Executive function is generally described as a set of integrated cognitive processes necessary to perform or accomplish everyday life activities. These cognitive processes allow individuals to plan or develop goals, initiate behavior, solve problems, anticipate consequences of actions, monitor progress toward goals, reason, strategize, direct attention to goal-relevant information, and manage time and space (Cicerone et al. 2000; Kennedy et al. 2008). Deficits in executive functions may include an inability to perform these cognitive processes or a lack of awareness that these or other cognitive and physical deficits exist and impede everyday life (Kennedy et al. 2008; Stuss 1991). Therefore, this chapter reviews the evidence for treatment of executive function in two main sections: awareness (i.e., deficits in self-awareness) and non-awareness (e.g., deficits in problem solving, planning, initiating behavior). Because executive function incorporates a number of subprocesses, and there is no consensus on precisely how to subdivide this complex domain, treatment development has typically focused on addressing individual subcomponents rather than the entire domain of executive function. Multiple approaches to the larger executive domain are sometimes included in comprehensive treatment programs. The committee’s conclusions are provided at the end of each section, in awareness and non-awareness.
The committee could not find any randomized controlled trials (RCTs) of mild traumatic brain injury (TBI) and awareness, perhaps reflecting the
fact that awareness deficits are more typically associated with more severe injuries. The committee reviewed four studies of participants with moderate-severe injuries who were in the chronic stage of recovery—two RCTs (Cheng and Man 2006; Goverover et al. 2007) and two single-subject, multiple baseline experiments of treatments intended to improve awareness of deficits (Sohlberg et al. 1998; Toglia et al. 2010). The committee did not find any nonrandomized, parallel group studies or pre-post designs on awareness. Table 8-1 presents a summary of all included studies in this review.
Chronic Phase of Recovery, Moderate-Severe TBI
Randomized Controlled Trials
Goverover et al. (2007) examined the effects of an awareness training protocol embedded within the practice of instrumental activities of daily living (IADLs) as compared to IADL training without any self-awareness training. The 20 participants had moderate-severe injuries that occurred an average of about 10 months prior to trial entry; participants’ phase of recovery ranged between the subacute and chronic stages. Participants were randomly assigned to either group, and treatments were provided in six, 45-minute sessions, two or three times per week, across 3 weeks. Tasks were identical in the treatment and control groups; however, the treatment group participants were asked to predict their own performance on the IADL tasks and to self-evaluate performance immediately after tasks. They received immediate feedback from therapists, as well as instruction to write about their experiences in a journal. Improvement in task-specific self-awareness (AAD scores) was not significantly different between the groups. Improvement in a self-regulation skill inventory was significantly greater in the treatment group, after adjusting for baseline scores. Functional performance as reflected by Assessment of Motor and Process Skills (AMPS) scores also improved significantly more for the treated group than for the control group. Distal outcomes (e.g., secondary measures) were not significantly different between the groups, including an Awareness Questionnaire.
Cheng and Man (2006) investigated a newly developed Awareness Intervention Program (AIP) compared to a conventional rehabilitation program. The AIP focused on improving awareness of the patient’s disease and related deficits such as physical or cognitive function. The AIP included educational sessions based on the types of deficits manifested by the patients and functional training sessions, in which patients practiced setting performance goals and then evaluating their own performance against those goals. The conventional rehabilitation program included physical, functional, and cognitive aspects of occupational therapy. The 21 subjects participating in the study were in the subacute phase of recovery from
|Study||N||TBI Severity Level||Brief Narrative||Comparator||Outcome Measures||Findings|
Cheng and Man 2006
|21||Moderate-Severe||Comparison of a systematic intervention, Awareness Intervention Program (AIP), versus conventional rehabilitation, to improve self-awareness.||Y
Other CRT Content: Conventional rehabilitation
• Self-Awareness of Deficits Interview (SADI)
• Lawton Instrumental Activities of Daily Living (IADL) scale
|Analysis showed the AIP training group improved significantly over conventional rehabilitation group.|
|Goverover et al. 2007||10||Moderate-Severe||Examined a self-awareness retraining program using practice of IADLs to improve task-specific self-awareness and self-regulation.||Y
Other CRT Content: IADL training
• Awareness Questionnaire
• Assessment of Awareness of Disability
• Self-Regulation Skills Inventory (SRSI)
• Assessment of Motor and Process Skills (AMPS)
|Improvements in self-regulation skill and functional performance were significant in intervention group.|
|Single Subject, Multiple Baseline Experiment
Sohlberg et al. 1998
|3||Moderate-Severe||Examined three categories of awareness to determine optimal outcome measures in awareness interventions for future research (a pilot study).||N||
• Caregiver rating of subject awareness
• Self/Other Rating Form
■ Caregiver rating
■ Subject rating
• Photograph (portion of the “Picture This!”) rating
• Behavioral measure
|Qualitative analysis suggested improved awareness after treatment via behavioral measure; no change in awareness via Self/Other measures.|
|Toglia et al. 2010||4||Not reported||Identified multifaceted task training approach to assess use of and skill in cognitive strategies to accomplish tasks and assess changes in general awareness (a pilot study).||N||
• Proximal outcomes:
■ Executive Function Performance Test subtask
■ Multiple Errands Test
• Distal outcomes:
■ Awareness Questionnaire
■ Behavior Rating Inventory of Executive Function
|Participants improved in self-regulatory skills and strategy use across tasks; general awareness deficits were unchanged.|
Constantinidou at al. 2008
Evaluating the Categorization Program (CP) to determine if it improves cognitive abilities.
