From: Haddock, G. & Slade, P.D. (Eds.) Cognitive-Behavioural interventions with Psychotic disorders, London: Routledge, ps. 103-115


Chapter 6


Douglas Turkington and David Kingdon


An integrated treatment strategy with a community key worker is the backbone of community management of patients with schizophrenia. A range of accommodation options. after discharge from mental hospital (Kingdon et al., 1991), befriending schemes (Kingdon et al., 1989) and crisis intervention accommodation (Turkington et al., 1991) complement neuroleptics, standard rehabilitative and family therapies. However, whether the patient will be able to engage in such an integrated strategy will depend on his or her own level of insight and attitudes to the symptomatology, as well as the attitudes of the psychiatrist and the community mental health team. The issues of engaging the psychotic patient in discussion about their symptoms is therefore fundamental to issues of recovery, relapse, compliance and coping. The classical psychiatric teaching that discussing psychotic symptoms with a patient is at best useless and at worst an exacerbating factor (Scharfetter, 1980) has percolated throughout psychiatry and psychiatric nursing. This has led to a stand-off approach to those psychotic symptoms which are most distressing to the patient, for example persistent voices or passivity phenomena. As a result of this approach, patients have often drawn catastrophic conclusions of madness and untreatability. Cascades of negative cognition, dysfunctional affect and disordered behaviour often follow with lack of treatment compliance and deteriorating social function. There is the need to address the key area of what we say to the schizophrenic patient and how to say it in order that they might be engaged as active agents in their own treatment. To this end, we felt that engaging the patient in the collaborative production of an explanatory normalising rationale/model of symptom emergence was a crucial first step in the relationship with the patient (Kingdon and Turkington, 1991). With the patient 'on board' a variety of techniques can then be used to reduce distress and improve compliance. It is only really in the last twenty years that an accumulating body of research evidence has started to indicate that such an approach is safe and feasible (Kingdon et al., 1994). Our study (Kingdon and Turkington, 1991), of sixty-four patients with schizophrenia treated with cognitive-behaviour therapy (CBT) revealed high acceptability and a low readmission rate with no suicides or homicides over a seven year period. More recently we have performed a study of CBT versus befriending in the treatment of schizophrenia on the basis of random allocation (Kingdon et al., in press ). This study shows that while cognitive therapy and befriending are both useful in acute psychosis the cognitive therapy was much more beneficial, and to a statistically significant degree, with those patients with chronic drug-resistant symptomatology. The cognitive-behavioural repertoire continues to expand into a diversity of therapeutic modalities including coping skills enhancement (Tarrier et al., 1993; see Chapter 11), cognitive remediation (Green, 1993) and early intervention (Birchwood et al., 1989; see Chapter 10).


This chapter covers the rapidly growing area of cognitive-behavioural therapy in schizophrenia (Kingdon and Turkington, 1994). Our approach stresses the use of developing a rationale with patients to explain symptom emergence' and decatastrophise the psychiatric diagnosis of schizophrenia.




It is worthwhile outlining the stages of therapy which are gone through with a schizophrenic patient when CBT is being used. The rate-limited step for all future progress is, of course, the initial one of engaging and building up rapport. This is worth exploring in some detail as, if the therapy gets off on the wrong foot, it can become increasingly difficult to test out the reality of psychotic symptoms. The following would seem to be the key points.

1. Empathy, warmth, genuineness and unconditional acceptance; these are features of all good psychotherapists and apply equally to psychosis as to neurosis.

2. The experience of interacting with psychotic patients and knowledge of the typical modalities of psychotic expression (for example thought disorder, hallucinosis, delusional perception, systematisation of delusions, etc. ). There is no substitute in this regard to spending many hours dealing with the problems of the severely mentally ill.

