewed three RCTs (Salazar et al. 2000; Vanderploeg et al. 2008; Zhu et al. 2007) of multi-modal/comprehensive CRT in patients in the subacute phase of moderate-severe TBI; one nonrandomized, parallel group study (Bowen et al. 1999) of multi-modal/comprehensive CRT included patients in the subacute phase of recovery from mild, moderate, and severe TBI. All four studies enrolled patients within 6 months of their injury. Most significantly, all three RCTs had some element of CRT in their comparator arms. Thus, the goal of these studies was to determine whether there was a benefit of one form or level of intensity of CRT relative to another, early after injury. These studies were not designed to assess efficacy relative to no treatment or relative to an inert or minimal control condition, such as a waitlist group. Table 11-2 presents all subacute phase studies by design and treatment comparator.
Comparator Group: Non-CRT Content
Bowen et al. (1999), a single, nonrandomized, parallel group study, included 104 patients in the subacute phase with TBI severity ranging from mild to severe. The aim of the study was to evaluate outcomes of services provided by a community-based, interdisciplinary team of specialists— clinical psychologist, occupational therapist, family support nurse—all
|Study||Design||No Treatment||Non- CRT||Other CRT|
|Salazar et al. 2000||RCT||×|
|Vanderploeg et al. 2008||RCT||×|
|Zhu et al. 2007||RCT||×|
|Bowen et al. 1999||Parallel||×|
supported by a clinical coordinator. Treatment took place either before discharge from an inpatient hospital stay (mean 5 days postinjury) or after discharge from an inpatient hospital stay (mean 37 days postinjury). Overall, the median contact time with team members was relatively small— fewer than 15 hours for the early group and fewer than 10 hours for the late group. A third group was offered no specialized interdisciplinary team services. All three arms continued to receive existing services or care as usual. Because of the nature of the program, individual-level randomization was deemed infeasible; randomization occurred by 3-month blocks of time and was rotated across the two hospital sites involved in the study. The study included assessment of a broad range of outcomes (e.g., social, cognitive, behavioral, employment, handicap, functional limitations) at 6 and 12 months postinjury. The extent of contact with different team members is well described in the study. There were problems with protocol compliance, in the form of crossovers from original group assignment, which may have been systematic. Using the significance of 0.01 in light of the multiple outcomes, and adjusting for coma duration and age (which differed across the groups), essentially there were no differences in assessed outcomes.
Comparator Group: Other CRT Content
Salazar et al. (2000)1 conducted an RCT involving 120 active-duty military personnel who had recovered sufficiently from a recent moderate-severe closed head injury (within 3 months of randomization) to participate in a cognitive rehabilitation program or safely return home with a care-giver. All were oriented and had a Rancho Los Amigos cognitive level 7. Most had headaches. About one-third of the participants were described as having aggressive behavior or major depression, although few were taking psychotropic medications. Participants were randomly assigned to a comprehensive, 8-week in-hospital cognitive rehabilitation program or, after receiving some inpatient memory training, were discharged to home for a program of education and counseling via weekly telephone calls from a psychiatric nurse. During the telephone calls, which were described as lasting 30 minutes, nurses inquired about the week’s events, offered support and advice in addressing problems, and checked on use of memory aids. Of the 67 participants assigned to the in-hospital program, 60 completed the program; 47 of the 53 assigned to the home program completed the trial. Six patients assigned to home rehabilitation required supplemental therapy. At 1 year posttreatment, more than 90 percent of the participants in both groups returned to work, the primary outcome measure (group difference
1 The committee reviewed Salazar et al. 2000, with Braverman et al. 1999, and Warden et al. 2000.
was 4 percent [95 percent confidence interval, 5 to 14 percent]). The proportion of participants between groups who were fit for duty was also not statistically different: 73 percent of the inpatient arm versus 66 percent of the home rehabilitation program. A range of neuropsychological tests, as well as behavior, social adjustment (belligerence, social irresponsibility, antisocial behavior, social withdrawal, and apathy), and mood measures did not differ across groups at 1 year, but only 32 of the intensive rehabilitation group and 28 of the home rehabilitation group had those assessments. The reasons for missing data were not reported. A post hoc subgroup analyzed the 75 study participants whose period of unconsciousness at the time of injury was more than 1 hour; 28 of 35 (80 percent) of the group randomized to the inpatient program and 23 of 40 (58 percent) of those randomized to the outpatient program were fit for duty at 1 year (p = 0.05).
Vanderploeg et al. (2008) conducted a comparative effectiveness study of patients enrolled in four U.S. Department of Veterans Affairs (VA) in-patient TBI rehabilitation programs. Both arms of the study were inpatient rehabilitation; participants received occupational therapy, physical therapy, speech therapy, TBI education, and social support for 2 hours per day. One arm also included 2 hours per day of cognitive-didactic CRT, while the other arm received 2 hours per day of functional-experiential CRT. CRT was given for up to 60 days (33 days was the mean). For both arms, the average quantity of inpatient interventions was 132 hours per patient. The study reported no difference in primary outcomes of independent living or employment, and no difference on any secondary outcome measures including the FIM, measures of mood and behavior, the Disability Rating Scale, or a self-rating of memory. In subgroup analyses, patients younger than age 30 had better school or work outcomes in the cognitive-didactic arm, while those with higher education and older than age 30 did better in the functional-experiential arm on that primary outcome.
Zhu et al. (2007) studied 68 TBI patients with the primary goal of determining whether a higher level of intensity of early inpatient rehabilitation that included CRT produced better outcomes than a lower intensity of the same intervention. Patients were a mean of 20 days postinjury. The intervention took place 4 hours per day, 5 days per week, for up to 6 months or until discharge, if rehabilitation goals were met. The intervention included social skills training, hearing and speech training, and physical therapy, with goals toward achieving independent living and integration into home and community. The comparator arm received the same content of intervention but at only 2 hours per day (versus 4). These investigators found that Functional Independence Measures (FIM) and Neurobehavioral Cognitive Status Examination (NCSE) scores were no different across the high- and low-intensity rehabilitation arms at 6 months, with substantial gains on average in both arms from enrollment to 6 months. However, the
maximum FIM was achieved by the third month in 47 percent of patients in the high-intensity arm compared to 19 percent of the low-intensity arm. This finding is statistically significant and suggests that early intensive in-patient rehabilitation including CRT may hasten recovery, with maintained long-term outcomes. There was no cost analysis so the value (i.e., health benefit relative to cost) is unknown. For example, it is unknown if earlier discharge translated to lower utilization costs.
