Cognitive-Behavioral Therapy in Severe Mental Illness

April 17, 2009

Psychotropic treatment can often prevent the relapse of psychotic and mood symptoms. However, many patients take medication intermittently or not at all; or the symptoms may be only partially responsive to medication. Therefore, there is a need for interventions that can supplement the effect of medication and improve treatment outcomes.

Psychiatric Times - Category 1 Credit


You must keep your own records of this activity. Copy this information and include it in your continuing education file for reporting purposes.

CME LLC is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians.

CME LLC designates this educational activity for a maximum of 1.5 AMA PRA Category 1 Credits™. Physicians should only claim credit commensurate with the extent of their participation in the activity.

CME LLC is approved by the California Board of Registered Nursing, Provider No. CEP12748, and designates this educational activity for 1.5 contact hours for nurses.

The American Nurses Credentialing Center (ANCC) accepts AMA PRA Category 1 Credits™ toward recertification requirements.

The American Academy of Physician Assistants (AAPA) accepts AMA PRA Category 1 Credits™ from organizations accredited by the ACCME.

Sponsored by CME LLC for 1.5 Category 1 credits.
Original release date 02/09. Approved for CME credit through April 2009.

Educational Objectives

• Strategies used in cognitive-behavioral therapy (CBT) for the treatment of severe mental illness
• Treatment approaches
• The benefits and challenges of using CBT

Who will benefit from reading this article?
Psychiatrists, psychologists, primary care physicians, nurse practitioners, and other health care professionals. To determine whether this article meets the continuing education requirements of your specialty, please contact your state licensing and certification boards.

Psychotropic treatment can often prevent the relapse of psychotic and mood symptoms. However, many patients take medication intermittently or not at all; or the symptoms may be only partially responsive to medication. Therefore, there is a need for interventions that can supplement the effect of medication and improve treatment outcomes.

Cognitive-behavioral therapy (CBT) has been found to be effective for those with severe mental illness.1 Recent meta-analyses have concluded that CBT has a definite effect on positive and negative symptoms in schizophrenia and there is evidence that it can have a sustained effect on attitudes about medication.2,3 The American Psychiatric Association practice guideline on schizophrenia refers to CBT as an evidence-based treatment, especially for persistent symptoms.4 Many other guidelines recommend CBT as well.5 Evidence of the effectiveness of CBT for bipolar and severe depression is more limited, but there are promising results for family work and social rhythm therapy.

Assessment and engagement
CBT involves working with specific symptoms: techniques to help manage delusions, hallucinations, and negative symptoms are based on the development of a good therapeutic relationship between patient and therapist and formulation of the experiences, competencies, thoughts, feelings, and actions of the patient.6

Engaging people who experience psychotic symptoms can be difficult: paranoia diminishes trust, and cognitive impairment and catatonia affect communication. Developing a relationship is a process that requires skill, attention to detail, and adaptation to the patient’s particular concerns.

A good clinical assessment is essential because it provides a wealth of valuable information to inform conceptualization and subsequent intervention and, in itself, can be therapeutic. Details gathered about the first episode of psychosis can be particularly informative. Analysis of the period preceding a psychotic episode can enable the patient to reconsider the conclusions he or she formed at the time, especially if the conclusions evolve into delusional beliefs.

Making sense of the experiences that led to psychosis involves assembling relevant information about predisposing, precipitating, perpetuating, and protective factors with key thoughts, feelings, and behaviors. Connections can be made between precipitating causes and symptoms (eg, stimulant/hallucinogenic drugs and initial psychotic experiences, often viewed as “bad trips”) that further evolved into persistent psychotic symptoms.

Formulation and process Once the precipitating causes are understood, work on delusional beliefs follows. Initially, the process of listening to the patient’s story of how his beliefs developed can improve the therapeutic relationship, allowing discussion and gentle probing about symptoms to begin. It is important for the therapist to begin by “suspending judgment” regarding beliefs. Beliefs may appear to be persecutory (“white vans are following me with intent to destroy me”) or grandiose in nature (“I can influence the weather”). They may include thought broadcast (“the world can hear me thinking”), thought insertion (“you put these ugly thoughts in my head”), or delusions of reference (“the TV show is referring to me”). All beliefs have meaning and function and must be broached with respect and compassion.

It is challenging for any of us to begin considering alternative explanations and ways of being when beliefs are deeply entrenched. A collaborative examination of the evidence offered in support of beliefs can be followed by inquisitive planning of simple experiments. A sample dialogue between the therapist and patient follows.

