Cognitive-behavioral therapy (CBT) in schizophrenia was originally developed to provide additional treatment for residual symptoms, drawing on the principles and intervention strategies previously developed for anxiety and depression. In the 1950s, Aaron Beck1 had already treated a psychotic patient with a cognitive approach, but thereafter the research in this specific area lay dormant for decades. Only after cognitive therapy had been firmly established for depression and anxiety, in the 1990s, did the research into psychological treatments for psychotic conditions gather force—again, with Beck in the forefront.
Pharmacologic therapy can leave as many as 60% of psychotic patients with persistent positive and negative symptoms, even when the patients are compliant with their medication instructions.2 Furthermore, medication compliance remains a major problem despite the introduction of modern atypical antipsychotics. Studies have shown treatment discontinuation in an estimated 74% of patients in both outpatient and inpatient settings.3
The evidence for the efficacy of CBT in treating patients with persistent symptoms of schizophrenia has progressed from case studies, case series, and uncontrolled trials to methodologically rigorous, randomized, controlled trials that include patients from both the acute4 and the chronic end of the schizophrenia spectrum.5-7 Subsequent meta-analysis8 and systematic reviews have further strengthened the evidence base.
CBT is now recognized as an effective intervention for schizophrenia in clinical guidelines developed in the United States9 and in Europe.10 In spite of the evidence base and absence of side effects, however, the general availability of this treatment approach within community settings is still low.11 This article will examine the procedure of CBT for psychosis, the evidence for its use, and the implications for practicing psychiatrists.
The therapeutic techniques used for patients with schizophrenia are based on the general principles of CBT. Links are established between thoughts, feelings, and actions in a collaborative and accepting atmosphere. Agendas are set and used but are generally more flexibly developed than in traditional CBT. The duration of therapy varies according to the individual’s need, generally between 12 and 20 sessions, but often with an option of ongoing booster sessions. CBT for psychosis usually proceeds through the following phases.
The assessment begins by allowing the patient to express his or her own thoughts about his experiences while the therapist listens actively. The use of rating scales—both specific and general—is encouraged to monitor progress, and the results are shared with the patient. Diagrams and written material can be most useful, especially for patients with chaotic lifestyles. The formulation of symptom causation and maintenance is also shared with the patient and evolves throughout the therapy as new information is considered.
Initially the therapist will state clearly what the therapy is about (including a safe and collaborative method of looking at causes of distress). Throughout the therapy, the use of Socratic questioning is emphasized. This involves drawing out the person’s own understanding of his situation and ways of coping with it through a process of guided discovery. Attempts are made to empathize with the patient’s unique perspective and feelings of distress and to show flexibility at all times. A vulnerability-stress model is used, so that the patient can understand that vulnerability is a dynamic concept that can be influenced by many factors, such as life events, coping mechanisms, or physical illness. The therapist stresses that he or she does not have all the answers but that useful explanations can be developed in cooperation. The typical nonspecific therapeutic factors of warmth, genuineness, humor, and empathy are of great value in this type of therapy, as in all other therapeutic encounters.
The ABC model, which was originally developed by Ellis and Harper,12 can be used to give the patient a way of organizing confusing experiences. It involves slowly and thoroughly moving the patient through the various steps using Socratic questioning to clarify the links between the emotional distress the patient is experiencing and the beliefs he holds. It includes the following steps:
- Based on a scale of 0 to 10, the patient rates the intensity of distress.
- The consequence (C) is assessed and divided into emotional and behavioral Cs.
- The patient gives his own explanation as to what activating events (As) seemed to cause C; and the therapist ensures that the factual events are not “contaminated” by judgments and interpretations.
- The therapist provides feedback to the patient to acknowledge the A-C connection.
- The therapist assesses the patient’s belief, evaluations, and images and communicates to the patient that a personal meaning is lacking in the A-C model; simple examples can be provided to facilitate understanding.
- The patient’s own belief (B), which is actually the cause of C, is then discussed; often, this can be rationalized, and a B such as “nobody will like me if I tell them about my voices” can be disputed and changed to “I can’t demand that everyone likes me. Some people will and some won’t…Maybe some friends might find it interesting.” This may lead to a change in C, ie, less sadness and isolation.
Realistic goals for therapy should be discussed early in the therapy with the patient, using the distressing consequences (C) to fuel the motivation for change. It is the therapist’s job to ensure that the goals are measurable, realistic, and achievable. The goals are revisited both during and at the end of therapy.
A normalizing rationale11,13 is helpful in decatastrophizing psychotic experiences. Education regarding the fact that many people can have unusual experiences in a range of different circumstances (stressful events, hyperventilation, torture, hunger, thirst, falling asleep, etc) reduces anxiety and the sense of isolation. By having the psychotic experiences placed on a continuum with normal experiences, the patient will often feel less alienated and stigmatized. As a consequence, the possibility of recovery seems less distant.
