Chapter One
Finding the Right Treatment Combinations: Changes in Rehabilitation Over the Past Five Years
Leonard Diller
Abstract
Progress toward finding the right treatment combinations has advanced along a number of fronts in the past 5 years. These include developments in identifying behavioral characteristics at both ends of severity in the recovery from traumatic brain injury. At the most severe end is the application of newer assessment devices, and at the opposite end is the clarification of the definition of minor traumatic brain injury. In the middle range there have been two major developments. First, there has been a proliferation of therapeutic modalities to establish competence in functional settings. Among them are the increase of group methods, the applications of the family coach model as a tool, the use of supported employment, and the introduction of computers for orthotic devices or cognitive aids. Second, there has been a large number of reports on varieties of cognitive remediation. These reports are reviewed with regard to the nature of outcomes that are achieved and their experimental designs. Along with the increase and diversity of procedures, there is a reemphasis on psychological constructs related to ego psychology such as awareness and self-efficacy as relevant modulating variables in facilitating response to treatment.
The ideal situation is one in which a patient with a known condition is treated with a known intervention toward a known outcome. In practice, we treat patients with partial knowledge of their conditions, with partial knowledge of the effects of interventions, and with imperfect knowledge of outcomes. We are still far from the ideal situation. Progress over the past 5 years has been slow for those who are impatient, but significant because there now exist more publications and conferences that are peer reviewed to address more precise questions and evaluate procedures more specifically and critically. There have been several major trends including (a) defining patient characteristics in a more salient way, (b) identifying newer treatment modalities, and (c) critical discussing the efficacy of cognitive remediation and clarifying factors, which may modulate responsiveness to treatment. Finding the right treatment combination is more difficult in the face of much diversity, but also more challenging in the light of opportunities offered by newer initiatives. In effect, interventions in psychological modalities must resonate with advances in neurophysiological developments to define biological parameters of the conditions being treated and with developments in service delivery in order to make interventions worthwhile. It also builds on procedures that are being developed with clinical experience.
Defining Populations More Precisely
In the field of traumatic brain injury (TBI), one major bottleneck in undertaking major clinical trials to assess treatments in fields as diverse as carotid artery surgery or management of depression on an outpatient basis has been a lack of agreement on a proper typology. Severity of TBI as a critical typology was supported by the emergence of the Glasgow Coma Scale (GCS). If there is a continuum for severity of TBI, progress has been made at opposite ends of the spectrum: The most severe patients are defined as minimally responsive, and the least severe persons are said to have so-called minor TBI. Definitions help specify problems that are being treated before interventions can be assessed.
Minimally Responsive Patient
The treatment of the minimally responsive patient via neurostimulation, sensory stimulation, or psychopharmacological means has been marked thus far by an absence of controlled studies. The recovery from coma is important for its own sake as well as its prognostic significance. A confounding problem is that diagnostic heterogeneity does not conform with behavioral homogeneity. Clinical syndromes (coma, vegetative state, persistent vegetative state, akinetic mutism, locked-in syndrome) and neuropathologic syndromes have to be reconciled with behavioral measures. To this end, a number of measures have been developed in the past 5 years: the Coma-Near Coma Scale, Coma Recovery Scale, Sensory Stimulation Assessment Measure, and the Western Neuro-Sensory Stimulation Profile (Giacino, 1992). These scales are more focused than previous measures; they are sensitive to more subtle changes and predict outcome. Whyte (1992) noted that variability is a common characteristic that may not be assessed in conventional rehabilitation instruments. Unlike other areas of rehabilitation, assessment is more transdisciplinary, based on ratings of targeted behavior.
Minor Traumatic Brain Injury
At the opposite end of the spectrum of severity, people with minor TBI have also become the subject of more intensive investigation. Zasler (1992) distinguished between separate effects due to whiplash, cranial trauma, and brain trauma, as well as posttraumatic stress disorder. Among the complaints with high incidence are vestibular disturbances, headaches, myofascial pain, dysfunction, depression, and anxiety. Although it has been suggested that people with minor brain injury who show symptoms immediately after the trauma tend to recover by 3 months, later study suggested that minor brain injury accompanied by radiographic evidence of lesion compromises recovery (Williams, Levin, & Eisenberg, 1990). Kay, Newman, Cavallo, and Resnick (1992) found that in 808 cases of minor TBI, 84% returned to work at 3 months, 7% during the first year, and 9% did not return to work. Complaints diminished, although at 1 year headaches still occurred in 23%, dizziness in 11%, and memory problems in 13% of the cases. Mateer (1992) and her colleagues identified deficit in attention as the primary problem in minor TBI. In a study of attentional disturbances in minor TBI, functional and dysfunctional people were distinguished by slowing of information processing and heightened susceptibility to interference. Problems of recall in memory were secondary to these disturbances. These factors distinguished people with subjective complaints from those without complaints (Newman, 1992). Progress in definitions depends on examining relationships between neuroimaging, clinical, and behavioral measures relative to each other.
With significant advances in measurement at opposite ends of the severity scales, the coming decade is positioned to examine interventions for these populations more precisely.
Advances in Treatment Modalities
In the 1,500 TBI treatment programs listed in the directory of the National Directory of Head Injury Services (1992), the vast majority deals with patients in the middle of the extremes. Because the variety of interventions is too great to review in a brief space, I divide them into two types: those emphasizing contextual approaches and those featuring cognitive remediation. The latter are grouped separately because the sheer number of empirical studies (N = 50) indicates an accumulating knowledge base and a heavy investment of resources.