Therapist: It sounds like you feel very distressed by the belief that your family is monitoring your phone calls. This is an understandable reaction to this belief.

Patient: It’s horrible; I can’t trust anyone. My own family is doing this to me!

Therapist: I’m wondering how much of you believes that your family is doing this, from 0% to 100%?

Patient: I am 95% convinced that they are. All the evidence is there!

Therapist: Okay, so you are pretty convinced. However, there is 5% of you that isn’t convinced. Because of the high cost (distress) of your belief, it may be a good idea for us to check out the evidence for this belief. How does this sound to you?

Patient: The most significant piece of evidence I have is that there is a recording device inside the phone. So perhaps we can plan some simple checks with it? I also saw my mom studying the phone bill last month, but at the time I was too scared to ask her why.

Therapist: So the recording device and your mother studying the phone bill are 2 things we could potentially explore further…?

Alternative explanations can be explored. For example, a patient who believes he is being electrocuted because his fingers tingle is given information about anxiety and a controlled experiment is undertaken to reproduce the symptoms. By modifying the misinterpretation of events that served to reinforce the delusional belief, the cognitive cycles that maintained the delusion are broken. Exercises that use Socratic questioning to examine beliefs and to foster discussion are very useful in assessment and relationship building. Gentle probing by the therapist can provide additional information and build trust on the part of the patient.

When a patient lacks an absolute conviction in his beliefs, the process of gentle questioning can begin to evoke alternative explanations. When beliefs are very strong and a “chink of insight” cannot be found to build on, such discussion eventually begins to become repetitive and it may be necessary to agree to disagree. Frequently, the patient is open to moving on to other, more important topics about underlying concerns and life goals. Once these are addressed, the beliefs may change. By focusing on why the beliefs are important to the individual and empowering him by teaching him what he can do if beliefs are causing distress (eg, through coping strategy enhancement), the therapist can improve collaboration and begin constructive behavioral change.

Addressing the voices
Hallucinations are frequently but not always distressing. Some patients describe positive aspects, such as companionship or guidance; nevertheless, hallucinations are disturbing and negative in content and effect for most patients. A structured approach to assessing, understanding, and reattributing voices or other hallucinations is used.

This involves establishing the nature of the perception (“like me speaking to you?”) and then the uniqueness of the experience (“just you, nobody else can hear them?”). Inquiry into beliefs about the origin of the perceptions follows (“why do you think no one else hears them?”); followed by discussing the reasons (“It’s God”; “But would you expect God to say such unpleasant things to you?”).

Alternatively, especially if the patient is uncertain of the origin of the voices, psychoeducation and normalization about the nature of voices can be helpful. For example, explaining the role that the Broca area of the brain plays in inner speech and the similarity to “dreaming awake” can help patients reattribute voices to internal phenomena (“my mind is playing tricks on me”) rather than from external sources (“the neighbors are talking about me”). It then becomes more relevant to use personal ways of coping with them (eg, relaxation, distraction, and taking medication). Coping strategies that involve direct engagement with the voices have been used successfully in patients who have developed insight into the voices but in whom they persist and continue to cause distress. These include the development of an assertive dialogue or use of mindfulness techniques.

Negative symptoms
There is accumulating evidence that CBT can have an enduring effect on negative symptoms, including amotivation and social withdrawal.7 Rehabilitation methods-such as social skills training, target setting, and behavioral experiments, supported by cognitive interventions-are used for negative symptoms. For example, a patient may set himself a task but be unable to complete it. Through inquiry, the patient’s belief that when he goes outside he is being talked about in the street is uncovered. Work on this belief involves reasoning, ie, reality testing. The emphasis is on ascertaining precisely who is talking about him, what they are saying, and why they may be saying it. Diaries or detailed recall of circumstances can help unblock resistance. Assessing expectations and pressures with families can also be useful. Remember, you cannot push patients out of negative symptoms. It is essential that they are allowed to reestablish control.

Readily achievable and collaboratively developed short-term goals establish momentum. The goals are often initially very simple (eg, go downstairs for coffee in the late afternoon). When the initial goals are achieved, further steps are built on that foundation. Consider suggesting that the person take time off when he has been trying but repeatedly failing to achieve a goal: he may feel less pressured.