The evidence is not informative for conclusions about the impact (efficacy) on patient-centered outcomes (quality of life, functional status) of multi-modal/comprehensive CRT in the subacute phase (Vanderploeg et al. 2008).
There is evidence not informative for conclusions about sustainment of treatment effects (through 6 months after treatment) of multi-modal/comprehensive CRT delivered in the subacute phase (Bowen et al. 1999; Salazar et al. 2000).
The evidence is not informative for conclusions about the impact (efficacy) on domain-specific psychometric measures of cognition or functioning of multi-modal/comprehensive CRT in the subacute phase (Zhu et al. 2007).
In summary, the committee identified and reviewed three RCTs of comprehensive or multi-modal CRT in the subacute phase (Salazar et al. 2000; Vanderploeg et al. 2008; Zhu et al. 2007), and one nonrandomized, parallel group study (Bowen et al. 1999). All three of the RCTs compared some form of CRT in all study arms and had no inert, waitlist, or usual care comparison. The nonrandomized, parallel group study included a usual services arm, but that study had challenges to validity due to the quasi-experimental design and crossover; furthermore, the contents of usual services were not reported. Because the three RCTs do not compare CRT to a group receiving non-CRT therapy or usual care, it is not possible to formulate conclusions about efficacy.
Subacute phase patients may not reflect the same patient pool as those who enter the chronic phase and need CRT. Salazar et al. (2000) appeared to have a ceiling effect because 90 percent or more of both treatment groups returned to work, the primary outcome. It is possible that since this study recruited subjects from the subacute phase, a nontrivial proportion might have improved substantially in the first year postinjury regardless
of intervention, and would not have been seeking or referred for CRT in the chronic phase. It is important to be clear that these subacute studies’ findings cannot be extrapolated to the population of TBI patients in the chronic phase.
The primary focus of the committee’s analysis was assessment of the evidence for efficacy. However, the three RCTs did provide information about two other questions:
1. Does CRT in the subacute phase affect rate of recovery? Two RCTs examined this question, but with conflicting results. One RCT (Zhu et al. 2007) found that more intensive rehabilitation led to earlier meeting of milestones for discharge (with outcomes at 6 months being no different). The other (Salazar et al. 2000) found no difference between inpatient and outpatient CRT for rate of readiness to return to duty at 1 year. From these two conflicting findings, it is inconclusive as to whether intensity of CRT in the subacute phase is associated with more rapid attainment of clinically meaningful outcomes.
2. Does CRT delivered in the inpatient versus outpatient setting affect recovery? One RCT (Salazar et al. 2000) showed no evidence of higher benefit to extending an inpatient, intensive, high-volume CRT program for 8 weeks compared to discharging to a less-intensive, outpatient follow-up program. All participants were eligible for discharge to the community at enrollment. A post hoc analysis suggested that those with severe TBI benefitted more from inpatient CRT.
The committee reviewed three RCTs (Cicerone et al. 2008; Ruff and Niemann 1990; Tiersky et al. 2005) of multi-modal/comprehensive CRT in patients in the chronic phase of TBI. One of the trials compared CRT to a similar volume of a non-CRT intervention (Ruff and Niemann 1990), and another to a waitlist control condition (Tiersky et al. 2005). Cicerone et al. (2008) compared one format of comprehensive CRT to another form of comprehensive CRT to assess relative or comparative effectiveness of alternate comprehensive approaches. Of six nonrandomized, parallel group design studies identified and described in this review of chronic phase TBI patients, three studies compared comprehensive CRT to a non-CRT program, and three were comparative effectiveness studies of alternate CRT approaches. Implications of study results are markedly different for studies that compare CRT to an inert comparison or to a non-CRT comparator group, as these studies provide knowledge about efficacy, versus the stud
|Study||Design||No Treatment||Non- CRT||Other CRT|
|Cicerone et al. 2008||RCT||X|
|Ruff and Niemann 1990||RCT||X|
|Tiersky et al. 2005||RCT||X|
|Chen et al. 1997||Parallel||X|
|Cicerone et al. 2004||Parallel||X|
|Goranson et al. 2003||Parallel||X|
|Middleton et al. 1991||Parallel||X|
|Parente and Stapleton 1999||Parallel||X|
|Sarajuuri et al. 2005||Parallel||X|
|Braunling-McMorrow et al. 2010||Pre-Post|
|Cicerone et al. 1996||Pre-Post|
|Huckans et al. 2010||Pre-Post|
|Klonoff et al. 2007, 2010||Pre-Post|
|Mills et al. 1992||Pre-Post|
|Murphy et al. 2006||Pre-Post|
|Rattock et al. 1992||Pre-Post|
|Walker et al. 2005||Pre-Post|
ies that compare alternative forms of CRT. The latter are comparative effectiveness studies, which do not yield knowledge about efficacy but instead show the relative impacts of the two different approaches. Thus, this section of this review is divided into two components: two RCTs (Ruff and Niemann 1990; Tiersky et al. 2005) and four nonrandomized, comparison group studies (Chen et al. 1997; Goranson et al. 2003; Parente and Stapleton 1999; Sarajuuri et al. 2005) that compare CRT to a non-CRT arm; and one RCT (Cicerone et al., 2008) and three nonrandomized, comparison studies (Cicerone et al. 2004; Middleton et al. 1991; Rattok et al. 1992) that compare two alternative forms of CRT. Table 11-3 presents all chronic phase studies by design and treatment comparator.
Comparator Group: Non-CRT Content
The committee reviewed one RCT of comprehensive CRT in patients with chronic TBI (Tiersky et al. 2005). A large majority of this small trial’s
participants (29 were randomized; 20 completed the trial) had mild TBI; all enrollees had to be at least 1 year postinjury (mean = 5 years). This study was a pilot trial of an outpatient intervention; no power calculations were reported. The intervention arm received about equal amounts of cognitive remediation (i.e., attention, information processing, memory) and individual cognitive behavioral therapy in two 50-minute sessions, 3 days per week over 11 weeks; the total intervention time is estimated at 55 hours. The comparator group was placed on a waitlist, and received two or three in-person meetings or phone calls with the principal investigator over the 11-week intervention period (2 or 3 hours total); no therapeutic activities were offered in these contacts. Outcomes were measured at 11 weeks, then at 1 and 3 months after treament. The primary outcome measures were the depression, anxiety, and general symptom indexes of the Symptom Checklist-90R, the PASAT (objective measure of attention), a coping measure, and a self-report measure of attention. There was a significant beneficial effect in favor of the intervention (p < 0.05) for the general symptom index, depression, anxiety, and the PASAT. Although the two groups did not differ statistically at baseline on a range of characteristics, the sample was small, and they were qualitatively different on several characteristics, for example, baseline General Symptom Index scores were 1.16 for treatment and 1.62 for controls (p = 0.19).
