by Michael D. Anestis, M.S.
Today's article aims to accomplish two tasks: first, to continue our coverage of effectiveness trials for empirically supported treatments (ESTs) and second, to add to our thus far limited coverage of schizophrenia. Let me clarify that first goal a bit. Many individuals who oppose the use of scientific research to guide treatment decisions are understandably concerned that the things we learn in controlled studies do not apply to "real world" clinical settings. Most of these controlled studies, known as efficacy studies, require that participants only meet criteria for one diagnosis and do not have currently high levels of suicide risk and these studies typically involve therapists for whom the clinical trial represents most if not all of their clinical obligations. Clients in the "real world," on the other hand, typically meet criteria for more than one diagnosis, often exhibit high levels of suicide risk, and only have access to mental health practitioners or primary care physicians who have large case loads and little time (of course, these critics often forget that the therapists in efficacy trials typically have many additional time-consuming professional demands, but this is another point altogether). Efficacy studies like this, however, are only the first step in establishing the empirical support for a particular form of psychotherapy. The next step - effectiveness studies - look at how well therapies that perform well in controlled conditions do in "real world" settings. In other words, the restrictions inherent in efficacy studies are removed and researchers, now confident that the therapy has the potential to be useful, examine whether or not it can be legitimately implemented in a more standard clinical environment. We have covered this point in other articles on PBB, but thus far research is repeatedly demonstrating that empirically supported treatments for a variety of diagnoses perform at least as well in "real world" practice as they do in efficacy trials.
All of this bring us to today's topic. We have spent very little time discussing schizophrenia on PBB, so it seemed like a good idea to expand our effectiveness discussion to this debilitating diagnosis in an effort to accomplish two important goals simultaneously. To do this, I would like to discuss a study recently published in the Journal of Clinical Psychiatry by Navdeep Malik, David Kingdon, Jeremy Pelton, Raj Mehta, and Douglas Turkington (2009) in which they examined the effectiveness of brief cognitive behavioral therapy (CBT) administered by mental health nurses for schizophrenia. This study is important for a number of reasons. First, as hinted at above, providing effectiveness data is important in countering the arguments of individuals who claim that research findings can not be applied to standard clinical care. Additionally, however, the study is important because it examines whether an EST can be provided effectively by mental health practitioners other than highly trained therapists with doctoral degrees. As much as I would like to believe that my pending Ph.D. will bestow upon me the insight of a wizard and turn me into a ninja therapist capable of solving all psychological problems with skills unique to my training, the evidence for such beliefs is sorely lacking and, even if it existed, might be irrelevant, as it seems unlikely that the majority of mental health care will be provided by such folks any time soon. In other words, it makes sense to see to what degree EST's can be taught to and implemented by individuals with less intensive training, as many clinicians fit within that category and there is a distinct possibility that, due to cost-effectiveness, they will represent an increasingly high percentage of mental health practitioners in the future.
Before we move further into discussing the study itself, let me provide a little bit of background information on schizophrenia, much of which can be found in the original Malik et al (2009) article. Schizophrenia is a chronic and debilitating mental illness that involves immense financial and emotional costs both for those who suffer from the illness and their loved ones. The disorder is best thought of as part of a spectrum of syndromes, including schizotypal personality disorder, schizoaffective disorder, and the various subtypes of schizophrenia. Within schizophrenia, there are many varieties of presentations. Symptoms are often classified as either positive - characteristics that would not be present without the disorder (e.g., hallucinations, delusions) - or negative - characteristics that are lacking or blunted due to the disorder (e.g., restricted range of emotions). Without question, the primary method of treatment for schizophrenia is medication, with atypical antipsychotics producing the most promising effects. Unlike older antipsychotic medications, which only addressed positive symptoms, atypical antipsychotics have been shown to reduce both the positive and negative symptoms of schizophrenia. Like their predecessors, however, these medications are quite potent and result in many aversive side effects (e.g., tardive dyskinesia, weight gain). As such, many individuals prescribed these medications do not properly adhere to their dosage and, even in those who receive maintenance medication, 30%-40% relapse within one year of hospital discharge (Davis, 1975, Hogarty et al., 1979). Given the relatively high relapse rate and the emotional and financial impact of relapse, the need for adjunct psychosocial treatments seems clear.
