Keywords

Impaired Self-Awareness After Traumatic Brain Injury

Incidence and Course of Recovery

All persons with moderate or severe traumatic brain injury (TBI) have impaired self-awareness (ISA) early after injury while confused. The confused state which is characterized by disorientation, cognitive impairment, restlessness, fluctuation in cognitive and other neurobehavioral functions, and other deficits [1] is not compatible with the ability to form accurate self-perceptions. Even after resolution of confusion, a number of studies have shown that patients in early recovery from TBI rate themselves as having less impairment than do treating clinicians or family members/significant others indicating some degree of ISA [2–4]. In a mixed sample of patients with TBI and aneurysm rupture, 97 % rated themselves as less impaired than they were rated by treating clinicians [5]. Remarkably, over 25 % of persons in early recovery from moderate and severe TBI actually rate themselves as having better cognitive and other neurobehavioral skills than they did prior to their injuries, clearly indicating substantial ISA [4].

Severity of ISA varies by the area assessed. Studies consistently show greater ISA for cognitive and behavioral functioning as compared to physical functioning [3, 6]. Overall degree of ISA is associated with injury severity so that persons with more severe injuries have greater ISA [4].

As for other neurobehavioral deficits, ISA shows a recovering course [2, 7]. In a study of 123 persons with moderate and severe TBI who were assessed an average of 45 days post-injury and again at 1 year post-injury, Hart and colleagues [8] found significant improvement over time for overall self-awareness as well as awareness for cognitive and behavior/affective functioning. Awareness for motor/sensory functioning did not improve over this time interval, but unawareness in this area was minimal at baseline assessment. In spite of the improvement over time, persons with injury still rated themselves as less impaired at 1 year post-injury than did caregivers for overall functioning, cognitive functioning, and behavioral affective functioning. Patients selected for treatment in post-acute community integration programs may be especially likely to have significant ISA as persons with relatively mild cognitive impairment but who are having poor outcomes due to ISA may be attractive candidates for such programs. In a study of 66 persons with TBI treated in a post-acute community integration program, Sherer and colleagues [3] found that 97 % had some degree of impairment of self-awareness.

Neural Substrate

Investigation of self-reflection in non-injured persons has indicated that the anterior medial prefrontal cortex and the posterior cingulate have increased activation during tasks requiring self-awareness [9]. In an fMRI study comparing typically developing adolescents to adolescents with TBI, Newsome and colleagues [10] found that injured adolescents showed greater activation of more posterior brain areas such as the cuneus, lingual gyrus, and parahyppocampal gyrus when performing a task requiring judgments about the self. While injured adolescents did not show greater activation of the anterior medial prefrontal cortex, they did show increased activation of posterior cingulate white matter. The authors interpreted these findings as indicating that disruption of prefrontal connectivity resulted in greater dependence on posterior regions for this task. This interpretation is consistent with the notion that self-awareness depends on a widespread neural network requiring integration of multiple brain areas. Diffuse axonal injury (DAI) is a key aspect of the neuropathology of TBI [11]. Widespread DAI results in partial disconnection of key cortical regions that normally function in an integrated manner to support complex neurobehavioral functions. Thus, it is to be expected that many persons with injury will show ISA even if they do not have focal contusions to areas such as the medial prefrontal and posterior cingulated cortices that are primarily responsible for self-awareness. Preliminary evidence [12] indicates that DAI is a key finding in injured persons with ISA.

Association with Cognitive Impairment and Other Neurobehavioral Deficits

As noted above, severity of ISA is associated with overall TBI injury severity and thus ISA tends to be associated with other neurobehavioral deficits. While findings are inconsistent, there is some evidence that persons with greater overall cognitive impairment have greater ISA after TBI [13]. Based on the importance of frontal brain regions for self-awareness, there has been interest in the association between ISA and executive function deficits. Findings have also been somewhat inconsistent in this arena, but investigations have generally found a positive association between degree of executive function deficit and degree of ISA [14].

There is growing evidence of a particular association of ISA with deficits in social cognition after TBI. Social cognition is the ability to understand the behavior of others and react appropriately in social situations [15]. As with ISA, impairments of social cognition have been associated with medial prefrontal lesions, cingulate lesions, and white matter lesions [16]. Also, as with ISA, there is some evidence that social cognition abilities are associated with execution functions [17]. Based on these findings, an association between ISA and social cognition is expected. Indeed, self-awareness of some attributes such as social skills and attractiveness only has meaning in a social context. An investigation of ISA and social cognition found that persons with greater impairment of social cognition showed greater impairment of self-awareness [18]. This finding suggests that impaired ability to judge the emotions and thoughts of others may deprive persons with injury of important feedback needed to form accurate judgments of their abilities and their social impact on others. Family members and friends often label the changes in social interaction style due to ISA and impaired social cognition as personality change.

Impact on Rehabilitation and Functional Outcome

Much of the interest in ISA after TBI is driven by the impact that ISA has on the rehabilitation process and long-term rehabilitation outcomes. For persons in early recovery from TBI, the key impact of ISA is decreased compliance with treatment and failure to observe safety precautions [19]. In addition, patients with more severe ISA require a greater intensity of service, and, thus, greater cost to achieve outcomes similar to patients with less severe ISA [20].

ISA has been associated with increased distress for caregivers of persons with TBI [21]. Since many persons with moderate and severe TBI will have an increased need for caregiver support for an extended period, if not permanently, after injury, preservation of caregiver mental health is a key issue. Caregiver burnout poses a threat to quality of life and community integration for persons with TBI [22].

Of greatest concern, patients with high levels of ISA have been shown to have poor functional outcomes. Sherer and colleagues [4] showed that at discharge from inpatient rehabilitation at a median of 42 days post-injury, patients with high ISA had only half the odds of having a favorable outcome as patients with more accurate self-awareness. This relationship obtained even after adjustment for age, years of education, injury severity, and functional status at admission for rehabilitation. Additional analysis of these data [23] indicated that while virtually every patient showed some degree of ISA, low levels of ISA were not associated with increased risk of poor outcome. With regard to long-term community integration outcomes after TBI, Sherer and colleagues [24] found that, at about 2.5 years post-injury, patients with high ISA had only one quarter the odds of being employed as those with more accurate self-awareness. These results obtained even after adjustment for injury severity, time since injury, degree of cognitive impairment, and pre-injury employment status. It should be noted that participants in this study were clients in a post-acute brain injury rehabilitation program and, due to admission criteria for such programs, the degree to which this finding would generalize to the larger population of persons with TBI is not known.

