Cognitive Behaviour Therapy and Cognitive Training in Work Rehabilitation for Persons With Severe Mental Illness

Overview

This study will compare the effects of cognitive behaviour therapy (CBT) and cognitive training in work rehabilitation of patients with severe mental illness (psychoses). These interventions will be compared with a matched control group ('treatment as usual'). The patients will be recruited in nine different counties in Norway. During a two-year period around 27-30 patients can be recruited from each county, meaning that each intervention arm will contain approximately 80 patients. The evaluation of the patients will include several validated diagnostic instruments as described below. The counties has been randomised to the two different intervention groups by a neutral institution: 1.group receiving work rehabilitation based on cognitive behaviour therapy, and 2.group receiving cognitive training. The control group will be matched according to gender and age. With Ntotal=160 for the comparison of two groups (80 patients x 2), a 0.05 level of significance and a power of 0.80, a standardised group difference of 0.44 can be detected. The standardised difference detected between supported employment and treatment as usual has been as high as 0.80 in comparable American studies, indicating that N is large enough in the present study. The participants in each county will be evaluated and followed by a local project coordinator and by a local psychiatrist/psychologist in a District Psychiatric Centre. Written evaluation protocols will be sent anonymously to the research centre for registration. Data will be stored according to current laws of person protection and data security. Pre-post differences in each group, differences between the control group and the intervention groups, and differences between the intervention groups will be tested using a mixed linear model programme. Research questions The project will compare the effects of CBT oriented work rehabilitation and work rehabilitation with cognitive training with regard to the ability of persons with severe mental disorders to cope with and keep a job. The two intervention arms will be compared with each other and with a matched treatment as usual control group. Main hypotheses 1. By the end of the project (T2) and by follow up examination (T3) the global psychosocial function of patients will be higher in the CBT work rehabilitation group than in the control group 2. The global psychosocial function of patients at T2 and T3 will be higher in the cognitive training group than in the control group

Cognitive behaviour therapy and cognitive training in work rehabilitation for persons with severe mental illness. A randomised controlled trial Introduction This study will compare the effects of cognitive behaviour therapy (CBT) and cognitive training in work rehabilitation of patients with severe mental illness (psychoses). These interventions will be compared with a matched control group ('treatment as usual'). The patients will be recruited in nine different counties in Norway. During a two-year period around 27-30 patients can be recruited from each county, meaning that each intervention arm will contain approximately 80 patients. The evaluation of the patients will include several validated diagnostic instruments as described in the detailed description below. The counties will be randomised to the two different intervention groups: 1. a group receiving CBT inspired work rehabilitation, and 2. a group receiving work rehabilitation integrated with cognitive training. The control group will be matched according to gender and age. With Ntotal=160 for the comparison of two groups (80 patients x 2), a 0.05 level of significance and a power of 0.80, a standardised group difference of 0.44 can be detected. The standardised difference detected between supported employment and treatment as usual has been as high as 0.80 in comparable American studies, indicating that N is large enough in the present study. The participants in each county will be evaluated and followed by a local project coordinator in a District Psychiatric Centre (DPS), and by a local psychiatrist/psychologist. Written evaluation protocols will be sent anonymously to the research centre for registration. Data will be stored according to current laws of person protection and data security. Pre-post differences in each group, differences between the control group and the intervention groups, and differences between the intervention groups will be tested using a mixed linear model programme. Background Vulnerability and stress in schizophrenia Schizophrenia is a mental disorder characterised by psychotic symptoms, apathy, social isolation and cognitive problems, which often reduce the functional capacity in central life arenas. Patients have problems in school and at work, with being a parent or friend, and in severe cases with being able to live on their own and cope with the challenges of daily life (Mueser \& McGurk 2004). Schizophrenia is one of the ten most important causes of severe chronic disability (Velligan \& Gonzales 2007). The course of the disease is perhaps best described and understood by the so-called stress/diathesis model. According to this model the person has a psychobiological vulnerability basically determined by genetic and perinatal environmental factors. The debut and course of the disease is determined by a dynamic interplay between biological and psychosocial