3. Word perfect accuracy and consistency. When building up a relationship with a patient with schizophrenia it is vital not to do anything either verbally or non-verbally to invalidate their experience. Confrontation of a delusional belief, for example, leads only to invalidation of the patient's account, weakening of the therapeutic alliance and often entrenchment of the delusion. If a patient reports that their house is bugged, the appropriate response is not to say that this is not so, but to acknowledge that this would be possible, but very unlikely, unless the patient has been involved in espionage or intelligence activities. The next step would be to collaboratively investigate the matter further using diaries and planned homework tasks. The tightrope between non-confrontation and collaborative reality testing needs to be walked in each treatment session.

4. Agreement to differ. At times patients who are very preoccupied will continue to attempt to persuade the therapist of the reality of, for example, their telepathic powers. This can become counter-productive and at times the therapist has to accept their right to their own opinion. This again should not be seen as confrontation, merely as an individual's right to express an alternative point of view. The following therapy extract illustrates this (the patient was very preoccupied with beliefs about his telepathic powers).


Patient: I need to know, do you believe in telepathy or not?

Therapist: The research position on this is that the possibility exists that some people under certain conditions of extreme emotion or stress may be able to pass feelings or images to a close relative but that it is not a viable method of communicating thoughts on a day to day basis. The homework from this session is for you to go up to the library and look up the research on this subject on reported occurrences of spontaneous telepathic phenomena. Once we have read this together we can then work on your own experiences. [Patient was given a brief reading list and an article on the subject. ]

5. Tactical withdrawal. If your patient is becoming distressed by any particular approach or subject then this should be noted and appropriate support given. .The subject matter may be very important and can be tentatively returned to later in therapy.

6. Allowing the waves of psychosis to roll over you. Therapists are very often put off by the large quantity of seemingly incomprehensible psychotic material produced by patients in the early sessions. This will gradually become comprehensible in light of the patient's life history, personal schemas and misinterpretations of normal phenomena. The cognitive therapist, through the engaging phase, is gradually working toward an early formulation of psychotic symptom emergence which will allow the mobilisation of specific techniques and must be able to tolerate a period when much of the data will be incomprehensible.

7. Teaching the cognitive model. This is done with the help of some reading material, numerous case examples and often personal disclosure of how the therapist has tackled one of his or her own problems using cognitive therapy, for example public speaking anxiety. The model describes how thoughts can cause feelings and behaviour. Thoughts may be amenable to change and can be tackled with a superficial emotion, for example anger at a member of staff or carer may be reduced to annoyance using less demanding rational responses.




The next phase after engaging is to provide an explanation of puzzling and distressing symptoms and to deal with catastrophic cognitions concerning insanity and possible treatment methods. Often patients with schizophrenia are not given the diagnosis themselves due to the clinician 's 'catastrophisation'. The patient may learn of the diagnosis from a carer, work it out for themselves or recognise the symptoms from an article or television programme. The conclusions that the patients draw from not being told the diagnosis are inevitably catastrophic. Typical cognitions which we have had reported include the following: "I am mad", "I will be locked up", "I will be beaten and tortured in an asylum", "there is no hope", "they will strap me down and give me electric shocks", "I will turn into a vegetable" and "they will give me very strong drugs and I will be a zombie". These cognitions are culture syntonic and reflect typical views of schizophrenia given in the media. Jack Nicholson's portrayal of psychiatric treatment in One Flew Over the Cuckoo's Nest contains many of the anxiogenic images which dominate a patient's thinking when they draw the conclusion that they must have schizophrenia. These cognitions are future orientated and catastrophic. They also contain some truth but are mostly anachronistic and terrifying images of treatment and outcome in schizophrenia which cause marked hyperarousal and anxiety. There is certainly evidence that such an emotional state can exacerbate psychotic symptoms (Slade, 1973). Depressive cognitions of hopelessness cause lowering of mood, withdrawal and suicidal ideation. The combination of hyperarousal and depression often leads to poor compliance with neuroleptic drugs and day care. The anxiogenic-depressive cognitive set can therefore become a self-fulfilling prophecy leading to maintenance of psychotic symptoms and a poor outcome. The rationale which is discussed with the patient describes the typical symptoms and possible genetic predisposition to respond in that way. The vulnerability-stress hypothesis (Zubin, 1987) is explained in detail to the patient in relation to the stressor or accumulation of stressors which preceded the emergence of symptoms. The patient is led towards an understanding that there is probably a discernible reason or reasons why the symptoms have occurred and the possibility that anyone stressed in certain ways could become psychotic. If there is a family predisposition to respond in this way this can be fully explored to help the patient to feel less different and isolated. It is often useful to itemise the types of stressors which can typically produce psychotic symptoms in 'you' or 'I'.


