Cognitive Behavior Therapy for Schizophrenia

Most of the studies on cognitive behavior therapy excluded patients who were judged not to be adhering to their regimens of antipsychotic medication and therefore not suitable for estimating adherence effects. There are a few randomized, controlled trials that used adherence rather than persistent symptoms as the primary outcome, and the results have been inconsistent (14). Kemp and colleagues conducted a study of a brief intervention using cognitive behavior therapy for acute inpatients in which the goal was to improve medication adherence after discharge (25). Their final cohort consisted of 74 inpatients treated for an acute psychotic episode who were randomly assigned either to supportive counseling or to six sessions of “compliance therapy” during hospitalization plus outpatient booster sessions. The patients assigned to compliance therapy had better adherence and fewer relapses than those assigned to supportive counseling (35% versus 60% relapse rate; hazard ratio = 2.1) (26). A study from Ireland failed to replicate the efficacy of compliance therapy (27) in relation to that for a supportive psychotherapy control group, although it is unclear whether the intervention included the components of cognitive behavior therapy that were used in the original study. Despite the inconsistency, the efficacy of cognitive behavior therapy for improving medication adherence seems to be more promising than that of traditional individual psychoeducation approaches, which have been consistently disappointing in their failure to show adherence benefits (28).

More recently, a series of “real-world” effectiveness studies have been published. Gumley and colleagues showed relapse prevention benefits from using cognitive behavior therapy for identifying prodromal signs of relapse (29). Durham and colleagues found modest benefits in relapse prevention and positive symptom control with cognitive behavior therapists who had limited prior training and supervision in cognitive behavior therapy for psychosis (30), although methodological problems make these findings equivocal. Further studies have shown that cognitive behavioral techniques can be used effectively in clinical practice by mental health nurses (31) given brief (2–3-week) training with ongoing clinical supervision. Overall, therefore, the studies of the effectiveness of cognitive behavior therapy generally favor the cognitive behavioral intervention, albeit less strongly than some of the predecessor efficacy trials.

The literature on cognitive behavior therapy has been favorable enough to make use of this intervention for schizophrenia a treatment recommendation in the United Kingdom. However, the evidence from the literature on cognitive behavior therapy is not definitive. There remain issues in interpreting the specificity of cognitive behavior therapy as well as the durability of any benefit beyond the period of the intervention itself. For example, the Tarrier studies demonstrated short-term specificity of the effect for cognitive behavior therapy over and above that of supportive counseling, but they did not show durability of that effect (22, 23). In contrast, the Sensky study showed just the opposite. This study failed to show short-term benefits from cognitive behavior therapy over and above those from befriending but did show greater durability of the improvements from the cognitive behavioral intervention (24). The reason for the contradictory findings is not known and therefore is unresolved. These unanswered questions point to the need for other randomized, controlled trials focusing on the specificity and durability of any putative benefits of cognitive behavior therapy.

Given these remaining uncertainties about the efficacy of cognitive behavior therapy in the United Kingdom, there are a number of questions when considering how to interpret the evidence for a U.S. treatment environment. There has been a dearth of controlled studies of the efficacy of cognitive behavior therapy for schizophrenia in the United States. The feasibility of transferring cognitive behavior therapy as currently practiced in the United Kingdom to U.S. practice settings is unknown, although successful U.S. and Canadian pilot projects are emerging (32–35). Nonetheless, it seems reasonable that greater consideration be given within the U.S. mental health system to supporting research and services to better understand the possible role of cognitive behavior therapy as a treatment option for persons with schizophrenia.

A therapeutic alliance is essential to any successful psychotherapy, including cognitive behavior therapy. Basic techniques include developing empathy, respect, unconditional positive regard, and honesty. One of the cardinal features of cognitive behavior therapy is its focus on subjective and behavioral connections among the patient's beliefs, feelings, and actions, irrespective of whether these beliefs are “reality based.” The approach involves collaboration without preconceived ideas through guided discovery (39, 40) and understanding of the person's experiences and beliefs.