Critical collaborative analysis
To proceed to this stage, the therapeutic relationship must have developed a degree of trust. The therapist uses gentle Socratic questioning to help the patient appreciate potentially illogical deductions and conclusions:
- “If your voices came from the radiator, why can’t anyone else hear them?” or
- “Hold on for a moment, this puzzles me. How do you explain your rape by this famous actor, since we know for a fact that he has never been to this country?”
Testing the evidence for and against maladaptive beliefs can safely be carried out without causing distress as long as the therapist remains nonjudgmental, empathic, and open-minded. An assessment is made of how the beliefs occurred—through inferences or cognitive distortions (eg, dichotomous thinking, selective inference, emotional reasoning, jumping to conclusions). Reviewing antecedents (stress, trauma, loss) that prepare the ground for psychotic change can be an eye-opening exercise for both patient and therapist. Identification of misattributions and attempts to reattribute are as productive as homework tasks.
Developing alternative explanations
It is of crucial importance to let the patient develop his own alternatives to previous maladaptive assumptions, preferably by looking for alternative explanations and coping strategies already present in the patient’s mind. It can be dangerously tempting to force the therapist’s readymade explanations onto the patient. The patient’s own healthier explanations might just be temporarily weakened by either external factors or dysfunctional thinking patterns. If the patient is not forthcoming with alternative explanations, new ideas can be constructed in cooperation with the therapist. Certain seeds might have been sown earlier in the therapy (from leaflets and previous discussions) that can now be used as building blocks.
Tina is a 64-year-old woman who has been hearing voices for 45 years. She has been hospitalized 4 times over the years, often in connection with life events, such as her mother’s death and her brother’s alcoholism and violence towards her. She has never accepted a diagnosis of schizophrenia but is compliant with medication and happily agreed to CBT, explaining that she would like to talk to a “nice young man” like the therapist. Following the assessment, it was clear that the patient’s main problems were lack of confidence, a tendency to isolate herself at home, and a belief that the voices had an external source (the central heating system). She was still working as a caregiver in a rest home, but after her partner’s and mother’s deaths 15 to 20 years previously had increasingly lost contact with friends and family. As protective factors, Tina was a healthy, likable woman with a good sense of humor. She was also very happy with her job.
Engagement was unproblematic, but the concepts of continuum and the ABC model proved to be difficult for Tina to understand. It helped when examples—such as temperature—were given to explain the concept of a continuum. To illustrate the ABC model to Tina, a book was held up and the therapist read aloud what he could see on the back of the cover, while she read the front of the cover. She subsequently understood that there could be different angles from which to view the same issue. While the therapist critically looked at her understanding, he also made extensive use of normalizing.
The patient’s general functioning, mood, and self-confidence improved significantly, but she remained adamant that the voices had an external source. She had also gained a much larger social life. After a couple of relapse-prevention sessions, it was decided to stop the therapy. Tina had had 22 sessions in all, and the voices were now largely benign and seen as conversational partners that conveniently would remind her about duties such as vacuuming and writing cards to her nephews and nieces.
The general finding has been that CBT significantly improves both negative and positive symptoms in different subgroups of patients with schizophrenia.8 The meta-analysis by Gould and associates14 indicated a strong effect size of 0.65 for positive symptoms. In spite of initial concerns, there is no evidence that suicidality develops during CBT; quite the contrary.15 At this stage, the factors mediating treatment success in these interventions are not clearly known and should be researched further. Compared with pharmacologic therapy, the dropout rates are remarkably low at around 12% in the randomized controlled trials.14 This is quite an achievement considering the severely unwell patient group, and it may indeed carry an important message to service providers about service users’ preferences. It must, however, be added that all studies so far have been conducted on patients receiving antipsychotic medication. The proven efficacy of CBT for schizophrenia is also cost-effective.10
Patient predictors of response to CBT remain uncertain. Although there is some evidence that a degree of cognitive flexibility and willingness to disclose are auspicious signs, cognitive variables were not found to be related to treatment response.16 Several of the comparison treatments (supportive counseling, befriending) have been shown to have some effect on a number of symptoms, but CBT has shown clear superiority in durability.5
It appears that CBT equips patients with a set of tools they can use to fight back the symptoms long after the therapy has been terminated. The effect in the comparison groups is thought to be a product of nonspecific therapeutic factors and the impact of being the object of caring attention. The studies have the same methodologic limitations as most other research into psychotherapy, including suitable comparison groups, blinding, and inclusion and exclusion criteria. Recent studies have, to a large degree, dealt with these issues, and the outcome generally remains positive.
It would be wrong to believe that CBT can only be used in formal therapy settings. Many aspects of the therapy can readily be implemented in the day-to-day management of patients with schizophrenia, including the ABC model, normalization, and the search for alternative explanations. The use of these approaches does not necessarily require formal training in CBT. On the other hand, the lack of supervision and of fully accredited therapists are major obstacles in the development of a service that lives up to the standard requirements.17 This area will undoubtedly receive further attention over the coming years, especially as patients and caregivers become more vocal about their needs and preferences.