In another RCT, Ruff and Niemann (1990) studied 40 patients with severe TBI 1 year postinjury. This outpatient CRT intervention was 8 weeks long and took place 4 days per week, 5 hours per day (for a total of 160 hours). Sessions included 2 weeks each of CRT targeting attention, spatial integration, memory, and problem solving. Also encompassed within the 5 hours of daily rehabilitation programming was a 50-minute group psychotherapy session and 30 minutes of wrap-up. The comparator arm was also 160 hours of treatment in an outpatient setting over 8 weeks. The difference was in the content, as this program included computer/video games, sessions on coping skills, group and didactic sessions on healthy lifestyle, small group discussion forums, lectures and workbook exercises on independence, and art. The comparator arm similarly included 50 minutes daily of group psychotherapy and 30 minutes daily of wrap-up. Cognition was measured in all 40 patients; behavior and adjustment were measured in a subset of 24 patients. Findings showed no between-group differences on outcomes in nine of nine attention measures, five of five spatial measures, five of nine memory measures, and four of four problem-solving measures; performance IQ was also measured. Verbal IQ scores and scores on four of the nine memory measures were better in the CRT arm than the non-CRT comparator arm.
In the Saajuuri et al. (2005) nonrandomized, parallel group study, 19 patients with moderate-severe TBI received an inpatient program that
included both neuropsychological rehabilitation and psychotherapy. The program took 210 hours (7 hours per day, 5 days per week, for 6 weeks). To be included, participants had to be judged as independent in daily life and have “adequate potential to achieve productivity” with “special” rehabilitation. At one rehabilitation facility, 23 patients (three were lost to follow-up) were identified for a comparison group out of a series of 213 patients at a different facility, who had sustained head injuries during the same time frame as those receiving the CRT intervention program; all 23 were judged as meeting the same criteria for the intervention program. The control group received care as usual, including both clinical and rehabilitation care services. A mailed questionnaire 2 years after completing the program (for the intervention group) or a comparable interval (for the comparison group) asked about paid and unpaid work or current student status; 2 of 19 receiving the intervention compared to 9 of 20 of the usual care group were not engaged in any productive activity at follow-up (p = 0.017). When categorized by full-time paid employment, only 1 of the 19 intervention compared to 7 of the 20 usual care group met this benchmark.
Chen et al. (1997) enrolled 40 patients in a study that compared hierarchical computer-assisted cognitive rehabilitation delivered in an outpatient setting to “various other therapies including speech therapy and occupational therapy.” Twenty patients who had received the computer-assisted cognitive rehabilitation program and had undergone pre-post evaluations of neuropsychological function were drawn from a database at one center; 20 patients from three other centers who had received other services were drawn from those centers’ records. The study was small, and the intervention and comparison arm participants differed substantially on several key characteristics including time since injury and length of coma. There were no significant differences between groups in pre-post score changes.
In the Parente and Stapleton (1999) study, outcomes were assessed among 33 TBI patients who had been referred to a rehabilitation program and given a program that included cognitive skills group sessions, computer training, training in use of electronic aids such as tape recorders or personal organizers, interviewing skills training, and peer teaching. Average participation duration was 4 months. However, the analysis sample only included 13 patients who had completed the program at the time the outcome evaluation was conducted. The comparison group was 64 subjects pulled from a database of 568 brain-injured patients who received services during the same time frame; the actual amount and type of services received by these subjects were unknown. While 10 of the 13 (76 percent) who received the intervention program were employed compared to 58 percent of the comparison group, the number in the intervention program analysis is very small, the comparison group could have differed significantly from the intervention group, and what the intervention impact is being compared to
(in terms of content and extent of services that might have included CRT) is completely unknown.
Goranson et al. (2003) retrospectively identified 42 mild TBI patients from existing clinical files. These patients were described as a small group of TBI patients seen at that clinic over 4 years. The study required patients have returned for follow-up outcome data collection at 6 and 18 months after initial collection. The intervention group comprised 21 patients who met the rehabilitation institution’s criteria for an outpatient CRT program that targeted attention, memory, reasoning, and problem solving, as administered by providers from multiple disciplines. Treatment was provided for 4 days per week and 5.5 hours per day, for an average of 4 months (range of program duration was 1 to 7 months). Another 21 patients were identified for the comparison analysis, selected to provide a similar distribution on age, education, and gender to the intervention group. Of note, however, most of the patients in the comparator “no rehabilitation” group did not meet inclusion criteria for the CRT program and thus were different from the group that did receive the CRT program. The study sample was in the chronic phase of recovery for mild TBI, on average 12 to 13 months postinjury. Those who received the CRT program had better Community Integration Questionnaire (CIQ) scores on the Home Integration scale at follow-up, adjusting for differences in baseline scores. There were no differences across groups on the CIQ Social Integration or Productivity scores. Again, the study is small, the intervention and comparison groups were not comparable because the majority of the comparison group was ineligible for the CRT program, and the sample selected for the analysis may have been prone to substantial selection bias because it represented a small subset who, for reasons not described, returned to the facility for follow-up outcome measurement.
Comparator: Other CRT Content
In an RCT, Cicerone et al. (2008) compared two alternative approaches to outpatient comprehensive CRT. One group of 34 patients was randomized to receive an intensive outpatient cognitive rehabilitation program, with an emphasis on metacognition and emotional regulation. The program included 11 hours per week of cognitive, communication, and life skill groups plus individual therapy (4 hours per week), over 16 weeks, for a total of 240 hours of outpatient CRT. Another group of 34 study participants were randomized to a different outpatient comprehensive interdisciplinary day treatment of standard neurorehabilitation, which included retraining of discrete cognitive functions through individual therapy and individualized physical, occupational, and speech therapy, as well as counseling and some group sessions. Treatment also took place over 16 weeks, 15 hours per week, for a total of 240 hours of outpatient CRT. The study found that
intensive cognitive rehabilitation yielded better scores on measures of community integration, life satisfaction, and self-efficacy, compared to the standard neuro-rehabilitation arm; neuropsychological functioning improved in both arms, but did not differ across groups at follow-up.