The following example illustrates how a clinician with a cognitive behavioral orientation might respond when a patient describes a delusional belief:

Patient: “The Mafia has my house under surveillance!”

Clinician: “Well, that is possible. … But why do you think it is the Mafia? Could it be some other organization? Or is something else happening altogether? How could we find out?”

The clinician oriented in cognitive behavior therapy is interested in the specifics of the patient's experience. He or she tries to learn more about them and does not challenge the patient's beliefs while at the same time being careful not to collude with the delusion. In contrast, a clinician using a biomedical approach would be more likely to ignore the specific content of the delusion and, rather, discuss the delusion as a symptom of a neurobiologic disorder.

Cognitive behavior therapy explores and develops the patient's own understanding of his or her symptoms. The goal is to find explanations of the patient's experiences that are acceptable to both patient and clinician. It aims to improve understanding of the psychosis by using a vulnerability-stress model. Strengths and vulnerabilities are identified. The antecedent period is explored carefully, any pertinent stressors are elicited, and the possible effects of stress are discussed. A formulation is drawn up collaboratively, with care to ensure that neither the patient nor the patient's caregiver is led to feel he or she is to blame for the symptoms or the illness. The following example illustrates working toward such a collaborative formulation:

Initially, such alternative explanations may be considered by the patient but not necessarily accepted. With time, however, explanations that are mutually acceptable to both the patient and therapist may evolve (40).

The goal of cognitive behavior therapy is not to try to persuade or force the patient to agree that he or she has symptoms of a mental illness. Rather, the goal is to reduce the severity of, or distress from, the symptom regardless of whether the patient accepts a diagnostic label.

Delusions are appropriate targets for a collaborative formulation approach. One commonly used technique to start the formulation process is known as “peripheral questioning” (41). The clinician begins by asking a series of peripheral questions about the person's belief system, with the goal of understanding how the patient arrived at his or her convictions (e.g., “How could others control your thoughts? What mechanism would they use?”). Peripheral questioning is linked with graded reality testing, which in turn can lead to the introduction of doubt and the generation of alternative hypotheses. Education about real-world issues can help patients understand the factual assumptions made to support their belief systems (e.g., “Can microchips really be inserted without your knowledge when you are asleep?”). Such ideas can be explored with appropriate homework exercises (e.g., “Shall we find out—perhaps on the Internet—what we can about the use of microchips in operations? Also we could check about regulations concerning such operations”).

For more systematized delusions, the clinician can use “inference chaining” (42). This technique involves a process of looking for the key personalized meaning underlying a delusion, e.g., it can be used to respond to a statement like “I am the Second Coming of Christ.” A reply might be, “What does that [being the Second Coming of Christ] mean to you?” Should the patient reply, “It means that the world will be put to rights,” the subsequent question might be, “Why is that so important to you personally?” The immediate answer, “All the wrongs from the past will be judged,” could be followed with, “And why is that so important?” The subsequent response, “I was always being bullied at school,” would represent a successful use of inference chaining. In this case, the patient was a victim of bullying at school and responded to this traumatic experience by always demanding “fair” treatment. He became very paranoid shortly after being fired from a job for what he believed to be unfair reasons. Inference chaining identified this current feeling of life being unfair and his powerlessness in relation to it and allowed specific discussion of it. This process in a sense bypassed the delusion and resulted in constructive engagement and discussion and a lowering of his distress from his delusional beliefs.

Hallucinations can also be better understood by discussing the details of the experience. The clinician might start with, “Is it like somebody talking to you? Or shouting?” Testing out the exact location of the voices can follow, as well as other details, such as “Do other people hear their voices? If not, why not?” Discussion of circumstances in which people without mental illness hear voices can be relevant (“normalizing”). These symptoms can be provoked in “normal” people, for example, by sleep deprivation, sensory deprivation, bereavement, trauma, and solitary confinement (4). Normalizing is commonly used during the initial engagement with an acutely psychotic patient. Rather than try to explain that hallucinations are caused by a mental illness, a clinician using cognitive behavior therapy will often focus on the effects of stress, such as sleep deprivation (as in the preceding example). This may be exacerbating or even triggering the patient's hallucinations. This explanation often brings improved understanding and hope, as well as reducing the sense of alienation from others. The functions of medication are described as improving sleep and acting directly on overactive regions of the brain.