• Part A included object categorization, teaching perceptual features to describe objects or living things, and higher levels of cognitive function including analyses, synthesis, linguistic flexibility, and abstract reasoning;
Other CRT Content: Typical treatment regimen of the rehabilitation site
• California Verbal Learning Test II
• Community Integration Questionnaire (CIQ)
• Control Oral Word Association
• Mayo-Portland Adaptability Inventory III
• Rey Complex Figure Test
• Scales of Cognitive Ability for TBI
• Symbol Digits Modalities Test
• The Booklet Category Test
• Trail Making Tests
• Wechsler Abbreviated Scale of Intelligence
|Both groups improved on a number of neuropsychological measures, and functional improvement was comparable between groups; the experimental group showed significant improvement on more measures, with better maintenance. A group comparison for change in neuropsychological measures was not performed.|
• Part B included new category learning, focusing on learning rules for categorization, with use of errorless learning principles and cueing hierarchies.
• Wechsler Memory Scale III
• Wisconsin Card Sorting Test
• Woodcock-Johnson III
|Couillet et al. 2010||12||Severe||Research on a rehabilitation program for divided attention, i.e., the ability to complete dual tasks simultaneously (e.g., walking and talking). Subjects randomized in a crossover design.||Y
Other CRT Content: Nonspecific cognitive training
• Divided Attention Tests:
■ Test for Attention Performance (TAP), subtest for divided attention
■ Go.no go, and digit span
• Executive and working memory tests:
■ Brown-Peterson Paradigm
■ TAP, subtest for flexibility
■ Trail Making Test
• Rating Scale of Attentional Behaviour
|Significant treatment effect shown for dual task on the TAP (divided attention subtest), go-no go, and digit span.|
|Evans et al. 2009||19||NR||Evaluated the efficacy of a cognitive-motor dual-tasking training program to improve performance with dual-tasking difficulties.||Y
No Content: Placebo (no training) control
• Spot the Word Test or National Adult Reading Test
• Divided Attention and Dual-Tasking Battery:
■ Speed of Comprehension Task, adapted
■ Tone counting
|Improved performance among treatment group shown for sentences and walking dual-task. Treatment group showed a reduction in self-reported difficulties (authors suggest caution in interpretation).|
|Fasotti et al. 2000||22||Severe||
Evaluates attention deficit rehabilitation program, Time Pressure Management (TPM). In brief, videotaped stories, a compensatory cognitive strategy is outlined with four objectives:
• Recognize time pressure in the task at hand;
• Prevent as much time pressure as possible;
• Deal with time pressure as quickly and effectively as possible; and
• Urge the patient to monitor himself while using TPM.
Other CRT Content: Concentration training from an existing memory training program
• Auditory Concentration Test
• Dutch version, Rey’s 15 Words Test
• Rivermead Behavioural Memory Test (RBMT)
• Visual Choice Reaction Time
• Visual Simple Reaction Time
|No significant difference between groups on use of anticipatory strategies or actual task performance. Treatment group showed some improvement on neuropsychological measures; authors do not identify how strategies could be evaluated by standardized tests.|
|Hewitt et al. 2006||30||Severe||Examines a method to aid individuals with “dysexuctive syndrome,” or problems with planning and problem solving, with overcoming related challenges.||Y
No Content: Control received no treatment
• Brixton Test
• Hayling Test
• RBMT screening score
• Speed and Capacity of Language Processing Test
• The Modified Six Elements Test
|Group by time interaction significantly improved for the treatment group, but no effect for group or time separately.|
|Levine et al. 2000||30||Moderate- Severe||Assessed the effects of Goal Management Training to improve disorganized, or maintenance of goaldirected, behavior.||Y
Non-CRT Content: Motor skills training
• Paper-and-pencil tasks
• Room layout
• Neuropsychological tests:
■ Trail Making Test
■ WAIS-R, digit symbol subtest
|Treatment group improved significantly from pre- to posttest compared to control. Treatment group reported to take more time to complete tasks—authors determined as evidence of care and double-checking.|
|Rath et al. 2003||60||Mild, Moderate-Severe||Group treatment focusing on emotional self-regulation strategies for problem orientation as well as problem-solving skills; including weekly “consolidation sessions” to review materials and notes from each group.||Y
Other CRT Content: Conventional treatment
• Problem solving tests:
■ Perseverative Response score
■ Problem Solving Inventory (PSI)
■ Wisconsin Card Sorting Test (WCST)
• Psychosocial functioning tests:
■ Brief Symptom Inventory
■ Community Integration Questionnaire
■ Problem Checklist
■ Recreation + Social Interaction composite score
■ Rosenberg Self-Esteem Scale
■ Sickness Impact Profile
• Cognitive Skills tests:
■ Stroop Color-Word Task
■ Watson-Glaser Critical Thinking Appraisal
■ Wechsler Memory Scale-III
■ Weinberg Visual Cancellation Test
■ Will-Temperament Scale
|Nearly 25 percent of participants overall (both groups) dropped out prior to post-test measures. Both groups improved on a wide range of measures generally subject to practice effects or expectation of improvement.|
|Webb and Glueckauf 1994||16||NR||Goal setting training involving orientation, goal setting, goal monitoring with use of worksheet and follow-up form.||Y
Other CRT Content: Less intensive goalsetting activities
|Goal Attainment Scaling (GAS)||Both groups improved to comparable degree. Treatment group maintained effect at 1-month follow-up.|
|Nonrandomized, Parallel Controlled Group
|8||Mild||Evaluating the efficacy of a treatment, based on conceptualization of deficits related to working memory, to address deficits of attention.||Y