Of the three nonrandomized, parallel group studies comparing alternative forms of CRT, Cicerone et al. (2004) enrolled 56 patients with TBI in a study that compared a 320-hour inpatient cognitive rehabilitation program that included individual and group cognitive remediation (4 days per week, 5 hours per day, 16 weeks) to a 288-hour standard inpatient neuro-rehabilitation program of physical, occupational, and neuropsychological therapies that “incorporated many of the principles of comprehensive neuro-psychological rehabilitation” but in a less structured, less intense fashion. The intensive CRT treatment arm had significantly better Community Integration Questionnaire scores after program completion, despite being in the chronic phase (mean = 34 months from injury) compared to the less intensive CRT arm, which was in the subacute phase (approximately 5 months postinjury).
Middleton et al. (1991) compared outcomes of two alternative forms of computer-assisted neuropsychological educational treatment at 8 weeks. Both treatment programs had 96 hours of training on attention, concentration, perceptual skills, and problem-solving skills. Of the participants, 18 received an additional 32 hours of computer-assisted attention and memory training, and 18 other participants received instead 32 hours of computer-assisted reasoning and logical thinking training. There is neither a description of how participants were allocated into each group, nor of the process for their selection out of eligible participants. Both groups had statistically significant gains in five of six neuropsychological test measures, but there were no between-group differences at follow-up.
Rattok et al. (1992) enrolled 59 patients with TBI in three different arms; all arms received 140–160 hours of attention training, community activities, and counseling. In addition, one arm received 220 hours of cognitive remediation and small-group interpersonal exercises, one arm received 200 hours of small group interpersonal exercises but no cognitive remediation, and one arm received 200 hours of cognitive remediation but no small group interpersonal exercises. The process for assigning participants to study arms was not described. The 400 hours of CRT were delivered over 20 weeks in an outpatient setting. Among the many outcome measures, no patterns of between-group differences emerged.
The committee reviewed seven studies of a pre-post design without any comparison or control group (Braunling-McMorrow et al. 2010; Cicerone et al. 1996; Huckans et al. 2010; Klonoff et al. 2007, 2010; Mills et al.
1992; Murphy et al. 2006; Walker et al. 2005). Study participants ranged from having only mild TBI (Cicerone et al. 1996) to only severe TBI (Walker et al. 2005), or included mixed participants. Three studies (Cicerone et al. 1996; Huckans et al. 2010; Walker et al. 2005) had 25 or fewer subjects. Most of these studies examined predictors or covariates of outcomes. Outcomes were measured at 3 months (Walker et al. 2005), through 12 months (Braunling-McMorrow et al. 2010), and through 18 months (Mills et al. 1992) after program completion. Three studies had highly variable follow-up outcome assessment times depending on program completion: Murphy et al. (2006) reported vocational status at discharge from the program, ranging from 1 week to 4.5 years; Cicerone et al. (1996) reported outcomes assessed from 1 to 6 months after treatment; and Klonoff et al. (2007, 2010) reported outcome assessment times at program completion, ranging from 2.8 to 23.5 months.
There was substantial heterogeneity in the content and duration of these CRT programs. Braunling-McMorrow et al. (2010) evaluated a comprehensive, community-based residential rehabilitation program providing multi-faceted behavioral and CRT strategies delivered by a multi-disciplinary team. Murphy et al. (2006) evaluated a vocational rehabilitation-focused program that included intensive cognitive rehabilitation followed by placement of participants in actual work settings with a job coach. Klonoff et al. (2007, 2010) assessed work, school, and driving outcomes of a holistic, “milieu-oriented work/school re-entry program.” Walker et al.’s (2005) 9-month community-based program including social skills training revolving around a group fundraising program to support an outdoor adventure course activity, practice on the outdoor adventure course, and group meetings to foster individual goal attainment. Cicerone et al.’s (1996) program of neuropsychological and cognitive remediation included a wide range of cognitive domain modalities tailored to the individuals’ needs. Mills et al.’s (1992) tailored program “emphasized improvement of the patients’ real-life functional abilities and psychological support.” The program took place 6 hours daily, 5 days per week, for at least 6 weeks; it involved both patients and family or friends, if appropriate.
There is limited evidence about the impact (efficacy) on patient-centered outcomes of multi-modal/comprehensive CRT delivered to patients with mild TBI in the chronic phase of recovery. One small but well-conducted trial demonstrated meaningful beneficial effects on
patient-centered outcomes (general symptom index, depression, anxiety) (Tiersky et al. 2005).
There is limited evidence about the sustainment of treatment effects on the general symptom index through 3 months posttreatment of multi-modal/comprehensive CRT delivered to patients with mild TBI in the chronic phase of recovery (Tiersky et al. 2005).
There is limited evidence about the impact (efficacy) on psychometric measures of cognition of multi-modal/comprehensive CRT delivered to patients with mild TBI in the chronic phase of recovery (Tiersky et al. 2005).
The evidence is not informative about the impact (efficacy) on patient-centered outcomes (quality of life, functional status) of multi-modal/comprehensive CRT in patients with moderate-severe TBI in the chronic phase of recovery.
The evidence is not informative about the sustainment of treatment effects (through 6 months after CRT) of multi-modal/comprehensive CRT delivered to patients with moderate-severe TBI in the chronic phase of recovery.
The evidence is not informative about the impact (efficacy) on psychometric measures of cognition for multi-modal/comprehensive CRT in patients with moderate-severe TBI in the chronic phase of recovery.
The committee found a paucity of studies of efficacy of comprehensive CRT, and the two RCTs of efficacy that the committee identified were small and intended as pilot studies. The lack of large trials with an inert or waitlist comparison group is the primary reason for the conclusions. In brief, there were a total of three RCTs and six nonrandomized, parallel group design studies of comprehensive CRT identified in the review. However, one of the three RCTs and three of the nonrandomized, parallel group studies were comparative effectiveness studies of alternative approaches to CRT and did not address efficacy. These trials compared one or more extensive programs of CRT; the amount of services in these programs ranged from a minimum of 96 hours to a maximum of 400 hours across all arms including the control arms.