Beliefs about the voices themselves can include omnipotence and omniscience (43). The content of voices can be usefully debated; for instance, if the voices are making abusive statements, the accuracy of these statements can be debated. Often patients are deeply ashamed and embarrassed by the voice content and will avoid social interaction because of the possibility that others might hear what the voices are saying. The “voice hearing” experience may be better understood by using a “voice diary” to look for variation among different points in the day or among different activities. Situations that trigger an increase in voice intensity can be identified, with the generation of improved coping strategies. Affective responses to hearing voices (usually anger and anxiety) are often linked to unhelpful behaviors that maintain and exacerbate the voices. Once this pattern is identified, patients can gradually learn to engage more constructively with their voices. Patients can be trained to take a mindfulness approach to their voices, leading to acceptance and increased commitment to tackling normal day-to-day activities (17).

One of the potential concerns is whether cognitive behavior therapy can be misunderstood as a substitute for antipsychotic medication, rather than an addition to it. Another possible concern is whether a complication of a cognitive behavioral approach is medication nonadherence among patients who otherwise would remain adherent to their medication, possibly by underemphasizing the relationship between symptoms and antipsychotic medication. There is no evidence that this is a complication of cognitive behavior therapy. In fact, while better adherence was not an outcome in the studies of treatment-resistant schizophrenia, some reviews showed better retention in the groups receiving cognitive behavior therapy than in groups receiving other supportive therapies (14).

Persistent denial of illness is the strongest predictor of medication nonadherence (44–49). Cognitive behavior therapy does not insist on acceptance of a diagnosis of schizophrenia. An agreement can be reached that treatment, both psychological and psychopharmacological, may be helpful to counter the continuing negative effects of past traumatic events, sensitivity to stress, or even use of illicit drugs, without forcing the issue of acceptance or rejection of a diagnostic label.

Not surprisingly, cognitive behavior therapy is not effective for all patients. Some are simply too thought disordered or agitated to use cognitive behavior therapy, although the effect of medication may lead them to become more amenable to treatment. Some may be too paranoid and unable to form a viable therapeutic alliance. There is no evidence at the current time that cognitive behavior therapy is of benefit for patients who consistently refuse antipsychotic medication. Differences between the cultural backgrounds of the clinician and the patient may also be problematic; for instance, there is some evidence of this in the United Kingdom, where therapists who were not African Caribbean were found to be less effective with African Caribbean than with white patients (50). There may be cultural barriers to forming an alliance or developing shared formulations.

There remain considerable barriers to implementation, especially when cognitive behavior therapy is not considered a standard psychosocial intervention for schizophrenia. The training experience in the United Kingdom indicates that learning basic cognitive behavior therapy for schizophrenia requires a minimum of 2 weeks of intensive training plus ongoing supervision with an expert cognitive behavior therapy supervisor for clinicians who are already experienced in treating schizophrenia (31). Introduction of cognitive behavior therapy is often met with skepticism, ranging from therapeutic nihilism (e.g., “individual psychotherapy of any form cannot work for schizophrenia” or “people with schizophrenia are too cognitively impaired for psychological approaches”) to concerns about some of the specific techniques used (e.g., concern that normalizing the interpretation of symptoms represents collusion with the patient's denial of illness) (51).

Supportive psychotherapy, which is an accepted approach for helping patients with schizophrenia, deals with the loss, disability, and stigma arising from having to live with an illness such as schizophrenia (37). A major difference between cognitive behavior therapy and supportive psychotherapy is that cognitive behavior therapy uses specific techniques with the goal of actively reducing the severity of some of the core symptoms leading to distress and disability.