Measurement of ISA

General Approaches

By pure definition, self-awareness is a difficult cognitive function to measure in the individual patient due to its subjective nature. Measurement involves evaluating the extent to which an individual is able to objectively recognize limitations arising from TBI while at the same time appreciating their subjective significance [25]. Rather than being assessed directly, the level of impairment of self-awareness must be inferred from either the patient’s self-report of his/her abilities and limitations, or from observation of some aspect of his/her behavior [26].

A distinction can be made between methods best suited to assessing different components of self-awareness. Metacognitive/intellectual awareness lends itself to assessment by knowledge-based methods such as questionnaires or interviews, and on-line awareness (emergent/anticipatory awareness) is more appropriately assessed by performance-based methods [27]. On-line awareness (i.e., the ability to monitor and modify behavior during actual performance) is best assessed by observation of task performance, for example counting the number of self-corrected errors. A number of authors have reviewed the topic of assessment of self-awareness in neurological rehabilitation [26–29] and have broadly identified three main approaches to assessment: These are the discrepancy method, clinical ratings, and observation of behavior on functional activities. Each of these approaches to the measurement of ISA is described below.

Discrepancy Method

Metacognitive awareness or self-knowledge is most commonly measured by comparing the patient’s self-report of his or her perceived functional abilities with the report or opinion of another, presumably more objective, source of information. This informant may be a significant other, such as a close relative or friend, or may be a therapist or staff member. Self-ratings of performance may also be compared with actual performance or test results. The most common method is to calculate the discrepancy between the patient’s self-ratings and the informant’s ratings on a questionnaire to give an indication of the direction and magnitude of self-awareness impairment. For example, when the informant score is subtracted from the patient score, a positive discrepancy score indicates that the patient overestimates his or her level of ability compared to the informant, or has an impairment of self-awareness. A zero or near zero discrepancy score indicates no impairment of self-awareness, and a negative score indicates the patient underestimates his or her level of ability compared to informants.

The first self-awareness questionnaire of this type to be developed for use with the TBI population was the Patient Competency Rating Scale (PCRS) [30]. The PCRS is a 30-item questionnaire which uses a 5-point Likert scale to rate the ease with which the patient is able to perform behavioral tasks including activities of daily living (e.g., how much of a problem do I have in preparing my own meals?), cognitive skills (e.g., how much of a problem do I have in remembering names of people I see often?), interpersonal skills (e.g., how much of a problem do I have in recognizing when something I say or do has upset someone else?), and emotional status (e.g., how much of a problem do I have in keeping from being depressed). The PRCS is available in both patient and significant other versions and self-awareness scores are usually calculated using the discrepancy method described above. Other less sensitive methods of scoring include comparing the average perceived competency score across all items (range = 1–5) for patients and informants (e.g., [31]) or classifying patients into three groups based on whether the highest number of items have patient self-ratings greater to, equal to, or less than the informant ratings [32]. Another approach has been to calculate the mean PCRS difference score on individual items [33]. The PCRS has a demonstrated test–retest reliability for the patient (r = 0.97) and informant versions (r = 0.92) [6] and inter-rater reliability for staff ratings (average r = 0.92) [31]. Borgaro and Prigatano [34] modified the PRCS to a 19-item scale for use in acute neurorehabilitation and demonstrated that the PCRS-NR had three psychometrically sound factors relating to emotional, interpersonal and cognitive functioning.

A range of other questionnaires have been developed to assess self-awareness in the TBI population using the discrepancy method. Examples of these include the Awareness Questionnaire [35], the Functional Self-Assessment Scale [36], the Head Injury Behaviour Scale [7], and the Dysexecutive Questionnaire (DEX) from the Behavioral Assessment of the Dysexecutive Syndrome (BADS) [37]. Other self-report questionnaires which have not been specifically designed for measuring awareness have been adapted for this purpose by creating a significant other version of the questionnaire and using the discrepancy method to compare it with self-ratings. For example, this method has been used with the Mayo-Portland Adaptability Index and the Sickness Impact Profile in self-awareness studies [38, 39], and to determine self-awareness for specific cognitive functions such as memory performance [40, 41].

The Awareness Questionnaire (AQ) developed by Sherer and colleagues [35] has emerged as one of the mostly widely used self-awareness questionnaires and is provided in Fig. 1 as an example of the discrepancy approach. The AQ has three versions (patient, clinician and family member) each consisting of 17 items which rate the patient’s functioning following TBI compared to pre-injury on a 5-point scale where 1 = much worse and 5 = much better. Total scores range from 17 to 85 with scores of 51 indicating the patient’s functioning is similar to pre-injury. Three subscales with strong internal consistency have been demonstrated using factor analysis—motor and sensory (four items), cognitive (seven items), and behavioral and affective (six items) [35]. Discrepancy scores are generated by subtracting clinician or family member ratings from patient self-ratings and can range from −68 to +68. The AQ has been used in numerous studies of self-awareness after TBI demonstrating its validity, for example relationship to long-term employment outcome [24]. In one study, the AQ showed only moderate correlations with the PCRS but both measures performed comparably in predicting employability at discharge [4]. Based on these results, preliminary cut-off scores were proposed with AQ discrepancy scores of <20 indicating mild or no impairment of self-awareness, 20–29 indicating moderate impairment, and >29 severe impairment of self-awareness [4].

Fig. 1
figure 1

Awareness questionnaire. Modified with permission from Sherer M, Bergloff P, Boake C, High W, Levin E. The Awareness Questionnaire: Factor structure and internal consistency. Brain Injury 1998;12:63–68

Full size image

Research using the discrepancy method to evaluate impairment of self-awareness has generally demonstrated that the majority of patients underestimate their impairments or limitations compared to ratings by family or staff [5]. However the accuracy with which patients’ self-ratings reflect their actual perceived brain injury-related limitations may be questioned as it can be influenced by a number of factors [42]. These include the contribution of psychological factors such as denial of disability, the desire for a favorable presentation of self, the degree of willingness to engage in self-disclosure, and caution about how the information may be used in clinical decision making. For example, a client with TBI may be concerned that disclosing details of difficulties may delay discharge or clearance to return to work or driving [43].

A second difficulty with assessing impairment of self-awareness is establishing an objective measure of functional competency or limitations against which to compare the patient’s self-report [42]. The reports of family members may be biased as a result of denial or unawareness of the extent of disability in the early stages following TBI, especially while the patient is still in hospital, and at later stages post-injury the emotional stress, strain and fatigue associated with caring for a person with TBI may lead family members to overestimate the extent of disability [39]. While an obvious solution may be the reliance on clinicians’ ratings as a more objective source of information, these too may be limited by a lack of knowledge of the patient’s premorbid personality and abilities, as well as limited exposure to the patient’s performance in real-life environments [42].