Improved concentration and cognitive ability enables work on positive symptoms that interfere with progress. When negative symptoms have emerged during periods of institutionalization, the identification of motivating factors-often long forgotten-can begin the process of reengagement in conversation and social interaction and subsequently set the stage for specifically therapeutic work. CBT techniques for behavioral activation that involve activity scheduling, mastery and pleasure rating (gauges how well the patient has mastered the task and how much he has enjoyed it), and graded task assignment can provide structured support.

Implementation of CBT in severe mental illness
Although sessional CBT is generally provided by clinical psychologists or nurse therapists, it is important for psychiatrists to be involved in the initial work with patients to develop agreement with them to be referred for psychological work. Work on attribution can be fundamental to this process: if the patient does not agree that there is anything wrong, why should he see a therapist to discuss anything?

Focused work on hearing voices, delusions, and negative symptoms can be successful in brief sessions with patients. This may be all that can be offered where availability of skilled therapists is limited. Psychiatrists can take the lead in alleviating the dearth of trained providers by advocating for improved availability through training and education.

After the patient has had sessions with therapists, psychiatrists often continue medication management and provide brief booster sessions with discussion and reinforcement of lessons learned and gains made.

Common pitfalls and challenges
Strong beliefs take time to change and CBT is rarely a quick fix. The major changes seen initially are usually behavioral (eg, increased social involvement rather than cognitive improvement and ability to cope with distressing symptoms). Most therapy courses involve 16 to 20 sessions over 6 to 9 months, and change often continues after completion of therapy. For acutely psychotic patients, CBT can improve engagement; combined with the effects of medication, CBT can allow work on symptoms to commence as agitation and thought disorder are reduced.

Patients with profoundly negative symptoms can be challenging to engage and often a “befriending” approach is helpful. Seeking out areas of interest with appropriate self-disclosure may be a necessary first step for the success of the work on formulation and symptoms, which will follow. Cultural issues are also important to consider and require adjustments to therapy.8


Ellie, 50 years old, was housebound and lived alone. She had a long history of mental illness and had been abused as a child. She stated she had an “unhappy childhood.” Also, she had had a “lost year” of drinking and isolation. As a teenager she had 2 hospital admissions related to drugs, alcohol, and depression, and 4 years ago she was hospitalized for alcohol-related hallucinations. Since then, despite a history of overdosing, self-harming, depression, and disengagement, she has managed to stay out of the hospital. A violent incident with her ex-partner coincided with the onset of auditory hallucinations.

Rapport with Ellie was established using a nondirective conversational style: befriending was quite successful in reducing her initial anxiety. Ellie and the psychiatrist worked to normalize and decatastrophize experiences using the stress vulnerability model and psychoeducation. The cognitive model was explained early in therapy: links between the thoughts (perceived as voices), feelings (distress and anger), and behavior (drinking and withdrawal) were creating a self-perpetuating cycle. A history and time line were established, and symptom assessment was completed using a standardized rating scale.9

Ellie was highly anxious and depressed, with suicidal thoughts and behaviors. She said, “My life as a whole is black and gray. I’d love some respite from myself.”

Ellie experienced one voice but minimal delusional belief. She was assessed as relatively independent (moderate competence and performance on the Social Functioning Scale) but very socially withdrawn, with a lack of relationships, no social and recreational activities, and unemployment.10 She viewed the voice she heard as malevolent, omnipotent with no benevolence, using the Beliefs About Voices Questionnaire.11

Based on the available information and further discussion with Ellie, a list of problems was formulated and agreed on (Table). Most problematic was the experience of hearing voices. Analysis of the last experience clarified that hearing the voice was an activating event. An emotional reaction to its content (anger, shame, anxiety, and guilt) was the consequence, and avoidance, isolation, and obeying the voice was the behavioral response.

The underpinning beliefs were, “What the voice says is true; it will never stop; it is in control and I have no control over my actions.”We formed the problem statement, “When I hear the voice it makes me angry, then I’m too terrified to go outside because I am frightened I could hurt someone, which affects my quality of life.” A plan was developed to address her psychotic symptoms.

Negative core beliefs linked to past events were identified. Ellie had developed unhelpful automatic thoughts and assumptions (eg, “I’m not good enough”) to manage the negative beliefs. During therapy, differing levels of belief were evoked using thought records completed during sessions and as homework. This gave her a simple yet powerful demonstration of the way thought influences and maintains feelings.