Over the past decade, CBT has emerged as an evidence-based intervention that provides a long-needed integrative approach to schizophrenia. The emergence of CBT for schizophrenia has added new optimism to the treatment of a highly stigmatized condition and may, in the long term, contribute to a change in the way the general public views people with schizophrenia. As the news about an effective talking therapy penetrates a wider audience, schizophrenia may no longer be seen as an essentially untreatable, incomprehensible, biologic condition beyond the reach of reasoning.
All psychiatrists should therefore at least be acquainted with the basic principles of CBT for schizophrenia in order to incorporate this knowledge into the daily management of severely mentally ill patients and to be able to appropriately refer patients for specialist therapy. Although the existing evidence base for CBT in schizophrenia shares some of the same limitations that exist for other psychotherapies, research has firmly established the evidence for reduction of symptomatology, low dropout rates, and cost-effectiveness. Despite this, widespread availability of CBT for psychotic patients is currently lacking, and providing sufficient availability of this method is one of the greatest challenges facing mental health services today.
Dr Hansen is a Danish-born consultant psychiatrist working in the department of psychiatry, University of Southampton, UK. He is an accredited member of the British Association of Behavioural and Cognitive Psychotherapists and holds a medical doctorate from the University of Southampton. He reports that he has no conflicts of interest concerning the subject matter of this article.
Dr Kingdon is professor of mental health care delivery at the University of Southampton and honorary consultant psychiatrist with Hampshire Partnership Trust. He reports that he has no conflicts of interest concerning the subject matter of this article.
Dr Turkington is a senior lecturer of psychiatry at Newcastle University with a special interest in cognitive-behavioral therapy for schizophrenia. He reports that he has no conflicts of interest concerning the subject matter of this article.
1. Beck AT. Successful outpatient psychotherapy of a chronic schizophrenic with a delusion based on borrowed guilt. Psychiatry. 1952;15:305-312.
2. Harrow M, Carone BJ, Westermeyer J. The course of psychosis in early phases of schizophrenia. Am J Psychiatry. 1985;142:702-707.
3. Lieberman JA, Stroup TS, McEvoy JP, et al, for the Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE) Investigators. Effectiveness of antipsychotic drugs in patients with chronic schizophrenia. N Engl J Med. 2005;353:1209-1223.
4. Lewis S, Tarrier N, Haddock G, et al. Randomised controlled trial of cognitive-behavioural therapy in early schizophrenia: acute-phase outcomes. Br J Psychiatry Suppl. 2002;43:s91-s97.
5. Sensky T, Turkington D, Kingdon D, et al. A randomized controlled trial of cognitive-behavioral therapy for persistent symptoms in schizophrenia resistant to medication. Arch Gen Psychiatry. 2000;57:165-172.
6. Tarrier N, Yusupoff L, Kinney C, et al. Randomised controlled trial of intensive cognitive behaviour therapy for patients with chronic schizophrenia. BMJ. 1998;317:303-307.
7. Turkington D, Kingdon D, Turner T, for the Insight Into Schizophrenia Research Group. Effectiveness of a brief cognitive-behavioural therapy intervention in the treatment of schizophrenia. Br J Psychiatry. 2002;180:523-527.
8. Pilling S, Bebbington P, Kuipers E, et al. Psychological treatment in schizophrenia: I. Metaanalysis of family intervention and cognitive behaviour therapy. Psychol Med. 2002;32:763-782.
9. Torrey EF. PORT updated treatment recommendations. Schizophr Bull. 2004;30:617-618.
10. National Collaborating Centre for Mental Health. Schizophrenia: Full National Clinical Guideline on Core Interventions in Primary and Secondary Care. NICE. Available at:http://www.nice.org.uk/page.aspx?o=289559. Accessed April 5, 2006.
11. Kingdon DG, Kirschen H. Who does not get referred for cognitive behavior therapy for schizophrenia? Experience from an area where availability has not been limited. Psychiatric Serv. In press.
12. Ellis A, Harper RA. A Guide to Rational Living. Chatsworth, Calif: Wilshire Book Co; 1961.
13. Kingdon DG, Turkington D. The use of cognitive behavior therapy with a normalizing rationale in schizophrenia: preliminary report. J Nerv Ment Dis. 1991; 179:207-211.
14. Gould RA, Mueser KT, Bolton E, et al. Cognitive therapy for psychosis in schizophrenia: an effect size analysis. Schizophr Res. 2001;48:335-342.
15. Hansen L. The Influence of Cognitive Behaviour Therapy on Suicidality in Schizophrenia [doctoral thesis]. Southampton, UK: University of Southampton; 2004.
16. Dickerson FB. Cognitive behavioral psychotherapy for schizophrenia: a review of recent empirical studies. Schizophr Res. 2000;43:71-90.
17. Turkington D, Kingdon D, Weiden P. Cognitive behavior therapy for schizophrenia. Am J Psychiatry. 2006;163:365-373