No Content: Control group did not receive treatment
• 2 and 7 Test
• Attention Rating and Monitoring Scale (ARMS) selfreport scale
• Continuous Performance Test of Attention
• Trail Making Test
■ Automatic Detection Speed
■ Controlled Processing Speed
|Difference between groups was significant, with treatment group demonstrating clinically meaningful improvement on majority of initially impaired measures. Treatment group demonstrated significantly greater reduction in the experience of attentional difficulties.|
|Fong et al. 2009||33||Moderate||Examined the effectiveness of a problem-solving skills training program, delivered to the treatment group, when provided with conventional cognitive training, which all participants received.||Y
Other CRT Content: Conventional cognitive training
• Key Search test
• Means-Ends Problem-Solving Measure (MEPSM)
• Metacomponential Interview (MI)
• Modified Six Elements test
• Raven’s Progressive Matrices (RPM)
• Social Problem-Solving Video Measure (SPSVM)
|Significant improvement was reported in the treatment group by the MI. No significant difference in change scores between groups for most outcomes.|
|Man et al. 2006||50||NR||Evaluated the efficacy of a pictorial-based analogical problem-solving skills training program to help patients better learn problem-solving skills through systematic, theoretically driven strategies.||Y
No Content: Placebo (no treatment)
• Halstead Reitan Test Battery (HRTB), Category Test
• Lawton IADL scale, Hong Kong-Chinese version
|Treatment group demonstrated some improvement in functional and overall problem-solving skills, immediately posttreatment and maintained at 4 weeks.|
|Manly et al. 2002||10||Mild, Moderate, Severe||Examines treatment for dysexecutive syndrome, or challenges in planning and problem solving for everyday life activities, using a brief auditory stimuli to interrupt activity and cue patients to consider the overall goal.||Y
Other CRT Content: Controls were tested on task without use of auditory cues
• Hotel Task
■ Compiling individual bills
■ Sorting the charity collection
■ Looking up telephone numbers
■ Sorting conference labels
■ Opening and closing garage door
• Version B same tests, different timing
|Treatment group did not perform significantly different than control.|
|Pre-Post Single Group
Constantinidou et al. 2005
|23||Moderate-Severe||Explores the effects of the Categorization Program (CP), a systematic treatment program, to improve categorization performance.||N||
• Categorization tests
■ Object categorization
■ New category learning tasks
• Community Integration Questionnaire (CIQ)
• Mayo-Portland Adaptability Inventory-3 (MPAI-3)
• Probe Tasks
|All participants showed improvement following CP training at and second posttests. Probe Tasks elicited much lower responses from TBI participants than at the first probe, and their performance improved systematically in a linear fashion as they received more training on the CP. The overall multivariate probe effect was also significant, reflecting the improvement on the three probes demonstrated by participants with TBI.|
|Fish et al. 2007||20||Severe||Examined use of an alerting strategy with a content-free cue to determine utility for a prospective memory task, comparing cued and noncued days for completion of telephone calls.||N||
• Cambridge Prospective Memory Test
• Cattell Culture Fair Test
• Hotel test
• National Adult Reading Test (NART)
• RBMT, story recall
• Rey Auditory Verbal Learning Test
• Sustained Attention to Response Test (SART)
• Trail Making Test
|Significant difference reported after cued calls versus non-cued days.|
|Marshall et al. 2004||20||Mild, Moderate, Severe||Examined the effects of behavioural intervention, an interactive strategy modeling training (ISMT), on challenges with problem solving. Participants asked to use a question-and-answer game to identify a picture.||N||
• Raven Colored Progressive Matrices
• Boston Naming Test
• Rapid Assessment of Problem Solving test (RAPS)
■ Number of questions
■ Percent of constraint-seeking (CS) questions
■ Question-asking efficiency scores
|Significant difference on pre- and posttest scores for number of questions and percent of CS questions. No differences between post training and follow-up on all measures.|
|Serino et al. 2007||9||Mild, Moderate, Severe||Investigated the efficacy of a rehabilitative program—Working Memory Training (WMT)—to improve working memory and other, dependent cognitive functions by targeting the central executive system.||N||
• General stimulation training (GST)
• Patient Competency Rating Scale
• Rivermead Head Injury Follow-Up Questionnaire
|Significant performance improvement reported from intermediate to final for WMT, as noted by tests for working memory, divided attention, executive function, but not in speed processing and sustained attention. No differences seen between screening and intermediate testing sessions (for GST).|
|Single Subject, Multiple Baseline Experimental Design
Dawson et al. 2009
|3||Mild, Severe||The study trained three adults to use a metacognitive strategy (Goal- Plan-Do-Check) to solve everyday performance problems.||N||Canadian Occupational Performance Measure||Changes according to self- and significant other ratings showed performance improvement (on seven of nine trained goals, four of seven untrained goals) as well as satisfaction with treatment results.|
|Delazer et al. 1998||3||Severe||N||Studies three patients with severe cognitive impairment leading to difficulty solving problems||
• Number of correct steps
• Number of correct answers
• Number of encoding errors
• Number of execution errors
• Time for completion
• Neuropsychological measures:
■ Munchner Gedachtnis Test
■ Rey Osterrieth Complex Figure
■ Digit span forward
■ Digit span backward
■ Tower of London
■ Trail Making Test
■ Raven SPM