factors. Medication and use of legal and illegal drugs are the most important biological factors. Antipsychotic medication can reduce symptom intensity and the probability of relapse, whereas drug abuse often has the opposite effects. The most important psychosocial factors are perceived stress, social support, and the patient's coping profile. High levels of stress can increase the biological vulnerability, symptom intensity, and relapse probability, whereas social support and the development of social abilities and problem solving skills tend to protect the individual. Cognitive behaviour therapy Cognitive behaviour therapy (CBT) in schizophrenia was originally developed to help the patient cope with delusions and hallucinations that did not respond to medical treatment. CBT assumes that delusions and hallucinations are influenced by the person's problems sorting and interpreting information, and that cognitive interventions can increase coping(Velligan \& Gonzales 2007). Therapist and patient examine the contents of the symptoms in detail, and explore whether the symptoms can be understood in new and more functional ways. Controlled studies clearly support the efficacy of CBT in schizophrenia (Tarrier \& Wykes 2004, Turkington et al 2004). The explanatory model of CBT, including core concepts as schema, automatic thoughts, socratic questioning, and alternative thoughts, now colour most of the rehabilitation programmes in schizophrenia. Cognitive training It was previously assumed that the reason why persons with psychotic diseases have problems in daily life is their reduced sense of reality, delusions and hallucinations. Today there is consensus that the problem is far more complex. Research shows that neurocognitive dysfunction is prevalent in persons with schizophrenia, and that this rather stable dysfunction is an important prognostic factor (Velligan et al 1997). The patients have problems with memory, attention, problem solving capacity, and reduced psychomotor speed. It has been demonstrated that neurocognitive dysfunction impacts more strongly on activities of daily life than do positive symptoms of psychosis, and that it predicts coping about as strongly as negative symptoms (Green et al 2000). Neurocognitive dysfunction impairs the person's ability to learn new skills, social relations, academic capacity and work capacity. It is particularly manifested as attention and concentration problems, when complex work tasks must be solved under time pressure. Consequently, many of these patients have problems handling jobs and academic courses that demand speed and mental flexibility. They have serious problems qualifying for a job, and if employed they tend to loose their job more often than other workers, with adverse social, economic and mental consequences. These patients also strive to cope with rehabilitation projects involving many people (Bell \& Bryson 2001). Today several researchers maintain that neurocognitive dysfunction is the key problem in schizophrenia. These findings have produced new rehabilitation programmes aiming at developing or restoring neurocognitive function (McGurk et al 2007). Cognitive rehabilitation focuses on improving function through a) reinforcement or reestablishment of previously learned behaviour, b) establishment of new cognitive abilities or compensatory mechanisms for impaired neurological systems (Harley et al 1992). These methods have for many years been used successively in treatment of patients with brain damages. Great progress has been made in this field during the last two decades, and training programmes are now also available for persons with mental disorders. The effects of cognitive rehabilitation have been examined in controlled studies showing that neurocognitive function can be improved by structured training (Ueland \& Rund 2005), and as a consequence psychosocial adaptation may be improved. Work rehabilitation in schizophrenia The National Norwegian Development Plan for Mental Health (Opptrappingsplanen for psykisk helse) states that work and meaningful activities are important sources of social identity and self esteem. Work rehabilitation for patients with schizophrenia and other psychotic disorders traditionally has been a great challenge both in Norway and other countries. Even though about half of the patients want a job, only a minority of these are employed (McGurk \& Mueser 2004). Chronic social dysfunction makes it difficult to compete effectively on the ordinary job market. Optimal treatment of symptoms combined with cognitive training that increase competence may somewhat improve their chances, but recent studies indicate that for many patients adjustments of job demands are necessary in addition to continuing support to cope with the job (Drake et al 1999). More than 50% of the participants cope in supported employment programmes in the US, compared to only 20% in control groups. Programmes combining cognitive training with supported employment seem to be particularly promising (Wexler \& Bell 2005, McGurk \& Mueser 2004). These programmes reduce stress through personal support and tailoring of job demands, and increase competence through cognitive training. Prejudice and lack of openness and understanding are still prevalent at the work place, and tend to prevent successful work rehabilitation of patients with psychotic disorders. Like family members leaders and colleagues may easily perceive symptoms as incomprehensible and frightening