The two efficacy RCTs of comprehensive CRT were small pilot studies, had no power calculations, and targeted different groups of TBI patients.
One of the two RCTs (Tiersky et al. 2005) included patients with predominantly mild TBI in the chronic phase demonstrated meaningful beneficial effects; notably, it was the sole RCT with an inert comparator arm—patients in that arm were waitlisted for the program. Therefore, there is preliminary evidence that an 11-week outpatient program of about 55 hours of both CRT and cognitive behavioral therapy is beneficial in patients with mild TBI in the chronic phase. However, while showing favorable findings on several primary outcomes, the study was a pilot, exploratory trial; no larger, follow-on trials were identified in this literature review. The second efficacy trial (Ruff and Niemann 1990) found few differences across CRT and non-CRT arms in a population with moderate-severe TBI, the non-CRT program was intensive and certainly included services and elements that could have also had a beneficial effect on the outcomes studied. In addition, the non-CRT arm received 160 hours of services over 8 weeks, an amount against which the lack of evidence of large benefit of CRT in this study must be taken into account. Because the control group received a substantial amount of rehabilitation and social services, the ability to detect a difference on clinical outcomes between the CRT arm and the control arm may be reduced. This study’s findings were not judged as evidence against efficacy of comprehensive CRT. The three nonrandomized, parallel group studies that had at least one non-CRT comparison group were small and had considerable design limitations. These conditions preclude findings from those trials having much bearing on interpretation of this literature in weighing whether or not there is benefit from comprehensive CRT for patients with TBI in the chronic phase.
About half of the studies the committee identified on comprehensive CRT did not answer questions about efficacy but rather compared one or more extensive programs of CRT; the amount of services in these programs ranged from a minimum of 96 hours to a maximum of 400 hours across all arms. Comparative effectiveness studies of comprehensive CRT may be premature without preceding efficacy trials of the interventions applied in each arm. Furthermore, without assessment of utilization and cost, the relative value (extent of health benefit relative to cost) of the programs being compared in these studies cannot be determined.
|Study||N||TBI Severity Level||Brief Narrative||Comparator||Outcome Measures||Findings|
|Cicerone et al. 2008||68||Moderate-Severe||Compared one format of comprehensive neuropsychologic (NP) CRT to another form of standard, multidisciplinary CRT to assess relative or comparative effectiveness of alternate comprehensive approaches.||Y
Other CRT Content: Standard, multidisciplinary neuro- rehabilitation
• Primary outcomes:
■ CIQ (community integration)
■ PQOL (life satisfaction)
• Secondary outcomes:
■ NP functioning
• Trail Making Test
• Controlled Oral Word Association Test
• Booklet Category Test
• California Verbal Learning Test-II
• Rey Complex Figure
■ Perceived self-efficacy
■ Community-based employment
• Vocational Integration Scale
|While both groups saw NP improvement, intensive CRT subjects, compared to the standard CRT group, saw greater improvement on the CIQ, PQOL, and self-efficacy for the management of symptoms. At 6-month follow-up, these gains were maintained. The standard CRT patients, however, showed increased productivity at 6 months, correlated with the need for continued remediation.|
|Ruff and Niemann 1990||24||NR||The effectiveness of cognitive remediation versus day treatment was compared in this study of emotional and psychosocial adjustment in subjects.||Y
Other CRT Content: Sessions emphasizing psychosocial adjustment, leisure, and activities of daily living
Katz Adjustment Scale (KAS) subscales:
• Social Obstreperousness
• Acute Psychoticism
• Withdrawn Depression
|There was no significant difference between groups, with both groups reporting decreased symptoms of psycho-emotional distress, i.e., lessened social withdrawal and depression.|
|Salazar et al. 2000
Braveman et al. 1999
Warden et al. 2000
|120||Moderate-Severe||Comparing in-hospital to at-home rehabilitation services for 120 active-duty military personnel who had recovered sufficiently from a recent moderate-severe closed head injury (within 3 months of randomization) to participate in a cognitive rehabilitation program.||Y
Other CRT Content: Home-based rehabilitation with TBI education and individual counseling from a psychiatric nurse, educational materials, and recommended strategies for enhancing cognitive and organizational skills
• Auditory Consonant Trigrams
• Buschke Selective Reminding Test
• Halstead-Reitan Neuropsychological Impairment Index
• Katz Adjustment Scale subscores
• Return to gainful employment and fitness for military duty at 1-year follow-up
• Trahan Continuous Visual Memory Test
• Wechsler Memory Scale Revised
• Wisconsin Card Sorting
|At 1-year follow-up, there was no significant difference between the in-hospital CRT program and home-based CRT program groups in return to gainful employment or fitness for duty; furthermore, there were no significant differences in cognitive, behavioral, or quality-of-life measures. A post hoc subset analysis showed that, in patients who were unconscious for more than an hour, the in-hospital group had a greater return-to-duty rate.|
|Tiersky et al. 2005||20||Mild, Moderate||Tested the efficacy of a comprehensive neuropsychologic rehabilitation program with cognitive remediation coupled with cognitive behavioral therapy sessions.||Y
No Content: Wait list
• Neuropsychologic functioning measures:
■ Assessment of Client Functioning Inventory (ACFI)
■ Attention Questionnaire
• Psychosocial/affective functioning measures:
■ Coping Response Inventory (CRI)
■ Symptom Checklist-90 Revised (SCL-90 R)
• Measure of community participation: CIQ
|The treatment group improved significantly in emotional functioning, which included reduced anxiety and depression, compared to the control; 1-month and 3-month follow-ups showed the greatest improvements. While performance on a divided auditory attention measure improved as well, no changes were observed in community integration scores.|
|Vanderploeg et al. 2008||360||Moderate- Severe||Evaluated two different cognitive rehabilitation approaches, cognitive didactic and functionalexperimental, to determine efficacy overall and relative effectiveness to specific subpopulations.||Y
Other CRT Content: Cognitive didactic or functionalexperimental
• Primary outcome measures at 1-year follow-up:
■ Functional independence in living
■ Return to work and/or school
• Secondary outcome measures:
■ Disability Rating Scale score
■ Neurobehavioral Rating Scale
|The study found no difference in primary outcomes of independent living or employment, and no difference on any secondary outcome measures including the FIM, measures of mood and behavior, the Disability Rating Scale, or a self-rating of memory. In subgroup analyses, patients younger than age 30 had better school or work outcomes in the cognitive-didactic arm, while those with higher education and older than age 30 did better in the functionalexperiential arm on that primary outcome.|
|Zhu et al. 2007||68||Moderate-severe||Evaluated the benefit of cognitive rehabilitation therapy, comparing varied intensity levels of training, on functional outcome.||Y
Other CRT Content: Fewer hours of rehabilitation, compared to the treatment group.