■ Graded Difficulty Arithmetic Test
|Number of correct steps increased for all three participants from baseline to session two. No notable changes were recorded on neuropsychological measures.|
|Ehlhardt et al. 2005||4||Severe||Evaluating the TEACH-M program to assess ability to learn multistep procedures utilizing specific cognitive rehabilitation principles.||N||
• Number of correct e-mail steps completed in sequence
• Number of correct e-mail steps completed, regardless of sequence
• Number of training sessions required to reach criterion for mastery
|Nott et al. 2008||8||Mild, Moderate, Severe||Occupational therapy alternated with a rehabilitation program, Perceive, Recall, Plan, and Perform (PRPP), to track information processing capacity during tasks.||N||PRPP system of task analysis||Difference in performance across phases was statistically sig for seven of eight participants. Post hoc testing different between baseline and intervention.|
|Vallat-Azouvi et al. 2009||2||Severe||Assess effectiveness of rehabilitation program to address working memory in two individuals with central executive deficits.||N||
• Central executive tasks:
■ Brown-Peterson paradigm
• Digit and visuo-spatial span
• Ecological questionnaire
• Non-target tasks
■ Go-no go
■ Trail Making Test
• Working memory tasks:
■ Arithmetic problem solving
■ Dual task
|Improvements noted in working memory and central executive tasks, as well as the ecological questionnaire. Nontarget measures were unchanged. Authors suggested that improvements were maintained in one subject.|
|Zencius et al. 1998||2||NR||Orientation questions presented to two individuals with executive deficits. Varied formats included individual sessions of flashcards with oral questioning, or group sessions with peer questions, written response, and verbal repetition.||N||Frequency of correct answers to questions||Improvement noted for both individuals from baseline to posttreatment, providing support for increasing orientation to person, place, and time using questions as well as for fading the prompting used during the training procedure.|
what was likely moderate-severe TBI. The AIP treatment program consisted of two individual sessions a day, 5 days per week, for 4 weeks. The AIP group demonstrated significantly improved awareness as compared to the conventional rehabilitation group. Functional outcomes did not differ between the groups.
Single-Subject, Multiple Baseline Experiments
Sohlberg et al. (1998) conducted a pilot study to assess three categories of awareness measures administered to three individuals with moderate-severe brain injury and their caregivers. Individuals were 7 to 21 years postinjury. This pilot study intended to determine which set of outcome measures would be more useful for further research in awareness interventions. Two groups of outcome measures were used to determine improved awareness in participants: behavioral indicators (e.g., increased independence, decreased interruptive behavior) and perceptions (self and others’ [e.g., caregivers’]) regarding awareness abilities (e.g., caregiver ratings and self-ratings of competency, self-judgments about likely cognitive breakdowns depicted photographically, or global ratings by a significant other). The treatment consisted of showing patients pictures of activities they were likely or unlikely to experience as cognitive failures (e.g., forgetting peoples’ names, forgetting to move the wet laundry from the washing machine to the dryer). To judge self-awareness, the examiner asked each subject whether the photographs represent problems they were likely or unlikely to experience. Qualitative analysis suggested dissociation between behavioral and perceptual indicators of awareness. Behavioral measures showed improved awareness after treatment; others’/self-perception measures showed no change in awareness.
Toglia et al. (2010) conducted a single-subject design trial with four subjects, using a multi-context approach to promote strategy use across situations and increase self-regulation, awareness, and functional performance. Treatment included nine 75-minute treatment sessions, provided twice a week for approximately 5 weeks. Sessions were divided into three phases: error-discovery, strategy training and mediation, and reinforcement of strategy. Each session included different multi-step (i.e., 10–15 steps) tasks, approached in various settings such as a kitchen or office. In qualitative analysis, participants demonstrated improvement in self-regulatory skills and strategy use. General awareness of deficits remained unchanged in these subjects.
The committee found no evidence from two RCTs (Cheng and Man 2006; Goverover et al. 2007) that self-awareness training produced
an overall increase in self-awareness beyond the types of tasks and activities that were the subject of self-appraisal (i.e., patient-centered outcomes).
The committee found no evidence from two RCTs (Cheng and Man 2006; Goverover et al. 2007) that measured posttreatment follow-up to show whether awareness treatment effects were maintained.
The committee found limited evidence from two RCTs (Cheng and Man 2006; Goverover et al. 2007) that showed an immediate increase in accuracy of self-assessment and self-regulation from treatments that involved practice in prediction and evaluation of task performance, for individuals with chronic stage, moderate-severe TBI.
The committee found no studies of cognitive rehabilitation therapy (CRT) for awareness deficits in mild TBI or subacute, moderate-severe TBI. The committee reviewed two RCTs and two single-subject, multiple baseline studies to address awareness deficits in patients with moderate-severe TBI in the chronic phase of recovery. The evidence provides no support for long-term treatment effect. Treatment effects show benefit for immediate/short-term outcomes, such as improvement in self-regulatory skills.