Ellie’s experience of hearing the voice was further clarified through assessing attributions, focusing on specific beliefs, and maintaining a voice diary to clarify connections with daily events. The voice was reminiscent of her grandfather, “an evil man.” Both he and the voice used her full name (Eileen), while no one else did. Relating to the voice was set within her interpersonal history: once the gap between negative perception of herself and overly positive attitudes to significant others decreased, it was easier to challenge the voice. She and her doctor discussed how voices and beliefs about voices can arise from childhood abuse and other precipitating events. A responsibility pie chart helped Ellie reattribute blame for the abuse from herself to the adults responsible. Focusing on the psychological origins and the relationship with her grandfather helped Ellie deconstruct the power beliefs and develop some boundaries with the voice.

Ellie achieved self-identified short-term goals related to the problem statement. These were rated regularly on a progress scale of 0 to 8 and plotted graphically to aid visualization of change. Avoidance decreased despite hearing the voice. As fears were disconfirmed, Ellie’s world expanded.

Behavioral experiments, properly designed and carried out, can modify beliefs and assumptions more powerfully than verbal techniques in the office. Ellie made regular late-night trips to the supermarket, where she would avoid speaking to anyone except the cashier. For the experiment to expand her contacts, she asked an assistant “where are the tomatoes?” The process of planning (designing the experiment), experiencing (the experiment itself), observing (examining what happened), and reflecting (making sense of the experiment) was used to maximize gains from this successful experience, and it was used as a foundation to increase levels of social contact.

Concentrating on Ellie’s experience of voice hearing positively changed the other problems: not getting out, difficulties in sleeping, and hopelessness. Reassessments at 6 months highlighted progress. Ellie had become more insightful regarding hearing voices, and her mood and selfesteem had increased. Ellie had become “her own therapist,” learning from and initiating new experiences. She felt hopeful about implementing positive change.

Recovery is a unique individual journey that may not mean being completely symptom-free. The ability to cope and reengage may be the most appropriate goal. This is illustrated by Ellie’s reflection that:

Life as a whole is no longer black and gray. I have color. Some days are brighter than others. There are times of love, light, and happiness; I can be lost in the moment. My life may not be anything special compared to other peoples’, but it is now my life, and the dark times make the okay times that much more special.


CME Post-Test


1. Wright JH, Turkington D, Kingdon D, Basco M. Cognitive Therapy forSevere Mental Illness: An Illustrative Manual. Arlington, VA: AmericanPsychiatric Publishing, Inc; 2008.
2. Wykes T, Steel C, Everitt B, Tarrier N. Cognitive behavior therapy forschizophrenia: effect sizes, clinical models, and methodological rigor.Schizophr Bull. 2008;34:523-537.
3. Rathod S, Kingdon D, Smith P, Turkington D. Insight into schizophrenia:the effects of cognitive behavioural therapy on the componentsof insight and association with sociodemographics-data on a previouslypublished randomised controlled trial. Schizophr Res. 2005;74:211-219.
4. American Psychiatric Association Steering Committee on PracticeGuidelines. American Psychiatric Association Practice Guidelines for theTreatment of Psychiatric Disorders Compendium 2006.Washington, DC:American Psychiatric Association; 2006.
5. National Institute for Clinical Excellence. Clinical Guideline: Schizophrenia(Clinical Guideline 1). London: Department of Health; 2002.
6. Kingdon DG, Turkington D. Cognitive Therapy of Schizophrenia. NewYork: Guilford; 2005.
7. Turkington D, Sensky T, Scott J, et al. A randomized controlled trialof cognitive-behavior therapy for persistent symptoms in schizophrenia:a five-year follow-up. Schizophr Res. 2008;98:1-7.
8. Rathod S. Expansion of psychological therapies. Br J Psychiatry.2008;193:256-257.
9. Lancashire S. Revised Version of the KGV Scale. Manchester, UK: Universityof Manchester; 1994.
10. Birchwood M, Smith J, Cochrane R, et al. The Social FunctioningScale: the development and validation of a new scale of social adjustmentfor use in family intervention programmes with schizophrenic patients.Br J Psychiatry. 1990;157:853-859.
11. Chadwick P, Birchwood M.The omnipotence of voices, II: the BeliefsAbout Voices Questionnaire (BAVQ). Br J Psychiatry. 1995;166:773-776.

Evidence-Based References
Kingdon DG, Turkington D. Cognitive Therapy of Schizophrenia. NewYork: Guilford Press; 2005.
Wright JH, Turkington D, Kingdon DG, Basco M. Cognitive BehaviourTherapy for Severe Mental Illness: An Illustrated Guide. Arlington, VA:American Psychiatric Publishing, Inc; 2008.