Current views characterizing schizophrenia as a brain disorder have dictated the use of a biomedical orientation for psychoeducation (52). Biomedical models share the concept of a specific diagnosis that has a final common pathway in significant abnormalities in CNS functioning. The potential complications that arise from communicating this concept may not be fully appreciated. Patients who accept a diagnostic label of schizophrenia have more depressive symptoms than those who do not (50). Investigators in two European studies using a randomized design found that the psychoeducation group experienced more depressive symptoms (53) and even suicidal ideation (54), and they expressed caution about using psychoeducation because of these risks. Cognitive behavior therapy tends to be more focused on symptoms than diagnosis and may help the patient accept necessary treatment without at the same time risking a worsening of affective and suicidal symptoms (24, 31). Cognitive behavior therapy is perhaps more acceptable—or less demoralizing—for patients struggling with the personal meaning of what is happening to them.

In the 1990s, Hogarty and colleagues developed and tested an individual psychotherapy known as “personal therapy” (55, 56). Personal therapy is a phase-specific individual treatment containing elements of psychoeducation, social skills training, and work on medication adherence (57). Psychoeducation, a cornerstone of personal therapy, is based on a biomedical model of illness causation, and so many of the key components of cognitive behavior therapy are not included in personal therapy, e.g., developing explanations for psychotic symptoms, normalizing, reality testing, and formulation. Unlike personal therapy, cognitive behavior therapy is designed to work directly on understanding and coping with the positive symptoms of psychosis rather than “containing” them (57).

Cognitive behavior therapy and cognitive remediation share the term “cognitive,” but they are very different treatments. Cognitive remediation is a rehabilitation approach whose techniques are adapted from the literature on brain injury. Schizophrenia is associated with cognitive deficits that impair social and occupational functioning (58). The goal of cognitive remediation is to improve neurocognitive function and teach patients strategies to compensate for deficits (59). Cognitive remediation and cognitive behavior therapy share a hopeful, optimistic approach that focuses on the patients' strengths. Cognitive remediation directly works toward improving neurocognitive functioning by using a model of brain injury and neuronal plasticity, whereas cognitive behavior therapy works through understanding the personal meaning of the content of the thoughts, sometimes known as “metacognition” or “thinking about thoughts.”

Cognitive behavior therapy differs from classic psychoanalytic therapies in that family conflicts and issues in childhood are viewed to be less causally related to schizophrenia. The classic psychodynamic therapy of schizophrenia as practiced in an earlier era uses free association within an open therapeutic space to allow the emergence of transference and countertransference phenomena. Patients with psychotic symptoms usually cannot tolerate an unstructured therapeutic environment, and classic psychoanalysis may therefore be too regressive for the psychotic patient (6). The structure and approach of cognitive behavior therapy in this regard are completely different. There remain some techniques that are shared by cognitive behavior therapy and more modern psychoanalytically oriented psychotherapy of schizophrenia. Both focus on the therapeutic relationship and the personal meaning of psychotic symptoms. However, a cognitive behavioral approach tends to be more structured and explicitly collaborative.

Cognitive behavior therapy can be used alongside most biologic models of schizophrenia. It is perfectly acceptable for a psychiatric practitioner to believe in a biologic/medical causation of schizophrenia and still embrace a cognitive behavioral model to use with patient care. Although a cognitive behavioral approach would not contradict a biologic point of view in a patient whose personal explanation fits that model, it does not insist on it for patients who prefer other explanations. Therefore, cognitive behavior therapy is not compatible with any kind of biomedically based intervention that requires using the diagnostic label “schizophrenia,” forbids any exploration of a personal meaning (formulation) of psychotic symptoms, or precludes the possibility of meaningful recovery. On the other hand, cognitive behavior and personal therapy seem contradictory and incompatible. Similarly, cognitive behavior therapy and psychodynamic therapy, as currently practiced, seem technically too different at least to be offered simultaneously, despite some commonality in relation to exploring meaning and developing the therapeutic relationship. Cognitive behavior therapy and cognitive remediation are more compatible because of their different and complementary goals, although one should be mindful of not overtaxing patients with two simultaneous interventions, both of which require homework and active participation.