The committee reviewed eight RCTs of treatments intended to improve cognitive aspects of executive function (i.e., aspects other than self-awareness). These studies speak primarily to treatments for individuals in the chronic phase with at least moderate injuries. Seven of them were conducted in the chronic phase, with one (Couillet et al. 2010) enrolling patients in both subacute and chronic phases. Seven of the studies enrolled only participants with traumatic injuries, while one (Evans et al. 2009) included a mixture of individuals with TBI and stroke, although a majority had TBI. Most studies included only patients with moderate or severe injuries, while two RCTs (Levine et al. 2000; Rath et al. 2003) included individuals with mild injuries; however, the results in these two studies were not separated by subgroup for analysis. One study (Evans et al. 2009) defined severity with respect to the executive impairment of interest, rather than injury severity. The ages of those treated ranged from the late 20s to early 40s. The studies enrolled a total of 218 participants, with sample sizes in each treatment arm ranging from 5 to 30. Two of these studies compared the experimental intervention to no treatment (Hewitt et al. 2006, used an unfilled waiting interval; Evans et al. , used “usual care”), one to a physical skill training intervention (Levine et al. 2000), and five
to other forms of cognitive treatment. Five of the treatments studied were compensatory in nature, two (Couillet et al. 2010; Evans et al. 2009) were restorative, and one (Constantinidou et al. 2008) was less clearly classifiable between restorative and compensatory. The committee also identified four nonrandomized, parallel group designs, four pre-post single group designs, and six single-subject, multiple baseline experiments.
Chronic Phase of Recovery, Moderate-Severe TBI
Comparator Group: No or Minimal Content
Evans et al. (2009) evaluated the effectiveness of a 5-week cognitive– motor dual-tasking training program developed to improve the performance of a group of people with divided attention difficulties arising from brain injury and thought to place demands on executive function. A treatment group of 10 people was compared with a control group of 9; the control group received no training. The intervention involved twice-daily exercises involving walking in combination with tasks of increasing cognitive demand over the course of the intervention. The primary outcome measure was a task requiring participants to walk and carry out a spoken sentence verification task simultaneously. Secondary outcome measures were measures of dual-tasking involving either two motor tasks or two cognitive tasks. A questionnaire measure relating to daily activities requiring divided attention was also completed. Compliance with the training program was good. Results showed evidence of improvement in performance on the primary outcome measure, but little evidence of generalization to other measures. Some evidence showed that participants believed their dual-tasking performance in everyday life improved after the intervention. The study was limited in terms of sample size, was not blinded, and did not control fully for therapist contact time, but it has produced valuable data relating to effect sizes associated with this form of intervention.
Hewitt et al. (2006) assessed participants’ ability to develop a plan to accomplish a minimally familiar task such as planning a trip. Participants were asked to list the steps required to accomplish a simulated task prior to treatment. They were randomized to then have a 30-minute break or 30 minutes of instruction in an approach to task planning that asked them to recall an example of a similar activity that they had planned in the past and consider that task in planning a new one. The outcome measures were number of steps listed and effectiveness of the new plan, and they were assessed immediately after the break/strategy training by raters blinded to the group assignment. Both groups improved on these measures, with the strategy training group improving more from pre- to posttest. This study suggests that such a strategy is useful in improving the planning of complex
activities, but does not answer the question of whether the strategy can be trained in such a way that it is retained and used in daily life.
Comparator Group: Non-CRT Content
Levine et al. (2000) assessed a strategy entitled Goal Management Training (GMT), in which an overt sequence of steps leading from a goal, to a set of actions to accomplish the goal, to a checking process that assesses progress toward that goal, is taught as a way to enhance the completion of goal-directed activities. Participants attempted to perform a set of laboratory-based simulations of real-world tasks, which were scored for time and errors. The participants were then randomized to receive either a motor skills training group or a GMT group for a single, 4- to 6-hour training session. In the GMT group, the training session involved didactic teaching of the GMT concept and practice applying it to a set of simulated activities similar to those used at baseline. Subsequently, both groups were reassessed on a similar set of simulated activities. The degree of improvement in errors from pre- to posttesting was significantly larger for the GMT group than the motor skills group, and GMT group members performed some activities more slowly, interpreted as evidence of care and “checking.” Although two of the trained activities were used in the assessment, another task that was not part of the GMT also showed differential improvement suggestive of short-term generalization of the strategy. This study suggests that GMT can be helpful when used, but does not answer the question of how to achieve regular spontaneous use of the strategy in daily life.
Comparator Group: Other CRT Content
Constantinidou et al. (2008) examined whether intensive training in categorization results in improvement in two untrained categorization tasks, a battery of neuropsychological tests, and a functional assessment scale. The comparison group received “usual care” including a range of cognitive rehabilitation activities, but without an intense focus on categorization training. Both groups received approximately 60 hours of training over about 13 weeks. The experimental group performed significantly better on both categorization tasks after treatment than the comparison group, whereas the two groups did not differ significantly prior to treatment. Also, the ability to categorize appeared better maintained across follow-up probes in the experimental group. Both groups improved on a number of the neuropsychological measures, and the experimental group improved significantly on more of them. However, a comparison of change in neuro-psychological measures was not conducted. Functional improvement was comparable between the two groups. These conclusions are tempered by the
small group size, the fact that direct tests that group by time comparison were not statistically significant, and the lack of direct comparison of the neuropsychological outcomes.