• Primary outcome measure: FIM
• Secondary outcome measures:
■ Glasgow Outcome Scale (GOS)
■ Neurobehavioural Cognitive Status Examination (NCSE)
|Though FIM and NCSE scores were no different across the high- and low-intensity rehabilitation arms at 6 months, with substantial gains on average in both arms from enrollment to six months, the maximum FIM was achieved by the third month in 47 percent of patients in the high-intensity, compared to 19 percent of the low-intensity arms. This finding is statistically significant, suggesting that early intensive inpatient rehabilitation including CRT may hasten recovery, with similar longer-term outcomes.|
|Bowen et al. 1999||104||Moderatesevere||Evaluated whether a community-based, interdisciplinary team (e.g., clinical psychologist, occupational therapist, and family support nurse) in addition to usualcare services was more beneficial for improving outcome than usual services alone. Based on a prespecified timetable, the study also aimed to determine whether early treatment was better than later intervention in 104 subjects assigned to a treatment group.||Y
Other CRT Content: Usual care, or noninterdisciplinary services (occupational therapy alone)
• Wimbledon Self-Report Scale
• Katz Adjustment Scale, modified form R1 (KAS-R1)
• Logical Memory 1
• Logical Memory 2
• Occupational status
• Functional Limitations Profile
• International Classification of Impairments, Disabilities and Handicaps (ICIDH)
■ Physical Independence
|After adjusting for potentially confounding factors, the investigators found a statistically insignificant superior outcome for both intervention groups, compared to the control group, in some but not all areas.|
|Chen et al. 1997||40||NR||Examined the effectiveness of computer-assisted cognitive rehabilitation (CACR) involving a hierarchical sequence of training steps beginning in fundamental functions and advancing to more complex cognitive processes.||Y
Other CRT Content: Control group received no or low doses of CACR, but did receive other therapies, like speech therapy and occupational therapy
• WAIS-R subtests:
■ Digit span, symbol
■ Picture completion
■ Picture arrangement
■ Block design
■ Object assembly
• Category test
• Trail Making Test
• Wisconsin Card Sorting test
• WMS and WMS-R
• Digit span forward, backward
• Logical memory immediate, delayed
• Visual reproduction immediate, delayed
• Paired associates immediate, delayed
|CACR made significant gains on the neuropsychological test on 15 measures, compared to seven by the comparison group. In contrast, the investigators found no significant differences between groups on posttreatment gains.|
|Cicerone et al. 2004||56||Mild, Moderate, Severe||This study compared the effectiveness of an intensive cognitive rehabilitation program (ICRP), a highly structured program integrating cognitive and psychosocial interventions.||Y
Other CRT Content: Standard neurorehabilitation including physical, occupational, and speech therapies, as well as neuropsychologic treatment
• Quality of Community Integration Questionnaire (QCIQ)
• Trail Making Test
• California Verbal Learning Test
• Rey Complex Figure
• Category Test
|Both groups improved significantly on the CIQ, with the ICRP participants twice as likely to show clinical benefits, compared to the SRP group. ICRP participants, as well as those with clinically significant improvement on the CIQ, showed significant overall improvement in neuropsychologic functioning.|
|Middleton et al. 1991||36||NR||Investigates outcomes following computerdelivered cognitive rehabilitation therapy, comparing two forms of computer-assisted neuropsychological treatment targeting attention and memory skills or reasoning and logical thinking skills.||Y
Other CRT Content: Control received computer-assisted training on reasoning and logical thinking skills
• WAIS-R, Digit Span
• Wechsler Paired Associates
• Knox’s cube
• Block Counting
• Concept formation
• Shipley Abstraction
|There was significant improvement on five of six measures by both groups at 8-week follow-up; no differential effect was observed in the treatment group.|
|Parente and Stapleton 1999||33||NR||In a pilot study, evaluated the effectiveness of the Cognitive Skills Group (CSG) providing thinking skill training as a precursor to vocational placement. The sequence of topics include problem solving, concentration/attention, decision making, remembering faces and names, study skills, functional mnemonics, prosthetic memory devices, social cognition, organizational skills, goal setting, non-verbal perception, specific study skills, and test-taking strategies.||Y
Other CRT Content: Similar services but not within CSG
|Return to work||Participants in the CSG had a 76 percent vocational rehabilitation rate, i.e., placement in competitive employment; 58 percent of the baseline participants returned to work.|
|Rattok et al. 1992||59||severe||Assessed three groups with varying treatment combinations of attention training, cognitive remediation, small-group interpersonal exercises, community activities, and personal counseling. Each group received a total of 400 hours of training. Cognitive remediation involved task and curing hierarchies, constructional praxis, visual information processing, and logical reasoning.||Y
Other CRT Content: Group 1 received cognitive remediation with small-group interpersonal exercises; Group 2 stressed smallgroup interpersonal exercises; Group 3 involved cognitive remediation without smallgroup time; all three groups received attention training.