Some studies have overcome the difficulties with the accuracy and objectivity of informants’ reports by using comparison with neuropsychological test performance. For example, Allen and Ruff [44] used a questionnaire to evaluate the self-awareness of patients with TBI and controls in the areas of sensorimotor function, attention, mathematics, language and reasoning, learning and memory, and reasoning. Self-ratings were compared with performance on neuropsychological testing to determine the level of self-awareness. Similarly, in a metamemory study, Livengood and colleagues [40] used comparison of performance on memory assessments with patient predictions of performance to measure the level of self-awareness. Self-ratings can also be compared to performance on functional tasks. For example, in acquired brain injury rehabilitation, the Assessment of Awareness of Disability (AAD) can be used following performance of activities of daily living to compare patient’s self-ratings of motor and process skills with therapist’s ratings [45].

A final issue raised with respect to using the discrepancy method for measuring self-awareness is the magnitude of the difference score required as a cut-off for identifying impairment of self-awareness [39]. As highlighted with the PRCS above, several different approaches can be applied to determine a discrepancy score, and the approach is likely to influence the number of patients who are identified as having an impairment of self-awareness [39]. The magnitude of discrepancy scores is also restricted if the informant rates the patient as being fully competent, thereby allowing little room for discrepancies in ratings to be achieved [4, 46].

Clinician Rating

The clinician rating method of assessing the level of self-awareness in patients with TBI relies upon clinical judgment to determine the extent of impairment of self-awareness using some form of rating scale. In essence this approach is not significantly different from the discrepancy approach, except that instead of informant ratings or test scores, the clinician uses his or her own knowledge of the patient’s performance for comparison with self-reports. This method is therefore reliant upon the clinician’s judgment and his or her ability not to be influenced by personal characteristics of the patient such as their likeability, attractiveness, and communication skills [29].

Clinician ratings of self-awareness are generally based on the patient’s responses to a structured interviewed, and there are several interview-based assessments including the Self-Awareness of Deficits Interview (SADI) [42], the Self-Regulation Skills Interview (SRSI) [47], and the Awareness Interview [13]. The SADI is provided in Table 1 as an illustration of the use of a clinician-rated structured interview for measuring metacognitive knowledge or intellectual awareness. The SADI measures the level of self-awareness on three subscales: (1) self-awareness of impairment, (2) self-awareness of functional implications, and (3) ability to set realistic goals [42]. The questions on the SADI build upon previous interview formats used in psychiatry [48] and social cognition research [49]. The patient's responses are transcribed verbatim by the interviewer during the interviewer, or alternatively, interviews may be audiotaped. The responses are rated on a 4-point scale similar in format to the scale used by Bisiach et al. [50] for rating anosognosia for hemianopia, but adapted to cover the range of impairments possible following TBI. On each dimension, a score of 0 indicates no disorder of self-awareness and a score of 3 indicates a severe disorder of self-awareness, giving a total possible range of scores from 0 to 9. Detailed scoring guidelines are provided.

Table 1 Self-awareness of deficits interviewa
Full size table

In designing the rating scale a number of points were taken into consideration. First, patients with TBI may display “borderline” awareness in which they acknowledge certain impairments (notably physical limitations), and ignore others (such as cognitive and personality changes), or they can describe problems that others have noticed but they are not convinced themselves that they exist [32, 51]. Second, understanding the functional implications of impairments may be limited by a lack of opportunity to try various tasks in the acute post-injury phase. Third, realistic goal setting is seen to reflect the degree of self-awareness [52], with the adjustment of pre-injury goals seen as an important step in the development of self-awareness after TBI [51]. Finally, in scoring an individual's responses on the rating scale, the interviewer needs some background knowledge of the patient's level of function. Therefore, a relative’s and/or clinician checklist can be used to gather collateral information to assist with assigning SADI scores. A full version of the SADI and the checklist are available from the authors.

An initial inter-rater reliability study indicated acceptable agreement between raters for total scores with an intraclass correlation coefficient (ICC) of 0.82 [42]. A second inter-rater reliability study where two raters were both present during the actual interviews yielded a higher ICC of 0.85 [53]. Test–retest reliability over a 2- to 4-week period was high (ICC = 0.92) [54]. The SADI has been significantly correlated with measures of frontal lobe functioning and injury severity [55], and with the AQ and measures of work status in individuals with acquired brain injury [56].

In contrast to the SADI which measures metacognitive awareness, the SRSI [47] was designed to measure on-line awareness skills in relation to a main area of difficulty identified by the patient. The SRSI has six items (emergent awareness, anticipatory awareness, readiness to change, strategy generation, degree of strategy use, and strategy use) which are scored by the interviewer using a 10-point rating scale. The items are grouped into three indices derived by factor analysis which include Awareness, Readiness to Change, and Strategy Behaviour Index [47]. The SRSI has test–retest (0.81–0.92) and inter-rater (0.69–0.91) reliability [47] and was significantly correlated with SADI scores and work status in adults with acquired brain injury [56].

The Awareness Interview [13] involves the use of both a clinician-rated interview and the discrepancy method by comparing neuropsychological test scores to patient responses in order to quantify self-awareness. Interview questions address the five specific areas of motor, intellectual, orientation, memory, speech or language, and visual perceptual impairment. Interview responses are scored on a 3-point scale to reflect the perceived amount of impairment on each function. These scores are then compared to ratings by a neuropsychologist on a comparable scale based on the results of neuropsychological testing of the same functions. Deviation scores are generated by comparing the two sets of scores to give scores ranging from 0 = no discrepancy to 2 = maximum discrepancy. Two additional questions (regarding awareness of the reason for hospitalization and awareness of general test performance and ability to resume normal activities) are scored using deviation scores which are then totaled with the other six deviations scores to give an Awareness Index (range 0–16). The Awareness Index therefore represents a more objective way of using clinician-rated interviews to measure self-awareness, but is related to only a limited spectrum of impairments seen following TBI and does not include items relating to executive dysfunction, and interpersonal, behavioral and emotional changes for which people with TBI often lack self-awareness.