Couillet et al. (2010) conducted a randomized crossover design addressing divided attention difficulties. The study included 12 patients at a subacute or chronic stage of recovery after severe TBI. Treatment consisted of training to perform two concurrent tasks using a hierarchical order of difficulty that progressively increased task difficulty following each patient’s individual improvement. A variety of task combinations were used during training. The control group practiced a range of computerized and paper and pencil tasks that did not require divided attention. Training lasted 6 weeks, with four, 1-hour sessions per week. Outcome measures included specific divided attention measures, other executive and working memory tasks, nontarget cognitive tasks to assess the specificity of treatment, and the Rating Scale of Attentional Behaviour addressing attentional problems in everyday life. The authors reported a significant treatment effect for divided attention measures and on the divided attention item of the Rating Scale of Attentional Behaviour. Less consistent effects were seen on other executive and working memory measures, and no significant effect was seen on nontarget measures.
Fasotti et al. (2000) studied a strategy training intervention entitled Time Pressure Management (TPM), which is based on the premise that slowed information processing leads to task failures and that strategies such as avoiding interruptions, taking the necessary time, taking pauses, etc., may lead to improved task performance. The experimental group was taught this strategy and practiced it for about 7 hours over 2 to 3 weeks. The comparison group was given didactic instruction in “how to concentrate.” Both were then assessed on two simulated tasks in which they had to recall directions provided via videotape or perform a computer task when given recorded directions. Performance on these tasks was coded with respect to specific TPM strategies that were performed in anticipation of task problems and in response to task problems, as well as quality of actual task performance. Both groups were also assessed on a range of neuropsychological and psycho-social measures. After treatment, the two groups did not differ on the use of anticipatory strategies; the TPM group using TPM strategies in response to task problems. Actual task performance did not differ between the groups. Interestingly, performance on the neuro-psychological test battery, but not the psychosocial measures, improved more in the TPM group, despite the fact that it is not obvious how the strategies taught can be applied during standardized testing.
Rath et al. (2003) compared two multi-component group treatment programs for problem solving deficits. Both groups received 2 to 3 hours of treatment per week over 24 weeks, although the experimental group
received treatment in a single, longer weekly block while the comparison group had shorter sessions across the week. The experimental group followed a structured lesson plan that started with problem orientation (i.e., identification of problems, attitudes toward problem solving, attribution of problem sources) and then focused on applying specific problem solving strategies to real-world problems. The comparison group’s treatment focused on several different cognitive domains as well as psychosocial adjustment, but without the specific focus on a problem solving framework. Multiple outcome measures focusing on attention, memory, problem solving, emotional adjustment, and physical symptoms, as well as caregiver reports, were assessed. Unfortunately, 5 of 32 participants assigned to the experimental group and 9 of 28 participants assigned to the comparison group dropped out prior to outcome assessment (nearly 25 percent overall). Moreover, the degree of improvement seen in the two groups was not directly compared statistically. Relative improvement between the two groups was impossible to assess because the outcome measures that improved significantly within each group (10 measures in the experimental group, 8 in the comparison group) were reported with effect sizes. However, no effect sizes were reported for those measures that did not improve significantly, nor were confidence intervals around the effect sizes reported. Both groups appeared to show significant improvement in a wide range of measures, but some of the measures are subject to practice effects and/or expectation of improvement.
Webb and Glueckauf (1994) assessed whether participant involvement in setting and reviewing treatment goals affected progress toward those goals or retention of improvement. Two groups participated in the identification of a priority behavioral goal, as well as a goal attainment scaling (GAS) exercise to anchor potential outcomes with respect to that goal into a five-point scale. One group was involved in more intensive discussion of the goal and more intensive review and reflection on the goal and progress toward it at weekly follow-up sessions. Both groups made progress on the GAS scale from pre- to posttreatment. The intensive goal group maintained this improvement at 2-month follow-up, whereas the other group regressed by the follow-up assessment. Each group lost participants; two dropped from the intensive training, and three dropped out from the other. Moreover, the degree of GAS improvement or maintenance was not statistically assessed head to head.
Nonrandomized, Parallel Group Designs
Fong and Howie (2009) studied a program of explicit problem solving training. Experimental and control groups were formed from pairs of participants matched on demographic and injury severity measures. All
participants received conventional cognitive training composed of functional skills training. The experimental intervention consisted of additional explicit training in problem solving skills with an emphasis on metacomponential strategies, delivered in 22 75-minute sessions over 15 weeks. The treatment was oriented toward the primary metacomponents of problem solving: defining the problem, representing the problem, planning problem solving strategies, monitoring selected strategies, and evaluating outcomes. Patients from the treatment group improved significantly on tests that assessed metacognitive ability. The significance level of this result would not have survived corrections for multiple comparison, and it was not clear which of the 22 outcome measures would have been considered sufficiently relevant to require correction.
This and the other nonrandomized, parallel group studies (Cicerone 2002; Man et al. 2006; Manly et al. 2002), single group pre-post studies (Constantinidou et al. 2005; Fish et al. 2007; Marshall et al. 2004; Serino et al. 2007), and single-subject, experimental designs (Dawson et al. 2009; Delazer et al. 1998; Ehlhardt et al. 2005; Nott et al. 2008; Vallat-Azouvi et al. 2009; Zencius et al. 1998) provided modest support for the conclusions of the RCTs. In general, the methodology of these studies was weaker, not only due to the nonrandomized nature of treatment assignment or single group design, but also due to very small sample sizes and inappropriate use of statistics in some cases. Like several of the RCTs, many were pilot studies or proof-of-principle trials that aimed to test the potential for a new intervention to be utilized in larger studies with more substantial statistical power.