Along these lines, researchers have spent a lot of time examining the impact of various psychosocial treatments in combination with medication for schizophrenia. Several studies have failed to find any meaningful impact of CBT on schizophrenia relapse (e.g., Drury, Birchwood, & Cochrane, 2000; Jones et al., 2004; Tarrier et al., 2004); however, all of those studies included very small samples, which vastly diminishes their statistical power and increases the likelihood of finding false negatives (data that erroneously indicate that there is no impact of treatment on outcome). Malik et al (2009) recognized this problem and, in the study I am about to discuss, addressed the sample size problem admirably. Additionally, the authors followed-up with patients 12 and 24 months after the end of therapy in an effort to measure the degree to which treatment effects were maintained.
Okay...on to the study itself.
Patients
All patients included in the sample met ICD-10 (essentially the European counterpart to the DSM) criteria for schizophrenia and many were recruited from six centers based in the United Kingdom. Other patients were recruited from primary care settings. 257 patients were randomly assigned to receive brief CBT and 165 were randomly assigned to receive treatment as usual (TAU).
Nurses
All of the mental health nurses who provided treatment in this study went through an intensive 10-day training program. All of the nurses were registered mental health nurses (RMNs), but none had received a master's degree or equivalent graduate designation. One of the nurses had prior training in CBT for schizophrenia and also served as a co-trainer/co-supervisor, but none of the other five nurses had any prior training in CBT. In addition to the training program, each nurse received short sessions of weekly supervision throughout the study.
Interventions
Each CBT patient received a total of 6 CBT sessions over a 2-3 month period with no follow-up or booster sessions. Treatment included the development of stigma reducing explanations for psychotic symptoms, the implementation of skills aimed at reducing distress, the improvement of treatment (medication) adherence, and the reduction of negative views of symptoms. Treatment also included the developments for early intervention plans should signs of relapse emerge. Additionally, with the permission of the patient, caregivers (e.g., family members) were offered 3 parallel CBT sessions during the course of treatment in an effort to increase their understanding of schizophrenia and CBT skills and to reduce expressed emotion.
Each TAU patient received regular review by a psychiatrist, free antipsychotic medications, access to a day hospital, and social support in the community.
Results
Only patients who were available for follow-up data at 24 months post-treatment were included in the analyses, resulting in a sample of 205 CBT patients and 165 TAU patients. 24.9% of patients who received CBT relapsed within 24 months compared to 34.5% of TAU patients. CBT patients spent an average of 32.7 days hospitalized post-treatment compared to 48.9 days for TAU patients. CBT patients averaged 356.8 days to relapse compared to 296.1 days for TAU patients. Each of these differences was significant, indicating that, while it certainly was not perfect, CBT was vastly superior to TAU in diminishing relapse for individuals suffering from schizophrenia.
10.2% of CBT patients made occupational recovery - meaning that they resumed full or part time work, education, or training - compared to 13.6% of TAU patients. This difference was not statistically significant, meaning that neither treatment successfully addressed this concern and neither treatment was better than the other on this particular point.
Cost-Effectiveness Evaluation
Approximately 22% of mental health costs within the United States - 1.5%-3% of total national health expenditures - are related to schizophrenia (Rice, 1999). In 2002, this translated to $62.7 billion in the US alone (Wu et al., 2005). As such, the degree to which treatments afford an opportunity to save money is a legitimate concern. Based upon their findings, the authors concluded that $808,000 could have been saved across the 6 sites from closing beds as a result of the reduced hospitalization in the CBT group. Across the 24 months, savings on individual patient care would have been approximately $1,661,400. This represents more than enough savings to cover the costs of training nurses in brief CBT and implementing the program.
Limitations
This study, while impressive, was not perfect. There was no active treatment condition to compare to CBT, so we can not use these results to conclude that CBT is better than anything other than TAU. Additionally, the definition of relapse in this study was hospitalization, which might not be a broad enough definition to fully capture the idea.
Summary
So, overall, what did we learn? First of all brief CBT in conjunction with proper medication is superior to TAU in reducing relapse rates in schizophrenia. Additionally, this protocol can be successfully implemented in "real world" settings by nurses with little to no background in CBT with a quick training program. Although neither CBT nor TAU successfully impacted occupational recovery, the overall results were very promising and the cost-effectiveness of the intervention is highly appealing.
***********
If you would like to learn more about CBT for schizophrenia, we recommend the following item, which is available through our online store of scientifically-based psychological resources:
- Schizophrenia: Cognitive Theory, Research, and Therapy by Aaron Beck, Neal Rector, Neal Stolar, and Paul Grant
Mike Anestis is a doctoral candidate in the clinical psychology department at Florida State University
Subscribe in a reader
