To Name :
To Email :
From Name :
From Email :
Comments :

Cases that Test Your Skills

Treating persistent psychosis with cognitive-behavioral therapy

Persistent paranoid delusions keep Mr. K housebound. He refuses to try clozapine after other antipsychotic regimens fail to reduce his symptoms. Can another treatment help him?

Vol. 2, No. 3 / March 2003

History: A secluded life

Mr. K, 31, immigrated to Brooklyn as a child with his family. At age 19, he was working as an auto mechanic in a relative’s garage when he had his first psychotic episode. He was hospitalized and diagnosed with paranoid schizophrenia, which was confirmed by his subsequent course of illness and a recent structured diagnostic interview.

Since becoming ill, Mr. K almost never leaves his house and socializes only with his mother and sister. Unable to work, he helps around the house and often cares for his nieces and nephews. He has no history of substance abuse and has been faithfully taking his medication since his last hospitalization 8 years ago.

Mr. K’s outpatient clinic chart reflects concern about persistent negative symptoms. He hardly speaks in session. Even when “stable,” his social and vocational functioning has been poor. Therapeutic dosages of oral haloperidol, haloperidol decanoate, and fluphenazine have not worked, and trials of risperidone and olanzapine—administered in therapeutic dosages for at least 3 months—were only slightly more effective. Attempts to treat his social withdrawal as a depressive symptom equivalent, with use of adjunctive selective serotonin reuptake inhibitors, also have been disappointing.

We eventually discovered that his so-called negative symptoms were in fact the manifestation of persistent positive symptoms (Table 1). His social withdrawal stemmed not from lack of motivation but from ideas of reference and constant paranoid fears. He believed that if he left the house, a street gang would kill him or his family. When he did venture outdoors, he thought that strangers were ridiculing or insulting him or intended to brutally attack him.

Mr. K spent much of his time at home reading, occasionally visiting the public library just long enough to check out a few books. Family members convinced him to attend church services, but he could not interact with other parishioners because he feared they would find out something was “wrong” with him.

How would you address Mr. K’s positive symptoms? Would you try another antipsychotic after lack of response to five other agents?

Drug treatment: A clinical trial

After a thorough evaluation, Mr. K entered a clinical trial during which he began taking another antipsy-chotic. But after 4 months at high therapeutic dosages, his positive symptoms showed no change from baseline.

We then recommended that Mr. K try clozapine. His persistent paranoid delusions and lack of response to other antipsychotics made him an ideal candidate for this agent. Because he was compliant, motivated, and had available family support, we were confident that he could surmount the vicissitudes of a clozapine trial.

Mr. K refused to try clozapine, however. After so many unsuccessful medication trials over the years, he said he felt some (albeit minor) benefits from his current study medication and wanted to stick with it.

What treatment options remain for Mr. K?

Dr. Weiden’s and Burkholder’s observations

Although the newer antipsychotics have greatly improved outcomes in schizophrenia over the past decade, many patients still battle persistent psychotic (positive) symptoms despite compliance with these medications. 1 While clozapine remains the treatment of choice for positive symptoms, some patients cannot—or will not—take it because of its burdensome side-effect profile.

It is well accepted that supportive psychotherapy can help a person with schizophrenia confront the secondary issues of loss, disability, and stigma. But psychotherapy is rarely considered as an adjunct therapy for treatment-resistant positive symptoms. Skepticism about the role of psychotherapy is understandable, because older studies that demonstrated psychoanalytic psychotherapy’s lack of effect on schizophrenia’s positive symptoms 2 have driven psychiatric medical education and practice for the past half-century. Other non-psychoanalytic therapeutic approaches had not been studied until recently, so most of us generalized from the disappointing results of the psychoanalytic psychotherapy research.

Still, there is a resurgence of interest in using cognitive-behavioral principles to treat core schizophrenia symptoms. Several randomized, controlled studies—almost all performed in the United Kingdom—have demonstrated that a cognitive-behavioral therapy (CBT) approach, modified for schizophrenia, is superior to more traditional supportive therapies in treating persistent positive and negative symptoms. 3-5 Until recently, CBT for schizophrenia has generated little interest or research on this side of the Atlantic.

Continued treatment: A new approach

At this point, we decided to address Mr. K’s paranoid symptoms with CBT-based psychotherapy. The patient, whom we’d been seeing twice monthly for medication management, agreed to weekly 45-minute CBT sessions across 3 months. A third-year resident who had treated Mr. K during the antipsychotic clinical trial administered the therapy as described in the literature. 4-6

During the first sessions, we found that some of Mr. K's fears of leaving the house stemmed from living in a poor inner-city neighborhood with a high rate of violent crime. A few sessions later, Mr. K was able to question some of his beliefs that assassins had targeted his family.

By the fourth week, the therapist discovered that Mr. K liked to read and viewed the local library as a reasonably safe place. Mr. K agreed to visit the library once a week and to record his experiences commuting to and from there.

After roughly 8 weeks of treatment, Mr. K was visiting the library twice a week, usually for 90 minutes at a time. He reported that he “had a good time,” but still suspected other library patrons were talking about him. Upon exploring these suspicions, we learned that Mr. K feared other library goers viewed him as “stupid” and that this fear was exacerbating some of his paranoid delusions.