• Orientation Remedial Module (ORM)
• Purdue Pegboard
• Visual processing skills
• Academic skills
• WAIS - verbal, performance
• Psychomotor dexterity
• Higher order and conceptual skills
|All three treatment mixes were equally effective in all areas, with some mixes producing superior results to others in some respects (e.g., with regard to intra- and interpersonal functions, the mixes that emphasized group interventions were superior to Mix 3 in some respects).|
|Sarajuuri et al. 2005||39||Moderate- Severe||Evaluated outcome following a comprehensive, 6-week neurorehabilitation program with psychotherapy and vocational interventions as well as follow-up support.||Y
Other CRT Content: Conventional care and rehabilitation
Structured self-report questionnaire:
• Gainful work (full or part time)
• Household management
• Supported or sheltered work
• Work trials
• Productive but nongainful work (full or part time)
• Volunteer work
|At two-year follow-up, the treatment group improved significantly more than the control; 89 percent of treated patients were productive, compared with 55 percent of the control group.|
|Pre-Post Single Group|
|Braunling-Mcmorrow et al. 2010||205||Severe||Investigators examined the effect multifaceted rehabilitation services had on functional outcomes for participants with major physical and cognitive difficulties (n = 129) and those with behavioral complications (n = 76).||N||
Functional Area Outcome Menu:
• Behavioral and Emotional Status
• Community Participation
• Educational Endeavors
• Global Quality of Life
• Invovement in Vocational or Educational Endeavors
• Level of Awareness
• Level of Independence
• Level of Self-Managed Health
• Residential status
• Vocational Endeavors
|There were significant functional gains of approximately 1.5 levels, using the rehabilitation treatment model, for neuropsychologically impaired participants, both with and without related behavioral and substance problems.|
|Cicerone et al. 1996||20||Mild||This investigation was a retrospective analysis of a neuropsychological rehabilitation program for 20 patients who were part of a larger sample, referred for treatment due to postconcussive symptoms and functional disability.||N||
■ Digit Span, Forward and Backward
■ Trail Making Test
■ Continuous Performance Test of Attention
■ Logical Memory I and II
■ California Verbal Learning Test
■ Rey Immediate Recall
• Higher cognitive function
■ Rey Copy
■ WCST Perseveration
■ Category Test
■ Verbal Fluency
|Neuropsychological rehabilitation led to improved cognitive functioning and a reduction of post-concussive symptoms for those participants able to resume productive functioning; these improvements most affected areas of complex attention and information-processing speed. Participants who were not able to resume productive functioning showed decline in functioning on various measures at 1- and 6-month follow-up.|
|Huckans et al. 2010||21||Mild||Investigators piloted a Cognitive Strategy Training (CST) group treatment to examine if an increase in compensatory strategies use was observed; they also examined the effect of CST on self-reported psychiatric symptoms, cognitive symptoms, and life satisfaction.||N||
• Beck Depression Inventory
• Cognitive Symptom Severity
• Community Integration Questionnaire
• Frequency of Cognitive Strategy Usage Scale (FCSUS)
• Memory Compensation Questionnaire (MCQ)
• Memory Questionnaire
• PTSD Checklist
• Satisfaction with Life Scale
• Severity of Dependence Scale
• TBI self-efficacy scale
• Usefulness of Cognitive Strategy Scale (UCSS)
|There was a significant increase in compensatory strategies (MCQ), use of class strategies and day planners (FCSUS), internal cognitive strategies, and use of extensive cognitive aids. Secondary analyses showed significantly lower levels of depression, lower levels of memory and cognitive impairment, no change in PTSD symptoms, no change in community integration levels, and a trend toward increased self-efficacy.|
|Klonoff et al. 2010||103||Mild, Moderate, Severe||Examines cognitive retraining offered within a therapeutic milieu and driving outcomes, the relationship between metacognitive processes and clearance to drive, and the relationship between a patient’s working alliance (WA) and driving status at discharge.||N||
• Cognitive Retraining Behavior Checklist (CRBC)
• Digit Symbol
• Driving status
• Letter Scan (letter H)
• Matching Shapes
• Word Fluency
• Working Alliance (WA) scores
|In all participants, the Matching Shapes cognitive test showed a significant difference between driving groups: drivers rated better on organization, independence, use of compensations than non-drivers, and drivers had higher mean and discharge WA ratings.|
|Klonoff et al. 2007||101||Mild, Moderate, Severe||Examines the effectiveness of cognitive retraining exercises within a milieubased rehabilitation on work or school status.||N||
• Cognitive Retraining Behavior Checklist (CRBC)
• Digit Symbol
• Driving status
• Letter Scan (letter H)
• Matching Shapes
• Word Fluency
• Working Alliance (WA)
• Work/school status
|More than 80 percent of participants returned to work or school associated with better cognitive performance in tasks requiring information processing speed, visual scanning, visuospatial skills, and memory.|
|Mills et al. 1992||42||Severe||Examines outcomes of a 6-week, structured outpatient rehabilitation program, individualized to each patient, including psychological support and emphasizing improvements in realworld functional abilities. Evaluations included family and patient interviews to determine day-to-day functioning. Follow-up evaluations were conducted at 6, 12, and 18 months post treatment.||N||
• Functional evaluation
■ Identifying community resources
■ Consumer skills
■ Leisure activities
■ Safety awareness
■ Social communication
■ Telephone skills
■ Time management
■ Use of public transportation
• Individual treatment goals
• Speech pathology evaluation
|Patients’ functional outcomes improved significantly post treatment; the overall percentage of goals achieved was 67.5 percent. A positive correlation between the initial functional evaluation and cognitive assessments was reported, however there was no significant correlation between the change in functional and cognitive outcomes posttreatment, although a trend was indicated. The number of days in treatment correlated with functional improvement.|
|Murphy et al. 2006||232||Mild, Moderate, Severe||Evaluates a rehabilitation program to contain elements of group-based training, followed by work site-based rehabilitation including variable amounts of hospital- or community-based neurological rehabilitation. As such, the program was designed to include elements of cognitive rehabilitation as well as site-specific work placement training and rehabilitation.||N||
• Paid competitive employment
• Education or training
• Voluntary work
• Discharge to other services
• Client withdrew
• Discharged for other reasons
|Overall, 72 percent of subjects resumed independent activity, such as paid competitive employment, education or training, or voluntary work. Of the remaining participants, 15 percent were referred to further rehabilitation services, and 13 percent of the sample withdrew. The authors reported that there was no significant difference on severity of injury relative to outcome.|
|Walker et al. 2005||11||Severe||A pilot study to evaluate an extensive, three-stage program that ultimately focused on goal identification and achievement through tasks analysis and problem solving in goal areas.||A pilot study to evaluate an extensive, three-stage program that ultimately focused on goal identification and achievement through tasks analysis and problem solving in goal areas.||
• Individual goals related to:
• Psychological measures:
■ Depression, Anxiety, and Stress Scales
■ General Well-Being Questionnaire
■ European Brain Injury Questionnaire, family rating
|At least one goal was achieved by 10 of the 11 participants, with overall, 21 of 26 goals attained. There were no significant differences on psychological measures.|
Bowen, A., A. Tennant, V. Neumann, and M. A. Chamberlain. 1999. Evaluation of a community-based neuropsychological rehabilitation service for people with traumatic brain injury. Neurorehabilitation 13(3):147–155.
Braunling-McMorrow, D., S. J. Dollinger, M. Gould, T. Neumann, and R. Heiligenthal. 2010. Outcomes of post-acute rehabilitation for persons with brain injury. Brain Injury 24(7–8):928–938.