Clinician ratings are also used to give scores on the ISA and Denial of Disability (DD) Clinicians’ Rating Scale [57]. This scale was designed to differentiate between ISA of neurological origin and DD of psychological origin in the individual patient with TBI. Both the ISA and DD scale consist of ten items which are rated as “yes” or “no” by the clinician with items rated “yes” then scored on a 0–10 scale of severity giving a maximum possible total score of 100 on each scale. Items on the ISA scale relate to a lack of spontaneous reports of difficulties, little affective reaction to feedback, “cognitive perplexity” in response to feedback or difficulties, other higher level cognitive problems such as impaired initiation, self-monitoring and planning. In contrast items on the DD include some minimal admission of difficulties, a negative affective reaction to feedback, use of arguments or excuses to explain behavior, a lack of severe impairment in initiation, planning, and self-monitoring on testing, and possible catastrophic reactions when faced with failure. Inter-rater reliability on the scales was generally high except when the clinician rated the degree of impaired self-awareness in the ISA group [57].

Observation of Behavior

The observation of task performance or behavior provides a third method of assessment of self-awareness, which particularly targets on-line awareness. Observational methods do not usually involve patient self-reports but focus on the patient’s task selection and avoidance, error detection, and error correction during task performance [26]. In one of the first studies of online awareness after TBI, Hart and colleagues [58] engaged participants in a naturalistic multi-level action task that involved making toast, wrapping a gift, and packing a lunchbox. Performances were videotaped and then analyzed to record instances of error correction (i.e., attempts to redress an error) and error detection (e.g., verbalizations, exclamations, facial expressions, and manual gestures signifying the participant’s awareness that an error had occurred). This study showed that aspects of on-line awareness could be reliably and objectively measured without reliance on self-report.

Ownsworth and colleagues [59] used a similar behavioral approach to measuring on-line awareness with a TBI participant during meal preparation and work activities. This included measures of error frequency (i.e., recording errors that compromised safety, outcome or time efficiency) and error behavior. Error behavior was systematically observed using a “pause, prompt, praise” technique which involved the therapist initially allowing a “pause” following an error to allow for self-correction, then a non-specific prompt, followed by a specific prompt if the error was not corrected. Errors were categorized as self-corrected, corrected with non-specific prompt, or corrected with specific prompt. This approach was adapted in a subsequent study [60] to classify errors as self-corrected errors (i.e., corrected after a 5–10 s pause) or therapist-corrected errors (i.e., corrected after a prompt) and checks (i.e., requests for advice or verification). In both studies, inter-rater reliability for frequency and classification of errors was established.

In summary, the standardized assessment of on-line awareness remains an under-developed area. It can be concluded that self-awareness can be measured in a number of ways which tap into different aspects of self-awareness. Therefore for any individual patient, it is advisable to use more than one approach to establish an understanding of his or her level of self-awareness across domains.

Empirical Studies of Interventions

General Approaches

The past two decades have seen an increasing emphasis on the development of interventions specifically targeting impairment of self-awareness after TBI, or incorporating self-awareness training into other cognitive rehabilitation approaches. A small but growing body of empirical studies provides evidence that self-awareness interventions can be effective in enhancing rehabilitation outcomes. Consequently, a review of the latest evidence concluded that metacognitive strategy training (i.e., targeting self-monitoring and self-regulation) should be a practice standard in the cognitive rehabilitation of people with executive dysfunction following TBI [61]. The treatment of ISA after TBI has been reviewed previously by several authors [26, 51, 62–64] and interventions described include feedback, education, behavior therapy, psychotherapy, milieu-oriented programs, game formats, group programs, and real world experiences. This section reviews the research relating to self-awareness interventions including the use of various approaches to providing feedback, predicted performance, occupation-based experiences, and the use of other techniques such as psychotherapy, education, and group programs.

Feedback Approaches

The provision of feedback to patients is a fundamental component of rehabilitation and includes feedback on test results, functional performance, and strengths and limitations. Arguably the primary rationale for providing feedback is to improve a patient’s self-awareness thereby enabling him or her to identify areas for improvement, or the need for strategies to improve performance. Timely, specific and consistent feedback is emphasized as being an important component of all awareness interventions [51, 65]. A systematic review of intervention studies which used a feedback component to improve self-awareness identified 12 studies of varied methodological quality including single case experimental designs [66]. Three studies were randomized controlled trials involving a total of 62 people with brain injury of mixed etiology [67–69]. A meta-analysis found a moderate effect size for the pooled estimate of improvement in self-awareness after completing a feedback intervention (Hedges adjusted g = 0.64, 95 % confidence interval 0.11–1.16). Furthermore, feedback interventions had large effect sizes for improving functional task performance and patient satisfaction with therapy [66]. Most feedback interventions use a combination of different forms of feedback including self-predictions and self-evaluations of performance, verbal feedback from a therapist, and in some cases, videotaped feedback. Peer feedback is also an important ingredient of group interventions.

Direct therapist feedback is a common approach to facilitating both intellectual and on-line awareness in patients with TBI [29, 51, 65]. Verbal feedback on performance is thought to be more readily accepted by the patient if provided by a trusted therapist in the context of a strong therapeutic alliance [70]. A “sandwich technique” in which negative feedback is preceded, and followed, by positive feedback is recommended [63, 71]. Klonoff and colleagues [70] described well-timed therapist feedback as an integral component of their cognitive retraining program which led to successful work placement in a case study of a patient with TBI.

While some authors have claimed that direct therapist feedback may be too confrontational and force patients to defend their confabulatory beliefs [72], research has demonstrated that, on the contrary, feedback of self-awareness assessment data led to a decrease in subjective reports of grief in participants with brain injury [73]. Another repeated measures study found that feedback from a consultant neurologist on the findings of brain scans and possible neurobehavioral outcome led to significant improvements in self-awareness as measured by the AQ and SADI in 17 patients with brain injury [74]. Interestingly, in this study, direct feedback was also associated with a decrease in self-reported symptoms of anxiety and depression.

Videotaped feedback has been recommended by several authors as an effective method for improving self-awareness of functional performance including awareness of behavioral and communication problems [62, 64, 75, 76]. Videotaped feedback has been used as an element of several self-awareness interventions with people with TBI (e.g., [59, 77]) and demonstrated to be more effective than verbal feedback and experiential feedback in a randomized controlled trial with 54 participants with TBI [78]. Schmidt et al. [78] used feedback on performance in a meal preparation task on four occasions over a 2-week period. The group that received a combination of video and verbal feedback had significantly greater gains in on-line awareness as measured by an error count and intellectual awareness measured by AQ discrepancy score. Interestingly, there were no changes in the level of emotional distress associated with any of the feedback interventions. McGraw-Hunter et al. [79] also described the use of video self-modeling to teach cooking skills to four individuals with TBI. In this study, the participants were videoed performing the cooking task with step by step direction from the researcher and any errors were edited from the video. In the experimental phase, the participants viewed the video prior to performing the cooking task with a graduated system of prompting, praise and corrective feedback. The approach was effective in achieving skill acquisition for three of the four participants within four training sessions, however the effect on self-awareness was not examined.