In addition, the generalizability of some of the studies was limited due to extensive methodological overlap between the intervention and the primary outcome measures (e.g., Constantinidou et al. 2005; Ehlhardt et al. 2005; Marshall et al. 2004). However, supportive evidence was provided for interventions that demonstrated early promise, some of them with implications for the functional consquences of the interventions. Externally originated alertness enhancement (random beeps during a reasoning task) facilitated attention and reasoning performance during a time-allocation task (Manly et al. 2002). The notion that metacognitive interventions such as context-free reminders could be successfully applied to facilitate memory for real-world tasks was also supported (Fish et al. 2007).
• The committee found studies of goal management training, intensive goal setting, familiar tasks as a planning template, and TPM
(Constantinidou et al. 2008; Fasotti et al. 2000; Hewitt et al. 2006; Levine et al. 2000) not informative for conclusions about the impact on patient-centered outcomes (quality of life, functional status).
• The committee found studies of goal management training, intensive goal setting, familiar tasks as a planning template, TPM, or training in divided attention (Constantinidou et al. 2008; Couillet et al. 2010; Evans et al. 2009; Fasotti et al. 2000; Hewitt et al. 2006; Levine et al. 2000) not informative regarding measures of posttreatment follow-up to show whether goal management training treatment effects were maintained.
• The committee found studies of goal management training, intensive goal setting, familiar tasks as a planning template, and TPM (Constantinidou et al. 2008; Fasotti et al. 2000; Hewitt et al. 2006; Levine et al. 2000) not informative to show benefit from goal management training beyond the training session for individuals with chronic, moderate-severe TBI.
• The committee found limited evidence for conclusions about the impact (efficacy) of training in divided attention on patient-centered outcomes (Couillet et al. 2010; Evans et al. 2009).
• The committee found limited evidence that training in divided attention led to immediate enhancement of divided attention performance beyond the combination of tasks trained (Couillet et al. 2010; Evans et al. 2009).
In summary, the committee evaluated a wide range of strategies, primarily compensatory, in patients with executive deficits related to moderate-severe TBI. There is evidence that GMT, using prior planned tasks as guides to planning new tasks, intensive involvement in goal setting, and delivery of content-free alerting stimuli during performance of complex tasks, may enhance task accomplishment. However, these studies did not establish the spontaneous use of these strategies after longer-term treatment or the breadth of tasks for which such strategies might be beneficial. The evidence for TPM is weaker since the use of the trained strategies did not result in clear improvements in performance, and, again, longer-term treatment with intent to generalize to daily life was not studied. The benefits of categorization training are less clear from research to date. Two of the trials (Hewitt et al. 2006; Levine et al. 2000) were essentially proof of principle studies, which assessed the immediate benefit of a single session of strategy training, as opposed to the longer-term benefit of a course of treatment.
Studies of divided attention training provided somewhat conflicting results. Both studies suggest improvement in performance of combinations of tasks that were performed together in training (Couillet et al. 2010; Evans et al. 2009), but only one (Couillet et al. 2010) suggested generalization to other task combinations. Because many combinations of tasks were used in training and their similarity to the outcome tasks is unclear, the degree of generalization implied by the outcome task performance improvement is unclear.
Other intensive executive treatments, such as those studied by Rath et al. (2003), are difficult to assess because of the lack of direct comparison to an alternative treatment (i.e., comparator included other CRT-like components). Because of the preliminary nature of most of the executive treatments studied, patient-centered outcomes were rarely included in the outcome measures. Thus, although several compensatory strategy training approaches show enhanced executive management of complex tasks on a short-term basis, there is limited evidence from two RCTs to document longer-term change to demonstrate the impact of such treatments on real-world performance (Couillet et al. 2010; Evans et al. 2009).
Cheng, S. K., and D. W. Man. 2006. Management of impaired self-awareness in persons with traumatic brain injury. Brain Injury 20(6):621–628.
Cicerone, K. D. 2002. Remediation of “working attention” in mild traumatic brain injury. Brain Injury 16(3):185–195.
Cicerone, K. D., C. Dahlberg, K. Kalmar, D. M. Langenbahn, J. F. Malce, T. F. Bergquist, T. Felicetti, J. T. Giacino, J. P. Harley, D. E. Harrington, J. Herzog, S. Kneipp, L. Laatsch, and P. A. Morse. 2000. Evidence-based cognitive rehabilitation: Recommendations for clinical practice. Archives of Physical Medicine and Rehabilitation 81:1596–1615.
Constantinidou, F., R. D. Thomas, V. L. Scharp, K. M. Laske, M. D. Hammerly, and S. Guitonde. 2005. Effects of categorization training in patients with TBI during post-acute rehabilitation: Preliminary findings. Journal of Head Trauma Rehabilitation 20(2):143–157.
Constantinidou, F., R. D. Thomas, and L. Robinson. 2008. Benefits of categorization training in patients with traumatic brain injury during post-acute rehabilitation: Additional evidence from a randomized controlled trial. Journal of Head Trauma Rehabilitation 23(5):312–328.
Couillet, J., S. Soury, G. Lebornec, S. Asloun, P. A. Joseph, J. M. Mazaux, and P. Azouvi. 2010. Rehabilitation of divided attention after severe traumatic brain injury: A randomised trial. Neuropsychological Rehabilitation 20(3):321–339.
Dawson, D. R., A. Gaya, A. Hunt, B. Levine, C. Lemsky, and H. J. Polatajko. 2009. Using the Cognitive Orientation to Occupational Performance (CO-OP) with adults with executive dysfunction following traumatic brain injury. Canadian Journal of Occupational Therapy/Revue Canadienne D’Ergotherapie 76(2):115–127.