Braverman, S. E., J. Spector, D. L. Warden, B. C. Wilson, T. E. Ellis, M. J. Bamdad, and A. M. Salazar. 1999. A multidisciplinary TBI inpatient rehabilitation programme for active duty service members as part of a randomized clinical trial. Brain Injury 13(6):405–415.
Chen, S. H., J. D. Thomas, R. L. Glueckauf, and O. L. Bracy. 1997. The effectiveness of computer-assisted cognitive rehabilitation for persons with traumatic brain injury. Brain Injury 11(3):197–209.
Cicerone, K. D., L. C. Smith, W. Ellmo, H. R. Mangel, P. Nelson, R. F. Chase, and K. Kalmar. 1996. Neuropsychological rehabilitation of mild traumatic brain injury. Brain Injury 10(4):277–286.
Cicerone, K. D., T. Mott, J. Azulay, and J. C. Friel. 2004. Community integration and satisfaction with functioning after intensive cognitive rehabilitation for traumatic brain injury. Archives of Physical Medicine and Rehabilitation 85(6):943–950.
Cicerone, K. D., T. Mott, J. Azulay, M. A. Sharlow-Galella, W. J. Ellmo, S. Paradise, and J. C. Friel. 2008. A randomized controlled trial of holistic neuropsychologic rehabilitation after traumatic brain injury. Archives of Physical Medicine and Rehabilitation 89(12):2239–2249.
Goranson, T. E., R. E. Graves, D. Allison, and R. LaFreniere. 2003. Community integration following multidisciplinary rehabilitation for traumatic brain injury. Brain Injury 17(9):759–774.
Huckans, M., S. Pavawalla, T. Demadura, M. Kolessar, A. Seelye, N. Roost, E. W. Twamley, and D. Storzbach. 2010. A pilot study examining effects of group-based cognitive strategy training treatment on self-reported cognitive problems, psychiatric symptoms, functioning, and compensatory strategy use in OIF/OEF combat veterans with persistent mild cognitive disorder and history of traumatic brain injury. Journal of Rehabilitation Research and Development 47(1):43–60.
Klonoff, P. S., M. C. Talley, L. K. Dawson, S. M. Myles, L. M. Watt, J. A. Gehrels, and S. W. Henderson. 2007. The relationship of cognitive retraining to neurological patients’ work and school status. Brain Injury 21(11):1097–1107.
Klonoff, P. S., K. C. Olson, M. C. Talley, K. L. Husk, S. M. Myles, J. A. Gehrels, and L. K. Dawson. 2010. The relationship of cognitive retraining to neurological patients’ driving status: The role of process variables and compensation training. Brain Injury 24(2):63–73.
Middleton, D. K., M. J. Lambert, and L. B. Seggar. 1991. Neuropsychological rehabilitation: Microcomputer-assisted treatment of brain-injured adults. Perceptual & Motor Skills 72(2):527–530.
Mills, V. M., T. Nesbeda, D. I. Katz, and M. P. Alexander. 1992. Outcomes for traumatically brain-injured patients following post-acute rehabilitation programmes. Brain Injury 6(3):219–228.
Murphy, L., E. Chamberlain, J. Weir, A. Berry, D. Nathaniel-James, and R. Agnew. 2006. Effectiveness of vocational rehabilitation following acquired brain injury: Preliminary evaluation of a UK specialist rehabilitation programme. Brain Injury 20(11):1119–1129.
Parente, R., and M. Stapleton. 1999. Development of a cognitive strategies group for vocational training after traumatic brain injury. Neurorehabilitation 13(1):13–20.
Rattok, J., B. Ross, Y. Ben-Yishay, O. Ezrachi, S. Silver, P. Lakin, E. Vakil, E. Piasetsky, E. Zide, and L. Diller. 1992. Outcome of different treatment mixes in a multidimensional neuropsychological rehabilitation program. Neuropsychology 6(4):395–415.
Ruff, R. M., and H. Niemann 1990. Cognitive rehabilitation versus day treatment in head-injured adults. Is there an impact on emotional psychosocial adjustment? Brain Injury 4:339–347.
Salazar, A. M., D. L. Warden, K. Schwab, J. Spector, S. Braverman, J. Walter, R. Cole, M. M. Rosner, E. M. Martin, J. Ecklund, and R. G. Ellenbogen. 2000. Cognitive rehabilitation for traumatic brain injury: A randomized trial. Journal of the American Medical Association 283(23):3075–3081.
Sarajuuri, J. M., M. L. Kaipio, S. K. Koskinen, M. R. Niemela, A. R. Servo, and J. S. Vilkki. 2005. Outcome of a comprehensive neurorehabilitation program for patients with traumatic brain injury. Archives of Physical Medicine and Rehabilitation 86(12):2296–2302.
Tiersky, L. A., V. Anselmi, M. V. Johnston, J. Kurtyka, E. Roosen, T. Schwartz, and J. Deluca. 2005. A trial of neuropsychologic rehabilitation in mild-spectrum traumatic brain injury. Archives of Physical Medicine and Rehabilitation 86(8):1565–1574.
Vanderploeg, R. D., K. Schwab, W. C. Walker, J. A. Fraser, B. J. Sigford, E. S. Date, S. G. Scott, G. Curtiss, A. M. Salazar, and D. L. Warden. 2008. Rehabilitation of traumatic brain injury in active duty military personnel and veterans: Defense and Veterans Brain Injury Center randomized controlled trial of two rehabilitation approaches. Archives of Physical Medicine and Rehabilitation 89(12):2227–2238.
Walker, A. J., M. Onus, M. Doyle, J. Clare, and K. McCarthy. 2005. Cognitive rehabilitation after severe traumatic brain injury: A pilot programme of goal planning and outdoor adventure course participation. Brain Injury 19(14):1237–1241.
Warden, D. L., A. M. Salazar, E. M. Martin, K. A. Schwab, M. Coyle, and J. Walter. 2000. A home program of rehabilitation for moderately severe traumatic brain injury patients. Journal of Head Trauma Rehabilitation 15(5):1092–1102.
Zhu, X. L., W. S. Poon, C. C. H. Chan, and S. S. H. Chan. 2007. Does intensive rehabilitation improve the functional outcome of patients with traumatic brain injury (TBI)? A randomized controlled trial. Brain Injury 21(7):681–690.