Contextually Driven Interventions
With the recognition that rehabilitation should focus on facilitating behaviors that are expressed in nontreatment or naturalistic settings, a variety of treatment modalities that differ considerably from each other have become prominent. Among them are (a) advances in group methods, (b) the use of coaches, (c) newer methods in vocational rehabilitation, and (d) applications of computers as assistive technologies. One could argue that the common element of all of these approaches is to make interventions contextually relevant (i.e., targets of interventions are focused to improve functioning within nonlaboratory settings).
Group approaches have tended to play a more important role in TBI than in other rehabilitation populations. Groups present in vivo opportunities in interpersonal contexts to overcome deficient social skills (Butler & Namerow, 1988) and to elicit behaviors that are not apparent in one-to-one sessions. Groups offer a modality for managing psychosocial as well as cognitive problems. The economy in delivering group approaches also enhances their desirability. Groups have proliferated for a range of purposes from orientation training (Corrigan, Arnett, Hovek, & Jackson, 1985) to training for community reentry (Ylvasker, Szekeres, Henry, Sullivan, & Wheeler, 1987). Groups are evolving for targeted deficits for different levels of patients. They range from building basic skills, such as help in organizing schedules or tracking proceedings in the group discussions, to stress relief or self-regulation (i.e., control of affective flooding and executive dysfunctions; Sherr & Langenbahn, 1992). Although there may be a hierarchy in terms of level of competence and patient needs in assigning membership to a particular group, there is little evidence as to how to proceed (Deaton, 1991). In effect, just as school systems learned that classes were a viable way to educate children, rehabilitation programs are using groups to help individuals learn to adapt to residual problems in living with TBI.
A more recent innovation to deal with community adaptation has been the use of the coach. In the United States, the vocational placement counselor, serving as a coach to facilitate obtaining and holding a job, has received the attention of the rehabilitation community. The use of a personal coach to assist patients and families with cognitive and personal problems in the home has received far less attention. In an early study, it was shown that it is possible to teach spouses of aphasics to serve as coaches for their aphasic partners (Goodkin, Diller, & Shah, 1975). The concept has been extended to show that graduate students could be trained to (a) teach patients and families to compensate for perceptual and cognitive deficits, (b) problem solve with regard to community reentry and family adjustment issues, and (c) teach family members to distinguish deficits from behaviors that may ordinarily be viewed as noncooperation, stubborn, anger, or emotional disturbance (Diller, 1992).
Contextual approaches in vocational rehabilitation have been given added impetus in the rapid growth of supported employment. First developed for use with developmentally disabled, this approach seeks to bypass traditional counseling, vocational aptitude testing, and even work sample approaches. It seeks to have the individual go directly to an employment situation, in which a job coach works with the employer and the patient (Wehman & Kreutzer, 1990) for as long a period of time as necessary. Supported employment has met with some success for people entering jobs with low-level skill demands, performing routine tasks. However, TBI requires more extended support than does placement for other disability groups, therefore its cost-effectiveness may not be as great. A method for assessing "job survival" following placement should provide useful information for assessing effectiveness of vocational programs (Fabian, 1992).
Computers for remedial exercises have been a major growth industry in TBI rehabilitation. Indeed, the proliferation of software for cognitive remediation has been so great that the commercial test corporations (e.g., The Psychological Corporation) offered competitions for development of software, much as the government offers research grants. However, computer applications as cognitive assists extend beyond providing material for remedial exercises. Computers may be used in a more direct, ecologically valid way. Computers can be used to address the multitasking demands of any activity that can be varied in terms of complexity. There is a rapidly growing literature on the use of computers (Levin, 1991; Parente & DiCesare, 1991) as devices to (a) assist in scheduling activities, (b) cue people who are subject to breakdowns in performing sequential tasks, and (c) coordinate streams of activities. Computers as orthotic aids may receive their greatest boost from their use with normal people. For example, when a simple device is triggered it will emit a signal from a parked car to indicate where it may be located in a shopping mall. "The study of forgetting in everyday life, which developed as an emerging field in cognitive psychology, is already beginning to pay dividends in terms of commercial uses that lap over to fit individuals with special needs (Hermann & Petro, 1991). Special accommodations may be needed to fit the needs of people with motor impairments (O'Leary, 1991).
Prediction Studies
On the face of it, prediction studies, which capitalize on individual differences at entry to a program, would appear to be a useful way to find the right patientātreatment combinations. Prediction studies are attractive because they involve collecting data at different points in time and simply correlating them. Earlier reviews for stroke and TBI patients in rehabilitation settings found that predictions were positive, but correlations were not powerful enough for individual use (Acker, 1986; Meier, Strauman, & Thompson, 1987). The criteria and populations varied widely. More recently, Neimann, Ruff, and Baser (1990) found that more severe patients did not do as well as less severe patients in cognitive retraining. Lam, Priddy, and Johnson (1991) found that in a TBI population more intact people did better vocationally than less intact people.
In general, prediction studies attempt to forecast outcome with little regard for the content of specific programs. Hence, the same variables are used to predict success in a vocational program as in a rehabilitation medicine program. Models that work best clinically in rehabilitation settings expose a candidate to a sample of tasks to be mastered and rules to be followed as screening and prediction tools. This holds true for activities of daily living sampled in physical and occupational therapy, learning samples in a school setting, and work samples in an employment situation. The sampling approach is more useful on a clinical level, in contrast to the traditional prediction models, which might be useful in terms of categories of programs or populations. For example, the diagnosis of mental retardation might be useful for eligibility for a given program, but in a population of retarded people within a program it may not be specific enough to provide a good personāprogram match. In a recent study of a TBI program, where acceptance of disability was both part of the treatment as well as a therapeutic goal, staff ratings of trainee acceptance during the program were the best predictors of vocational outcome (Ezrachi, Ben Yishay, Kay, Diller, & Rattok, 1991).
Outcome Studies in Cognitive Retraining
There are now approximately 50 outcome stu...