Predicted Performance

The technique of predicted performance involves asking patients to self-predict their performance prior to completing a task (e.g., the amount of difficulty expected, how much assistance will be needed, how long it will take, or the need for strategies). Following task performance, patients complete a self-evaluation in which they rate their performance of the task, which can then be compared with the predicted performance as well as with therapist feedback on the performance [68]. While predicted performance is most commonly used in conjunction with the performance of functional activities, it has also been used successfully to improve awareness of performance on memory tasks in single case studies [80, 81] and for verbal recall and arithmetic tasks in a single group session [82].

Occupation-Based Interventions

Experiential feedback or occupation-based interventions involve participation in real-life activities that allow the person with TBI to discover his or her own errors [71]. The selection of meaningful occupations to improve self-awareness contains elements of “supported risk taking” or “planned failure” and needs to be well-structured and supported by the therapist to minimize any emotional distress [51, 65, 83]. Even without specific intervention targeting self-awareness, individuals with TBI have reported that their self-awareness developed as a result of comparing their current ability to perform familiar occupations with their pre-injury status [83]. An occupation-based approach acknowledges that self-awareness training involves “rebuilding a sense of self” [76, p. 181]; while engaging in structured occupational experiences, the person with TBI uses self-monitoring techniques to discover strengths and weaknesses, and to develop strategies and new ways of doing things, thereby promoting self-efficacy.

Toglia [76] described a dynamic interactional approach to improving self-awareness using engagement in meaningful occupations. In the pre-activity phase, this approach involves the use of techniques such as self-prediction, guided anticipation of challenges, and strategy generation. During the activity, techniques include “stop and check” periods, self-questioning (e.g., Am I keeping track of everything?) and therapist feedback to reinforce strategy use. The post-activity phase can include various forms of self-assessment such as video feedback, self-ratings, guided questioning, and comparison of outcomes with a template or model. Journaling or structured logs may also be used to promote self-reflection on the activity, as well as broader identification of cognitive failures and strategies in daily life. Role reversal is another technique developed by Toglia for promoting self-awareness using functional activities. In role reversal, the therapist performs the activity and deliberately makes errors, while the patient observes the performance and identifies errors and suggests strategies, an approach which may be less cognitively demanding and less threatening than identifying errors during one’s own performance.

Several studies have used single case study designs to illustrate the effectiveness of occupation-based approaches using many of the above techniques to improve self-awareness after TBI [59, 77, 84, 85]. Usually occupation-based interventions are used in conjunction with other techniques such as feedback to promote self-awareness. The three randomized controlled trials in the Schmidt et al. [66] systematic review described above each involved occupation-based interventions.

In the first of these, Cheng and Man [67] used a combination of education, experiential feedback, self-prediction and goal setting to significantly improve intellectual awareness as measured by the SADI in 11 intervention participants compared to a control group. In the second study, Goverover et al. [68] used a combination of occupation-based intervention and predicted performance with ten participants with TBI when performing instrumental activities of daily living over six therapy sessions. Compared to a control group, the intervention group improved significantly in task performance and self-regulation skills. However, no significant differences were found for task-specific or general self-awareness.

The third study [69] was a randomized controlled trial with three treatment arms, one of which involved individual occupation-based support. Occupational activities were selected on the basis of participants’ goals and performed in their home or community with a focus on self-monitoring and self-correction of errors, and use of self-regulation strategies. After eight weekly intervention sessions participants in this group showed significant improvements in goal attainment, however the level of self-awareness was not explicitly evaluated in the original study. In the subsequent meta-analysis [66], it was found that compared to a wait-list control group, the standardized mean difference of discrepancy scores on the PCRS was not significant for this study by itself (Hedges adjusted g = 0.48, 95 % confidence interval of −0.41 to 1.38).

In another study, Ownsworth et al. [60] used a single case study design to demonstrate improvements in self-regulation skills in two participants with brain injury during meal preparation activities. The metacognitive skills training approach involved a “pause, prompt, praise” approach and incorporated sessions of role reversal, videotaped feedback, and post-task discussion. Specifically, during task performance, the therapist waited (pause) if the participant started to make an error to allow self-correction of the error and provided non-specific direction (prompt) if the error continued. The therapist then affirmed the participant for correct performance (praise). Compared to baseline, both participants showed a significant increase in self-corrected errors, decrease in therapist corrected errors, and decrease in number of times the participant checked to ensure accuracy of task performance. In contrast, another participant who engaged in an extended baseline of behavioral practice without metacognitive skills training showed no significant changes in error correction, and an increase in the number of checks suggesting an increased reliance on therapist support. Interestingly, the participants receiving metacognitive skills training rated themselves lower on the PCRS following the intervention suggesting more accurate self-awareness, whereas the behavioral practice participant perceived greater self-competency following the intervention.

Ownsworth et al. [86] also described the use of a similar metacognitive contextual intervention component as part of a larger program for two individuals with TBI and one with stroke to facilitate achieving paid work after long-term unemployment. Techniques used included self-prediction, self-monitoring, and self-evaluation of performance of functional tasks in the participants’ homes and workplaces. The program also included group education and support activities, family involvement, sessions with disability support counselors, and a work trial with employer education and support, so it not possible to determine which component or components of the intervention were effective. However, all three participants had maintained paid employment 6 months later.

Other Approaches

Other approaches to facilitating self-awareness following TBI include group therapy, education, adjustment counseling, and psychotherapy. Many of the studies pioneering these techniques and others described above have occurred in the context of comprehensive neuropsychological community integration programs [29, 87–89]. These programs employ holistic milieu-oriented approaches which combine cognitive retraining with psychotherapeutic interventions to address both neuropsychological and emotional or adjustment issues, and to enhance community integration outcomes. While the development of self-awareness has been a major focus of the rehabilitation programs, outcomes generally have been reported in terms of better emotional adjustment and higher levels of productivity, rather than improvements in self-awareness per se. A study by Malec and Moessner [88] however, found that 62 graduates of a comprehensive day treatment program had diminished impairment of self-awareness compared to pre-intervention. They also found that improvements in self-awareness and distress levels were associated with positive behavioral changes but not vocational outcomes. Studies such as these illustrate the difficulties associated with determining the effectiveness of specific self-awareness intervention techniques, as in practice they are usually used in combination, as well as the difficulty determining the extent that gains in self-awareness impact upon community integration outcomes.