1. Sleep deprivation. There is evidence that lack of sleep can lead to illusions, hallucinations and paranoid ideation (Oswald, 1974); as an example, medical staff working without sleep for prolonged periods have been reported to behave in bizarre and irrational ways. Again this is often best backed up with appropriate literature and given to the patient. As psychotic patients have often been sleep deprived at the time of symptom onset this can be taken as part of the explanation of what has been happening. This may lead to a reduction in anxiety and improved compliance with neuroleptics.

2. Post-traumatic stress disorder. As very significant life events often precede the development of psychosis it is useful to discuss the types of psychological reactions which people can have to extreme stress. Again the patient is given the literature showing that hallucinations were very common amongst veterans of the Vietnam war (Wilcox et al., 1991). This again can help patients to feel less alienated by their symptoms.

3. Sensory deprivation. Bed rest in a darkened room and water tank immersion can both lead to the development of simple and eventually complex hallucinations. The latter more intensive experimental situation is by far the more powerful of the two in inducing hallucinosis. Often patients show interest in reading this material (Slade, 1984) as they may identify their long-term isolation, albeit much less intense, with such short-term sensory deprivation.

4. Hostage situations. This work is described by Siegel (1984), and patients often find it helpful to discuss this material. In particular the communication problems and isolation of the hostage are often identified with by the schizophrenic patient.

5. Solitary confinement. Grassian (1983) describes the emergence of psychotic symptoms in prisoners kept for prolonged periods without contact with others. Schizophrenic patients often describe similar periods of isolation prior to psychotic breakdown.

6. Sexual abuse. Recent work has shown that hallucinosis is surprisingly common in those patients who have undergone repeated or particular brutal sexual abuse (Ensink, 1992). Most patients will not be ready to disclose such traumatic events early in treatment but when they eventually do, explaining that there is a possible connection between symptoms of psychosis and such treatment can help to explain symptom development.


This can improve the therapeutic alliance, help the patient feel more 'normal' and less alienated and allow work on non-threatening and explanatory areas prior to tackling the patient's own personal symptomatology. Therapist and patient can begin to work out collaboratively a formulation of how the symptoms may have emerged and also to decatastrophise the term schizophrenia, which is a descriptive term for those patients suffering from certain of these symptoms. The importance of rationale has been well described by Romme and Escher (1989, and see also Chapter 8). They showed that many people who do not have schizophrenia hear voices and those who cope well with them may be those who have developed clear rationales to explain them. The most useful rationales included mystical, parapsychological, Freudian and Jungian explanations. Biological rationales were not strongly favoured and indeed it is our experience that patients are loath to accept biological rationales early on although they may adopt them later in treatment.