or as expressions of laziness or hostility, and react negatively and correcting, with the same adverse consequences. In a pilot project we investigated whether the psycho-educative method could be adapted and used in work rehabilitation of patients with severe mental disorder. The goal of this project was to examine whether the psycho-educative method used in work rehabilitation can increase the chances that persons with severe mental disorders will cope with a job, and, secondly, to develop good models of collaboration between the welfare institutions to improve work rehabilitation for persons with severe mental disorders. Two external evaluations of the pilot project concluded that both the psycho-educative method and the other cognitive methods can be adapted to the work rehabilitation context, and that the organised collaboration between clinicians and job consultants has a great potential. The participants were generally very stable and evaluated the experience positively. The engagement in the pilot project grew over time both in the work and welfare system and in the psychiatric health care system (DPS), and so it was decided to spread the project to nine different counties. Research questions The present project will compare the effects of cognitive behaviour therapy and cognitive training with regard to the ability of persons with severe mental disorders to cope with and keep a job. The two intervention arms will be compared with each other and with a matched treatment as usual control group. Hypotheses 1. By the end of the project (T2) and by follow up examination (T3) the global psychosocial function of patients will be higher in the CBT work rehabilitation group than in the control group 2. The global psychosocial function of patients at T2 and T3 will be higher in the cognitive training group than in the control group Material and Method Material Persons with DSM-IV schizophrenia spectrum disorders about to start work rehabilitation will be recruited in nine different counties in Norway. From each county around 27-30 patients can be recruited during a two-year period, giving a total of a little less than 250 patients. Method Instruments The following instruments will be used: Diagnosis: The MINI interview (Sheehan et al 1997). Intensity of positive and negative symptoms: The Health of the Nation Scale (HoNOS, Wing et al 1996), The Apathy Evaluation Scale (self rating version) (Marin et al 1991), and the PANSS (Key et al 1987). Global functioning: The Global Assessment of Functioning Scale (GAF) (APA 1987). Complicating drug abuse: The Alcohol Use Disorders Identification Test (AUDIT) (Saunders et al 1993) and Drug Use Disorders Identification Test (DUDIT) (Bermann et al 2005). Cognitive functioning: The Measurement and Treatment Research to Improve Cognition in Schizophrenia (MATRIX)-battery (Marder et al 2004). Relations to leaders and colleagues: The Felt Expressed Emotion Rating Scale (FEERS, Bentsen 1999). Work performance: The Work Behavior Inventory (Bell et al 2004) Intervention CBT in work rehabilitation: The contents of the psychoeducative classes in the pilot study has been adjusted and standardised. Classes will be offered to all employees in the participating companies. Central themes are symptoms, course, methods of treatment, rehabilitation, relapse prevention and prognosis in schizophrenia and other psychoses. The core elements and the basic ways of reasoning in CBT (schema, automatic and alternative thoughts) will be particularly emphasised. The DPS supervisor in each county will meet the job consultants weekly through the whole project period and be easily available for consultations on telephone. The central supervision focus will be current communicative and behavioural problems at the work site. Attempts to solve such problems will continuously include the participants. Each participant will be offered 10 months in the project. Cognitive training: The cognitive training will be computer based and use a programme package developed by dr. psychol. Torill Ueland. The programmes include training of attention, memory, executive functioning, and psychomotor speed. Computer based training allows for multi-sensoric feedback and positive reinforcement. Moreover, computer based training can be used individually in that the degree of difficulty can be varied . This is important since neurocognitive profile and functioning vary among patients. Group training is economically attractive since one person can administrate the programme to six persons simultaneously. The programme package has been adapted to the Norwegian context, but the selection of tasks is based on international research. It combines tasks originally developed for patients with brain injuries (Cog Rehab) and tasks developed for psychiatric patients (Cogpack). Tasks from both sources have been used in several previous studies (Bell et al 2001, McGurk et al 2005). Design Six counties have been randomised to one (n=80) of the two intervention groups. A matched control group (treatment as usual, n=80) will be established in three other counties. With a total N=160 for comparison of two groups, a significance level of 0.05 and a power level of 0.80, a standardised group difference of 0.44 can be detected. The standardised difference between supported employment and treatment as usual in comparable American studies have been as high as 0.80 (Velligan \& Gonzales 2007), meaning that N should be acceptable in the present project.