The use of group programs is an important component of the above comprehensive treatment programs, as it allows for valuable peer feedback, role modeling and support during group discussions [29]. There have, however, been few studies which have specifically examined the use of group interventions to facilitate self-awareness. Ownsworth, McFarland and Young [90] conducted a 16-week group support and psychoeducation program with 21 participants with long-term acquired brain injury. The participants showed significant improvements in levels of on-line self-awareness and strategy use, as well as improved psychosocial function, and gains were maintained at a 6-month follow-up.

Psychotherapy is another integral component of many neuropsychological rehabilitation programs that aim to develop self-awareness after TBI. The aim of psychotherapy is to assist patients to explore feelings of loss and anger [83], and to establish realistic goals and re-establish meaning in life [89]. Both individual and group psychotherapy can been used following TBI [29]. For patients who display denial or minimization of information that is too painful to acknowledge, psychotherapy has been recommended to strengthen their emotional readiness to cope with the rehabilitation process [83]. The successful use of a psychotherapeutic approach following TBI has been demonstrated in a case study [91] and as part of a comprehensive day program [29].

Psychotherapy, along with other counseling approaches to facilitate adjustment and acceptance of disability, emphasize the importance of a strong therapeutic alliance. Establishing trust and providing a safe and accepting environment are at the foundations of successful awareness interventions and enhance engagement in rehabilitation [29, 51, 64]. This process may also include a focus on goal setting, in particular discussing and acknowledging the patient’s personal goals and incorporating them into meaningful therapy goals [52]. Motivational interviewing has been presented as a goal setting approach which may be incorporated into holistic rehabilitation programs to enhance the therapeutic relationship and facilitate self-awareness and acceptance of disability [92].

Education is another fundamental component of many rehabilitation programs which aim to enhance self-awareness. Educational approaches may include written materials, individual information sessions, groups, and game formats, and content may cover brain function, brain impairment, the effects of TBI, and strategy use. Often family members are included in education sessions [51]. It is thought that, by maintaining an informational format in group education sessions, individuals with TBI can discuss typical brain injury problems in a less threatening way, and hear from others who may be prepared to discuss their personal experiences and cognitive difficulties, thereby promoting self-acceptance [29]. The research is unclear, however, as to whether education alone leads to improvements in self-awareness. For example, a randomized controlled trial of a single 30-min educational session for improving knowledge and awareness in children with TBI did not find any significant improvement [93]. Educational board games have also been developed to enhance knowledge and awareness in patients with brain injury [94, 95]. These were designed to provide opportunity for patients to learn about the cognitive and behavioral sequelae of brain injury through repetition of information in a non-threatening way; however, the approach was demonstrated to be effective only for improving general knowledge and not accuracy of self-appraisal [95]. When combined with facilitated discussion before the game, the approach led to improved self-awareness of personal strengths and limitations in three participants with severe TBI [94].

In summary, this section has described a number of intervention approaches to improving self-awareness following TBI and overviewed the empirical studies that have investigated their effectiveness. It is evident that most self-awareness interventions are comprised of more than one technique and are delivered as programs which aim to improve not only self-awareness, but psychosocial functioning more generally. Further research is required to establish which techniques, or combinations or techniques, are most effective, and with what type of patients. Indeed for some individuals with severe cognitive impairment, attempts to improve self-awareness may not succeed, and interventions should focus more on compensation for the impairment in self-awareness (e.g., in order to prevent the individual from engaging in unsafe or risky behaviors) [51] or use behavioral interventions which do not require self-awareness to improve function [72]. For other individuals, interventions may lead to improved self-awareness but this may not necessarily translate into functional gains (e.g. due to increased emotional distress), and further research is needed to determine what type of intervention is best suited to particular awareness sub-types.

Recommendations for Clinical Practice and Case Examples

The Big Picture

As the review above indicates, there are multiple possible approaches to address ISA with the goal of improving treatment compliance and patient outcome. Unfortunately, none of these approaches has extensive empirical support. The current state of knowledge does not support specific clinical guidelines for treatment of ISA. Rather, the clinician must rely on experience and clinical judgment to develop an individualized approach to meet the needs of the patient based on the patient’s clinical status, phase of recovery, social support network, and other considerations. All approaches to treatment will rely on feedback to the person with injury, to some degree, and all approaches will be enhanced if the treating clinician(s) can establish an effective therapeutic alliance with the patient.

Therapeutic, or working, alliance is a construct commonly described in the psychotherapy literature. It is generally defined as having three components: the bond between the clinician and the client, agreement on the means of therapy, and agreement on the goals of therapy [96]. This bond may be based on the therapist’s ability to convey empathy, the client’s confidence in the therapist’s expertise, the therapist’s genuine commitment to facilitating the client’s best interests, and/or other aspects of the relationship. However, this bond is not based on friendship or personal affection. Clinicians with limited training and/or experience in counseling may fail to appreciate this difference. In persons with brain injury, a stronger alliance between the patient and therapist has been shown to result in more accurate self-awareness for the patient [97]. An effective therapeutic alliance can be regarded as a necessary though not sufficient condition for improvement in self-awareness in rehabilitation of patients with TBI.

One key issue that can drive the treatment approach selected is the primary goal of the awareness intervention. If the primary goal is treatment compliance and acceptance of obtainable long-term goals, the approach will be different than if the goal is re-establishment of a positive sense of self after catastrophic injury.

Enhancing Treatment Compliance in a Patient with ISA

In a different context (compliance with wearing a splint), Kuipers and colleagues [98] surveyed rehabilitation clients to determine factors that the clients believed affected their compliance with treatment. These factors were (1) client and therapist collaboration in treatment, (2) client trust in therapist expertise, (3) client understanding of the purpose of treatment, (4) fit of the therapeutic approach with client goals and lifestyle, (5) support from family and friends, and (6) any discomfort, including feelings of stigma, caused by being in treatment. With a little modification, these factors are applicable to interventions to improve self-awareness. Interventions pursued with these issues in mind are likely to maximize therapeutic alliance between the therapist(s) and client. We will describe a clinical case of a person with profound ISA after TBI that illustrates how therapy guided by these principles can facilitate compliance with treatment and lead to a favorable outcome.