In developing a rationale, a close examination of the antecedents of psychotic breakdown may be necessary .The importance of this was shown as far back as the early 1950s (Beck, 1952). The crucial period leading up to the psychotic breakdown should be worked through most often with inductive questioning. Other useful techniques for exploring this period include imagery and role play. Key cognitions can be detected from this period pointing to underlying schemas concerning achievement, approval and control. Sessions around this stage of the therapy can be very rich with important material. If the patient does not have paranoid delusions it is often helpful to tape record sessions for playback as homework. Care is needed as inductive questioning performed in relation to the immediate pre-psychotic period can hit upon sensitive and painful material. It may be necessary in some cases to tactically withdraw if this occurs. It is our experience that in perhaps 70 per cent of cases this stage of therapy is viewed as being integrative and emotionally helpful. In the other 30 per cent there can be some degree of flare-up of psychotic symptoms. However, we have always found this to be containable using a variety of supportive techniques. It is also worthwhile stating that informal follow-up has not detected any suicides or homicides in our schizophrenia cohort over a nine-year period. This is an important stage of therapy which needs to be handled sensitively.




This would generally be the next stage of therapy and its importance lies in the fact that affective symptoms often seem to be delusionally misinterpreted and can also lead to poor compliance with psychosocial and biological treatments. Anxiety symptoms should have reduced with the collaborative production of a normalising rationale. Some persist, however, as illustrated below.


Patient: Planes that pass over my house have the power to move my intestines. There are communist agents on board working a ray gun that causes this.

Therapist: Let's keep a diary of exactly how often this happens in the next week and exactly what it feels like for you.

Patient: [next session] This has happened at least five or six times each day but the diary shows that the planes could be linked to two or maybe three episodes in total.

Therapist: What does the diary show was the cause of the other episodes?

Patient: Doors banging, dogs barking and loud music.

Therapist: I am going to give you a handout describing the main anxiety symptoms. Can you recognise any of these in yourself?


The patient identified sensitivity to sound, abdominal churning, tension, worry, irritability and episodic palpitations as being symptoms which he himself suffered from. As he began to learn about anxiety and practice progressive muscular relaxation on a daily basis the particular delusional misinterpretation concerning the planes and communists was dropped. In relation to depression, mild and reactive forms are best tackled cognitively. However, antidepressant drug treatment is sometimes required to complement CBT at this point in treatment. Propanolol can also be a useful adjunct to relaxation for somatic anxiety symptoms.




The appropriate techniques can be listed in relation to the main symptom groups, i.e. delusions, hallucinations and thought disorder.




These are defined as false beliefs held in spite of evidence to the contrary and out of keeping with the patient's social, cultural and educational background. Whilst it is certainly true that a patient cannot be argued out of a delusion (confrontation generally leading to entrenchment), the effectiveness of collaborative gathering and assessment of evidence in relation to the delusion has not yet been fully tested. Pioneering early work in this area (Watts et al., 1973) did show that delusional beliefs could be modified by psychological intervention (see Chapter 2 for a review of this work). This classical definition of delusion cannot therefore be regarded as scientifically evaluated. Strauss's analysis of schizophrenic delusions showed them to be points on continua of function (Strauss, 1969). He believed that this spectrum extended from normal belief through overvalued ideas to delusions. Delusions may then be regarded as fluid structures which, given the appropriate therapist strategies and attitudes, are amenable to gradual weakening. Those delusions which are held most rigidly frequently involve high investment of self-esteem. The following CBT techniques may be used in working with schizophrenic delusions.


1. Peripheral questioning. This is a technique for sensitively helping the patient to draw his own conclusions about some of the inconsistencies surrounding the delusion. It is important not to start questioning at the heart of the delusion as this may be too threatening unless these are the key issues that the patient wishes to discuss. Initially the approach is to gather as much pertinent information as possible. Peripheral questioning can then proceed on to some of the 'nuts and bolts' of the delusion. Socratic questioning can then be sensitively used in relation to the implications of the delusion. Peripheral questioning is illustrated by the following therapy examples.

Patient: My house is bugged and this interview room is too.

Therapist: What would the bugs look like? Let's try and find them.

Patient: They are set inside the concrete of the walls.

Therapist: What kind of bugs can hear through concrete?

Patient: I don't know.

Therapist: But I understand bugs to be like small microphones, they need to be out in the open or at least only superficially hidden. Will we look for them here and then perhaps we can check out at home? [Homework - reality testing with diary.]