Delazer, M., T. Bodner, and T. Benke. 1998. Rehabilitation of arithmetical text problem solving. Neuropsychological Rehabilitation 8(4):401–412.
Ehlhardt, L. A., M. M. Sohlberg, A. Glang, and R. Albin. 2005. TEACH-M: A pilot study evaluating an instructional sequence for persons with impaired memory and executive functions. Brain Injury 19(8):569–583.
Evans, J. J., E. Greenfield, B. A. Wilson, and A. Bateman. 2009. Walking and talking therapy: Improving cognitive-motor dual-tasking in neurological illness. Journal of the International Neuropsychological Society 15:112–120.
Fasotti, L., F. Kovacs, P. A. T. M. Eling, and W. H. Brouwer. 2000. Time pressure management as a compensatory strategy training after closed head injury. Neuropsychological Rehabilitation 10(1):47–65.
Fish, J., J. J. Evans, M. Nimmo, E. Martin, D. Kersel, A. Bateman, B. A. Wilson, and T. Manly. 2007. Rehabilitation of executive dysfunction following brain injury: “Content free” cueing improves everyday prospective memory performance. Neuropsychologia 45(6):1318–1330.
Fong, K. N. K., and D. R. Howie. 2009. Effects of an explicit problem-solving skills training program using a metacomponential approach for outpatients with acquired brain injury. American Journal of Occupational Therapy 63(5):525–534.
Goverover, Y., M. V. Johnston, J. Toglia, and J. Deluca. 2007. Treatment to improve self-awareness in persons with acquired brain injury. Brain Injury 21(9):913–923.
Hewitt, J., J. J. Evans, and B. Dritschel. 2006. Theory driven rehabilitation of executive functioning: Improving planning skills in people with traumatic brain injury through the use of an autobiographical episodic memory cueing procedure. Neuropsychologia 44(8):1468–1474.
Kennedy, M. R. T., C. Coelho, L. Turkstra, M. Ylvisaker, M. M. Sohlberg, K. Yorston, H.-H. Chiou, and P-F. Kan. 2008. Intervention for executive functions after traumatic brain injury: A systematic review, meta-analysis and clinical recommendations. Neuropsychological Rehabilitation 18(3):257–299.
Levine, B., I. H. Robertson, L. Clare, G. Carter, J. Hong, B. A. Wilson, J. Duncan, and D. T. Stuss. 2000. Rehabilitation of executive functioning: An experimental-clinical validation of Goal Management Training. Journal of the International Neuropsychological Society 6:299–312.
Man, W. D., W. Y. L. Soong, S. F. Tam, and C. W. Y. Hui-Chan. 2006. Development and evaluation of a pictorial-based analogical problem-solving programme for people with traumatic brain injury. Brain Injury 20(9):981–990.
Manly, T., K. Hawkins, J. Evans, K. Woldt, and I. H. Robertson. 2002. Rehabilitation of executive function: Facilitation of effective goal management on complex tasks using periodic auditory alerts. Neuropsychologia 40(3):271–281.
Marshall, R. C., C. M. Karow, C. A. Morelli, K. K. Iden, J. Dixon, and T. B. Cranfill. 2004. Effects of interactive strategy modelling training on problem-solving by persons with traumatic brain injury. Aphasiology 18(8):659–673.
Nott, M. T., C. Chapparo, and R. Heard. 2008. Effective occupational therapy intervention with adults demonstrating agitation during post-traumatic amnesia. Brain Injury 22(9):669–683.
Rath, J. F., D. Simon, D. M. Langenbahn, R. L. Sherr, and L. Diller. 2003. Group treatment of problem-solving deficits in outpatients with traumatic brain injury: A randomised outcome study. Neuropsychological Rehabilitation 13(4):461–488.
Serino, A., E. Ciaramelli, A. D. Santantonio, S. Malagu, F. Servadei, and E. Ladavas. 2007. A pilot study for rehabilitation of central executive deficits after traumatic brain injury. Brain Injury 21(1):11–19.
Sohlberg, M. M., C. A. Mateer, L. Penkman, A. Glang, and B. Todis. 1998. Awareness intervention: Who needs it? Journal of Head Trauma Rehabilitation 13(5):62–78.
Stuss, D. 1991. Disturbance of self-awareness after frontal system damage. In Awareness of Deficit After Brain Injury: Clinical and Theoretical Issues. Edited by G. P. Prigatano and D. L. Schacter. Oxford: Oxford University Press. Pp. 63–83.
Toglia, J., M. V. Johnston, Y. Goverover, and B. Dain. 2010. A multicontext approach to promoting transfer of strategy use and self regulation after brain injury: An exploratory study. Brain Injury 24(4):664–677.
Vallat-Azouvi, C., P. Pradat-Diehl, and P. Azouvi. 2009. Rehabilitation of the central executive of working memory after severe traumatic brain injury: Two single-case studies. Brain Injury 23(6):585–594.
Webb, P. M., and R. L. Glueckauf. 1994. The effects of direct involvement in goal setting on rehabilitation outcome for persons with traumatic brain injuries. Rehabilitation Psychology 39(3):179–188.
Zencius, A., M. Wesolowski, and I. Rodriguez. 1998. Improving orientation in head injured adults by repeated practice, multi-sensory input and peer participation. Brain Injury 12(1):53–61.