Tom was a 25-year-old male who sustained TBI at age 15 in a motor vehicle collision. Records indicated an initial Glasgow Coma Scale score of 7 indicating severe injury. CT imagery indicated a left temporal bone fracture with hemorrhagic contusion of the left temporal lobe, widespread subarachnoid hemorrhage, diffuse brain swelling, and a number of small contusions primarily in the frontal lobes. At initial evaluation, Tom was found to be oriented. His speech was mildly dysarthric and he spoke at a rapid rate causing some problems with intelligibility. Testing showed impaired comprehension for complex language tasks, moderate impairment of verbal memory, decreased dexterity with the right upper extremity, and poor problem solving. Tom’s social presentation was awkward. Comments were occasionally off topic. He answered questions impulsively and frequently interrupted the interviewer.

Four years after his injury, Tom eventually graduated from high school with tremendous support from his widowed mother. He was an only child. Tom stated that his goal was to find employment as an insurance agent. Tom had no history of any paid employment. He had completed previous rehabilitation programs of various types including training to be an upholsterer, but was never placed in a job due to his insistence on working as an insurance agent and the collective judgment of all others involved, including his mother, that there was no chance that he could successfully do this type of work.

Tom was admitted to a comprehensive, postacute brain injury rehabilitation program and received 3.5 h of therapy a day, 4 days a week. Therapy goals were to improve speech intelligibility and social skills, implement use of compensatory strategies to improve memory, and develop a plan for placement in paid work. Tom denied having any difficulties at all with intelligibility, social interactions, or memory and indicated that he already had a plan to work as an insurance agent. The team avoided confronting Tom on any of these issues. To address communication issues, we videotaped Tom interacting with other program clients and reviewed the videotapes with him in individual sessions. Since Tom declined to make any notes or use any memory strategies that he would need to initiate, we developed cue sheets to prompt him through multistep tasks. These cue sheets were small enough to easily fit in Tom’s pocket. To work on developing a plan for job placement, we tried Tom out in various simulated jobs we could create in the associated rehabilitation hospital. Tom assured us that it was pointless to work on any job activity that was not related to becoming an insurance agent, but with coaxing we were able to get him to participate in activities, particularly if these activities gave him the chance to interact with therapists at the hospital. On a few occasions, he made inappropriate remarks to female therapists and we firmly redirected him. He denied doing anything wrong, but because he was motivated to continue this opportunity for interaction, we were able to get him to modify his behavior.

Therapists were always focused on being supportive of Tom. We complimented him any time he spoke more slowly or used one of the cuing sheets. We looked for any opportunity to praise his performances on simulated jobs and generally did not comment on errors. Rather, we simply had him redo the task until it was correctly done. Tom was naturally deferent to male authority figures so the program director always had Tom address him as doctor when they spoke. Even though Tom denied any need for treatment, therapists explained the purpose behind each task on which we worked. While Tom continued to focus on work as an insurance agent, we repeatedly emphasized that we were working to help him achieve his goal of obtaining a job. Early on, we fully explained the rationale behind our approach to Tom’s mother. She was very supportive of our approach with Tom and she had great influence on Tom as she was the only person that Tom interacted with regularly prior to admission to our program. Finally, we normalized all tasks to the greatest extent possible. Staff made a point of speaking more slowly around Tom to show that a slower rate was easier to understand. Staff used cuing lists in front of Tom to help them remember tasks that they were doing. We described all simulated work tasks as preparations to help Tom achieve his eventual goal of working.

After several weeks of working with Tom, we identified work tasks that he could complete and the amount of cuing needed to guide him through these tasks. We also determined that while Tom very much enjoyed interacting with others, such interactions frequently got him off task. By negotiating with the manager of a local restaurant, we were able to place Tom in a job doing custodial work after hours at the restaurant about 10 h a week. We provided onsite job coaching for the first 4 weeks of employment. Tom initially resisted this placement, but we emphasized that this would be his first paid job and that it might eventually lead to another job that was more similar to his eventual goal. Tom’s mother played a key role in convincing Tom to give the job a chance. While Tom always saw the job as transitional, he remained at the placement. He found the paychecks rewarding and began to go on outings without his mother to the movie theater and the local mall, activities that he had never engaged in prior to treatment.

By focusing on compliance as our goal rather than insisting that Tom have the same view of his situation as we did, we were able to make great progress in our work with Tom. We believed that if Tom could become convinced that we were on his side and trying to do the best we could to help him that he would be more likely to comply with our requests. At discharge assessment, we could not demonstrate that Tom’s speech intelligibility, social skills, and/or memory were improved. However, he was consistently using the simple cue cards we created for his work duties.

Facilitating Establishment of a Positive Self-Concept in a Patient with ISA

Formation of an accurate self-perception is a complex cognitive task requiring integration of multiple cognitive systems [99]. Self-awareness requires, among other skills, accurate observations of one’s behaviors and their outcomes, integration of individual behavioral performances into a coherent self-concept, introspection and reconsideration of self perceptions over time and across settings, and assimilation of direct and indirect feedback from social interactions. Achieving more accurate self-awareness is only a preliminary step toward re-establishing a positive sense of self after a catastrophic injury. Pioneers in community re-integration programming for persons with TBI such as Ben-Yishay et al. [100] and Prigatano [101] have viewed psychotherapy as a key aspect of therapy intended to facilitate improved functioning and eventual return to work. Following TBI, injured persons may have a profound sense that something is not right. In the presence of ISA, this vague sense of something being wrong is not connected to an accurate appraisal of the cognitive and behavioral deficits caused by the injury. Rather, the injured person is confused and frustrated [101]. Caregivers and family members will often try to help the injured person by pointing out deficits due to the injury as an explanation for the feeling that things are not right. Often this feedback will simply result in more frustration leading to strain in relationships with the primary social support network which in turn leads to greater feelings of despair and abandonment.

For persons with mild TBI who are distressed after injury, psychotherapy may be the only therapy needed. However, in our experience, the most successful approach to addressing a lost sense of self in patients with significant cognitive and/or behavioral deficits combines therapy to address the deficits, feedback to improve self-awareness, and psychotherapy to address the injury to the self. Prigatano’s model [101] for psychotherapy after TBI focuses on the importance of work (productive activities), love (passion, intimacy, and commitment to others), and play (a sense of the inner playful self). Description of all the intricacies of providing psychotherapy to persons with cognitive and social communication deficits is beyond the scope of this chapter, but we will describe a clinical case to illustrate how cognitive rehabilitation, self-awareness, and psychotherapy interventions can be integrated to facilitate a favorable outcome.