This excerpt from a therapy session reveals the kind of deficit in real world knowledge, i.e. understanding of microphones, often exhibited in relation to delusional thinking.


Patient: I am being followed everywhere I go.

Therapist: What kind of people are normally followed in this way?

Patient: People that are being terrorised by someone.

Therapist: Think about films that you have seen and books you have read. What kind of people get followed?

Patient: Spies, terrorists, politicians, royal family, pop stars.

Therapist: You are not any of those sorts of person are you?

Patient: No, but I felt like royalty when I was younger as my twin sister was always with me and we did everything together.

Therapist: Perhaps this thing about being followed has in some way arisen from your separation from your twin sister. What we really need to know is whether it is a feeling of being followed or whether there are actually people there. Will you record the exact appearance of the followers (what they are wearing, facial expressions, etc.) and circumstances of each episode in which you feel you are being followed and record it in a diary on a day to day basis [reality testing homework].


The tactics here are to work in a non-challenging and gradual way from the periphery of the delusion towards the centre.


2. Reality testing. As above, tests of specific delusions need to be set up in a collaborative way. It should not be done in an abstract way, but with a firm record kept of findings. In the early stages it is often useful to involve a community key worker to go with the patient on these reality testing exercises and to help him or her to record the findings. In analysing the findings a 'guided discovery' mode is used and the patient helped to generate possible explanations for the findings. This kind of work is normally enjoyed by patients and can be applied to many situations and beliefs.


3. Working through the emotional investment of a delusion. Delusions of persecution are often invested with fear or anger. Similarly, grandiose delusions are often invested with elation, and delusions of control with anger and depression. If it is proving difficult to tackle a delusion in the ways outlined above, a cognitive therapy approach aimed at reducing the emotional investment may be effective. Techniques used are those which have been well described elsewhere, as in the treatment of anxiety and depression (Blackburn and Davidson, 1990). Once the emotional investment has been reduced it is possible to proceed with reality testing.


4. Using schema level CBT with delusions. The pertinent schemas are often detected at the time of examining the antecedents. Other techniques include identifying common themes in cognitions, examining meanings of specific events and use of the downward arrow technique (this allows the therapist to follow a distorted automatic thought down to the underlying belief which generated it). It often seems to be the case that delusions emerge at times of invalidation of key personal schemas. A delusion would then be regarded as a distorted schema which serves the function within the psyche of protecting against devastating loss of self-esteem. Where this is the case, schema work could be expected to help patients gradually weaken the intensity with which a delusion is held, or start to behave in a less restricted way. This may, however, be at the expense of precipitating depressive symptoms. Whilst this does not occur in all cases it should be watched for and may require appropriate support in terms of structured activities, supportive counselling and the use of antidepressant drugs. Standard techniques for schema level CBT are described elsewhere, for example Beck et al. (1979).




It is surprising how often voices are accepted in schizophrenia without the development of a rationale to explain them or research to identify their exact location. A vital stage in the treatment of hallucination is to help the patient do exactly this. The activation of appropriate coping strategies often follows. A critical collaborative analysis of the voices begins by generating hypotheses about their origin. Very often patients will have no immediate answer as to what the voices might be, but using a guided discovery modality, possibilities can be generated, percentage belief allocated and then homework devised to test out the options. This usually leads to a reduction in belief that the voices actually exist in the external world and may lead to an increased belief that the voices could be of mystical and parapsychological origin. Consequently patients may start to accept, particularly when considering the content of the voices, that they might in fact be the product of their own mind manifested in an unusual way, due to severe stress. If either a variable external rationale or a belief that the voices could be internally produced can be tentatively accepted, then patients are often more enthusiastic at pursuing coping strategies (Falloon and Talbot, 1981). In relation to the voice content, this is best dealt with either by considering the voices as automatic thoughts or by acting directly in contradiction to them. In the former approach the patient records ( often with key worker help) the main statements made by the voices. The truthfulness of these statements is then considered and rational responses generated. These responses can then be recorded onto an audio tape for the patient to listen to when the voices are active. Behavioural homework experiments, e.g. acting against the voices, has a similar effect to taped rational responses. The voices may initially become more severe, but generally, after a period, fade into the background. This would appear to be a phenomenon related to cognitive dissonance. Patients need to be warned that voices may become a bit worse before improving. The following case relates to a young lady with neuroleptic resistant hallucinosis who had come to believe that she was a witch because the voices told her so, and dressed accordingly in long black gowns etc.