Table 1


Apparent negative symptom

Underlying psychotic symptom


  • Too preoccupied with hallucinations to be engaged in external environment
  • Avoids television or radio because of idea of reference when either is turned on

Social withdrawal

  • Fears leaving house because of perceived threat or danger
  • Stimulation of being with other people increases psychotic symptoms

Poor hygiene

  • Will not disrobe for a shower or bath for fear of sexual assault
  • Believes water is poisoned
  • Clothes considered protective, too risky to change

Once Mr. K acknowledged that his mental illness was not outwardly recognizable, we could convince him that his fear of appearing “stupid” should not keep him housebound. He began to visit the library more frequently. Although he was still anxious, it no longer took him all day to summon the courage to leave home.

After about 10 weeks of treatment, Mr. K began going on bike rides twice a week. We were concerned that he was making too many changes at once, but he insisted he felt more “comfortable” and enjoyed the exercise.

Did cognitive-behavioral therapy contribute to Mr. K’s improvement? How did the therapist’s treatment differ from accepted protocols?

Dr. Weiden’s and Burkholder’s observations

You might be thinking, “This sounds no different from what I do in practice!” The patient was reassured and encouraged to go out and live his life despite having symptoms.

In practice, however, some techniques used in the CBT approach to psychosis are quite different; some techniques are not intuitive, and some contradict most standard teachings of supportive psychotherapy in this country.

The CBT approach used for Mr. K differed greatly from traditional “medical model” supportive psychotherapy. The therapist:

  • rejected a “brain disorder” approach to describing his illness 7
  • used the stress-vulnerability model to explain positive symptoms
  • viewed psychotic symptoms as normal reactions rather than pathologic response
  • considered psychological factors behind specific psychotic symptoms (Table 2). 8

The “brain disorder” explanation. The Kraepelinian model, which characterizes schizophrenia as a degenerative brain disorder, drives patient education. For example, one brochure for patients and their families refers to schizophrenia as “a brain disorder like Alzheimer’s disease.”

Such a comparison could devastate a young adult who is overwhelmed by symptoms and after being told he had a “brain disorder” that “interfered with his cognition.” While this corresponded with our knowledge of schizophrenia, he took this to mean he is “retarded” and “stupid.”

Table 2




Psychosis lies on one end of a continuum

Psychosis may be the extreme end of normal cognitive, perceptual experiences

Delusions can be modified by others under some circumstances

Consider reducing delusional belief or distress by verbal interventions*

The stress-vulnerability model represents a more appropriate explanation of symptoms

Emphasizes stress as a possible cause of symptoms; specifying a diagnosis of schizophrenia can be contraindicated

Psychotic symptoms may be a normal response

Normalize symptoms, behavior as much as possible

Nature of psychotic symptoms is based on specific circumstances

Explore life events that have a specific psychological context for the individual

Psychotic symptoms may be secondary to congnitiive dysfunction

Trace the origins of hallucinations or delusions to specific cognitive overload or stress

* This is not meant to endorse a confrontational approach.

Telling the patient that he or she has schizophrenia– known in some clinical circles as “the S word”–is not necessary and may even be harmful in some cases. We’re not saying that giving a diagnosis of schizophrenia or using a medical model approach is wrong. However, patient education based on symptoms instead of diagnosis may be more conducive in some cases.

The stress-vulnerability model explains psychosis without having to use a diagnostic label. Mr. K’s previous understanding of schizophrenia dovetailed with his low self-esteem. His self-perceived stupidity also had discouraged him from confronting his paranoid anxieties.

Once he realized that unrelenting psychotic symptoms—not his IQ—held him back, we could form a treatment plan. We explained Mr. K’s paranoid symptoms with the stress-vulnerability model: His fears and suspicions worsened whenever he was under stress. This allowed us to sidestep the “brain disorder” model that demoralized him.

By no means, however, did we reject the notion that schizophrenia has a biologic basis. Even if “schizophrenia” is never mentioned, this diagnosis still guides treatment.

Viewing psychotic symptoms as normal reactions. The therapist extensively examined Mr. K’s safety concerns for himself and his family. The therapist acknowledged that these concerns were legitimate, as the patient lives in an inner-city neighborhood plagued by violent crime. Mr. K was then praised for his devotion to his family.

By applying normalization techniques, 9 the therapist found fact-based aspects of the delusional beliefs. Normalizing Mr. K’s safety concerns made him feel validated. From there, we could map out a plan for him to periodically leave his home. We devised a routine that addressed his safety concerns: He went to the library only in daylight. He chose a subway route that was less convenient but made him feel more comfortable. Once at the library, he called home to make sure his family was safe, then sat at an open table so he could watch other library patrons come and go.

Once this routine was established, we could address the more idiosyncratic delusions that caused Mr. K considerable stress and anxiety. We asked him to explain the evidence behind his belief that trained assassins were targeting his family. He could not do so and eventually admitted that this thinking was misguided. His fears gradually shifted from specific threats targeted at him and his family to nonspecific fears of the randomness and unfairness of life.

Placing psychosis in psychological context. We tried to understand the psychology of his paranoid thoughts while viewing his symptoms as part of a neurobiologic disorder.

What initially looked like “garden variety” paranoid delusions had a psychological meaning to Mr. K. His dread of public humiliation was intertwined with his fear of assassination. Once understood, these two fears could be isolated and became easier to treat. 10 We traced Mr. K’s fear for his family’s safety to his being the oldest male in a matriarchal household. Because his illness prevented him from assuming the role of breadwinner, all that was left was for him to guard his family from the imagined threat of assassination.

Did you miss this content?
Botulinum toxin for depression? An idea that’s raising some eyebrows