Joan was a 35-year-old divorced female who sustained TBI in a motor vehicle collision. Her Glasgow Coma Scale on admission to the Emergency Department was 10 indicating a moderate injury. Duration of post-traumatic confusion (amnesia) was 8 days. She received 2 weeks of inpatient rehabilitation that addressed cognitive and behavioral deficits as well as her fractured right forearm. Goals included improving functional skills such as dressing, cooking, etc. while she remained with a cast on her right arm. By 3 months post-injury, the cast was removed and she returned to full-time work as a sales clerk in a department store. She immediately noticed difficulty with fatigue and asked to be cut back to 20 h a week from her usual schedule of over 40 h a week.

Cognitive testing revealed mild impairment of verbal memory and mild cognitive slowing on motor and verbal tasks. Performance on motor tasks was influenced by residual weakness in her right hand and arm that was thought to be related to her fracture rather than to her brain injury. Interview with Joan’s sister revealed that she thought that Joan was irritable and distractible since her injury. She described Joan as having a total change in personality. This resulted in conflict when they were together and the sister was spending less time with Joan as a result. Joan was initially reluctant to allow program staff to speak to her work supervisor, but eventually gave permission. The supervisor indicated that Joan had been a top performer prior to her injury frequently receiving bonus checks based on her high sales totals. Initially on return to work, Joan had difficulty recalling procedures for processing coupons and exchanges. Her supervisor was not surprised by this as procedures had changed shortly before Joan had been injured and Joan had been away from work for 3 months. However, Joan became very frustrated when processing coupons and exchanges and she rejected help from others. The supervisor indicated that even though Joan had been short with fellow employees causing some hurt feelings, he still regarded her as one of his best sales clerks.

Since Joan was working 20 h a week, her therapy program was scheduled around her work schedule. She received 3 h of therapy a week working on memory compensation, 2 h a week working on strengthening and increasing flexibility in her right hand and arm, and 1 h of psychotherapy. The cognitive therapist obtained materials from Joan’s job site that described steps in processing purchases in which coupons were used, processing exchanges, and various other work tasks. Joan resisted efforts to review these materials in therapy sessions as she indicated that she knew them perfectly well and that her problems at work had been due to fatigue or computer malfunction. In a process that took three or four sessions, the therapist was able to convince Joan to write down the procedures and then compare her description of the procedures to information provided by her supervisor. Joan was only willing to do this if the therapist did not look at her description, but allowed Joan to make the comparison independently. While Joan silently looked over her work and the work materials, the therapist noticed that Joan added some notes in the margins of her page. The therapist commented that she frequently used cue sheets to help her recall the steps in complex processes. She challenged Joan to see how clever she could be in developing a cue sheet that would allow even the therapist who was inexperienced in these procedures to complete them. Reluctantly, Joan accepted the challenge. Joan’s initial efforts involved a good deal of text description and the therapist worked with Joan to create briefer cuing strategies that emphasized key words and acronyms.

Concurrent with Joan’s work with the cognitive therapist, Joan was doing strengthening and stretching activities for her right hand and arm with an occupational therapist. As her rapport with Joan increased, the therapist engaged Joan in a conversation about the benefits of regular exercise for reducing fatigue and improving cognitive abilities. Joan had been an avid cyclist before her injury, but had not engaged in any regular exercise since her injury. She agreed with the therapist to set a goal of walking an hour day, 4 days a week. The therapist was confident that Joan could easily reach this goal based on the intensity of her past cycling program.

The psychotherapist initially focused on building a therapeutic alliance with Joan. He encouraged Joan to talk about her feelings about the accident and how she felt she had been changed by the accident. He reflected back to her the feelings of despair and confusion. After a couple of sessions he asked Joan to describe herself as she had been before her injury. Key attributes that Joan identified for herself were that she was a high energy person and a tireless worker. She became tearful when describing the overwhelming fatigue that she had felt at times since her injury. Knowing that Joan had started a walking program, the therapist commented on how remarkable it was that she could walk for a full hour in spite of her fatigue. Joan responded that it was easy, that anyone would walk for an hour. She went on to add that she planned to increase her walks to 1½ h starting the next week. A transforming moment in therapy occurred with the therapist asked Joan what she liked to do for fun. She had to give this a good deal of thought before she answered. She mentioned cycling and shopping with her sister and admitted that she had done neither of these things since her injury. With the therapist’s encouragement she committed to ask her sister to go shopping with her the next weekend.

Within 6 weeks of treatment, Joan had perfected her cuing sheets for work and regained normal strength in her right hand. She had reconnected with her sister who had been her primary source of social support since her divorce. The treatment team negotiated a gradual return to fulltime work with Joan’s supervisor with Joan increasing her work hours by 5 h a week based on feedback from her supervisor that her work was satisfactory. Four weeks later Joan was working fulltime and felt that she was making progress in adjusting to the “new” Joan she found herself to be after her injury.

Joan’s treatment addressed the triad of work, love (her relationship her sister), and play (taking the initiative to engage in activities that had been fun for her prior to her injury). Since Joan’s cognitive impairments were relatively mild, she was quickly able to improve her skills in developing self-cuing strategies and she showed good benefit from these strategies. Walking partially replaced cycling as a form of exercise as her physician had restricted her from return to cycling. Joan’s energy level seemed to improve as she increased the duration of her walks. Rather than reassuring Joan that she was not that severely injured and that she would likely continue to improve, the psychotherapist accepted the level of distress that Joan said she was experiencing and looked for opportunities to encourage her to engage in activities that would likely decrease her distress. The coordinated approach taken by the treatment team maximized the social influence that the team could have in modifying Joan’s behavior.

Summary

Research with persons with TBI shows that impairment of accurate self-awareness is common early after injury and in the post-acute period. Nonetheless, the level of self-awareness improves over time. Preliminary findings suggest that the degree of impairment of self-awareness must reach a critical level before it adversely affects outcome. For an important subgroup of persons with TBI, ISA is a key deficit that affects compliance with treatment and outcome. Given the impact of ISA, it is important that therapists working with persons with ISA are familiar with approaches to assessment and intervention. While there are a number of proven approaches to assessment, treatment approaches require additional investigation though it seems clear that any successful treatment will involve provision of feedback and that client acceptance of feedback and compliance with treatment is facilitated by a strong therapeutic alliance between the client and treatment team. When designing interventions to address ISA, it is important to have a unified approach involving the entire treatment team and to tailor the approach to address compliance or self-identity depending on the needs of the client.