Therapist: Our work on the voices seems to show that we are unsure where they are coming from.

Patient: I think they are either ghosts [50 per cent belief] or thoughts [50 per cent]. I think they might be thoughts because quite often I hear my mother's voice and she is still alive.

Therapist: Have you tested out whether it could be your mother?

Patient: Yes, she denies it and my friend couldn't hear them either.

Therapist: If they are your own thoughts then maybe they aren't so scary and maybe you don't need to believe them if they are unreasonable.


Rational responses were formulated and used with a measured reduction in voice frequency and intensity. The next stage was the behavioural homework of dressing in her normal clothes and wearing make-up etc. Voices initially flared up and then further settled.


Thought disorder


One of the main CBT techniques involved in treating schizophrenic thought disorder is thought linkage. In this technique the patient is repeatedly asked to clarify the links between disconnected fragments of speech. Neologisms are not accepted and the patient is asked to explain where the word comes from and what it means. Video/audio taping of 'incomprehensible' interviews can be extremely helpful in case formulation. Often the main themes, including stressors, can be disentangled and addressed individually with more focus leading to increased comprehensibility (Turkington and Kingdon, 1991).




The maintaining factors for negative symptomatology are frequently preoccupation with positive symptoms, neuroleptic side effects and depression. The process of therapy is described in tandem with the use of low dose medication and standard behavioural rehabilitation approaches. This may allow negative symptoms to gradually improve as positive symptoms ameliorate and motivation improves. However, the absence of a range of community psychiatric facilities and support options makes attempts to alleviate negative symptoms much more difficult (see Chapter 9 for an account of negative symptom management).




CBT booster sessions and delineation of the specific relapse prodrome for individual patients do much to arm the patient against the possibility of future relapse. A list of telephone numbers to access staff and other response options needs to be formulated.




The use of cognitive-behaviour therapy with a normalising rationale in schizophrenia is currently being scientifically evaluated. Studies have been designed and funded to measure compliance with neuroleptics and outcome, and to measure efficacy on an 'intention to treat' basis. The techniques described would appear to be safe and acceptable to schizophrenic patients. This form of treatment cannot, however, be maximally beneficial without the full support of a comprehensive community psychiatric service including sensitive neuroleptic prescribing and monitoring, drop-in and day care facilities and a full range of accommodation options.




Beck, A. T. (1952) 'Successful out-patient psychotherapy of a chronic schizophrenic with a delusion based on borrowed guilt', Psychiatry 15: 305-12.

Beck, A. T., Rush, A. J., Shaw, B. F. and Emery, G. (1979) Cognitive Therapy of Depression, Chichester: John Wiley.

Birchwood, M., Smith, I., Macmillan, F., Hogg, B., Prasad, R., Harvey C. and Bering S. (1989) 'Predicting relapse in schizophrenia: the development and implementation of an early signs monitoring system using patients and families as observers', Psychological Medicine 19: 649-56.

Blackburn, I. M. and Davidson, K. (1990) Cognitive Therapy for Depression and Anxiety, Oxford: Blackwell.

Ensink, B. I. (1992) Confusing Realities: a Study on Child Sexual Abuse and Psychiatric Symptoms, Amsterdam: Free University Press.

Falloon, I. R. H. and Talbot, R. E. (1981) 'Persistent auditory hallucinations: coping mechanisms and implications for management', Psychological Medicine 11: 329-39.

Grassian, G. (1983) 'Psychopathology of solitary confinement', American Journal of Psychiatry 140: 1450-4.

Green, M. F. (1993) 'Cognitive remediation in schizophrenia: is it time yet?', American Journal of Psychiatry 150: 178-87.

Kingdon, D.G. and Turkington, D. (1991) 'The use of cognitive behaviour therapy with a normalising rationale in schizophrenia', Journal of Nervous and Mental Disease 179: 207-211.

Kingdon, D.G. and Turkington, D. (1994) Cognitive Behaviour Therapy of Schizophrenia, New York: Guildford Press.

Kingdon, D. G., Turkington, D. and John, C. (1994) 'Cognitive behaviour therapy of schizophrenia. The amenability of delusions and hallucinations to reasoning', British Journal of Psychiatry, (editorial) 164: 581-7.

Kingdon, D. G., Turkington, D. and Judd, M. (1989) 'Befriending: cost effective community care', Psychiatric Bulletin 13: 350-1.

Kingdon, D. G., Turkington D., Malcolm, K., Szulecka, K. and Larkin, E. (1991) 'Replacing the mental hospital: community provision for a district's chronically psychiatrically disabled in domestic environments', British Journal of Psychiatry 158: 113-16.

Kingdon, D. G., Turkington, D., John, C., Rowlands, P. and Hope-Gill, M. (in press) , A controlled study of cognitive-behavioural therapy vs befriending in the treatment of schizophrenia', British Journal of Psychiatry.

Oswald, I. (1974) Sleep (3rd edition), Harmondsworth: Penguin.

Romme, M. A. I. and Escher, A.D.M.A.C. (1989) 'Hearing voices', Schizophrenia Bulletin 15: 209-16.

Scharfetter, C. (1980) General Psychopathology (translated H. Marshall), Cambridge: Cambridge University Press.

Siegel, R. K. (1984) 'Hostage hallucinations', Journal of Nervous and Mental Disease 172: 264-71.

Slade, P. D. (1973) 'The psychological investigation and treatment of auditory hallucinations: a second case report', British Journal of Medical Psychology 46: 293-6.

Slade, P. D. (1984) 'Sensory deprivation and clinical psychiatry', British Journal of Hospital Medicine 32: 256-60.

Strliuss, I. S. (1969) 'Hallucinations and delusions as points on continua of function " Archives of General Psychiatry 21: 581-6.

Strauss, I. S., Walker, W. and Brenner, H. D. (eds) (1987) Psychosocial Management of Schizophrenia, Toronto: Hans Hoober.

Tarrier, N., Beckett, R., Harwood, S., Baker, A., Yusupoff, L. and Ugarteburu, I. (1993) , A trial of two cognitive-behavioural methods of treating drug resistant residual psychotic symptoms in schizophrenic patients: 1. Outcome', British Journal of Psychiatry 162: 524-32.

Turkington, D. and Kingdon, D.G. (1991) 'Ordering thoughts in thought disorder', British Journal of Psychiatry 159: 160-1.

Turkington, D., Kingdon, D. G. and Malcolm, K. (1991) 'The use of an unstaffed flat for crisis intervention and rehabilitation', Psychiatric Bulletin 15: 13-14.

Watts, F. N., Powell, G. E. and Austin, S. V. (1973) 'Modification of delusional

beliefs', British Journal of Medical Psychology 46: 359-63.

Wilcox, J., Rionase, D. and Suez, L. (1991) ĎAuditory hallucinations, post-traumatic stress disorder and ethnicity', Comprehensive Psychiatry 32: 320-3.

Zubin, J. (1987) 'Possible implications of the vulnerability hypothesis for the psychosocial management of schizophrenia', in J. S. Strauss, W. Walker and H. D. Brenner (eds) Psychosocial Management of Schizophrenia, Toronto: Hans Hoober.