Article Text

Download PDFPDF

Cognitive behavioural therapy improves psychotic symptoms at 18 months in people with schizophrenia
  1. Patrick K Kingsep, BA (Hons), MA (Clin Psych)
  1. Centre for Clinical Interventions, Perth, Western Australia, Australia

Statistics from

OpenUrlAbstract/FREE Full Text

Q Does cognitive behavioural therapy improve symptoms in people with schizophrenia?


Embedded ImageDesign:

Randomised controlled trial.

Embedded ImageAllocation:


Embedded ImageBlinding:

Assessors blinded to treatment allocation.

Embedded ImageFollow up period:

18 months.

Embedded ImageSetting:

11 mental health units in Manchester, Salford, Liverpool and North Nottinghamshire, UK.

Embedded ImagePatients:

309 people, admitted for the first or second time, presenting with schizophrenia, schizophreniform disorder, schizoaffective disorder, delusional disorder, or psychosis (DSM-IV criteria). Participants had psychotic symptoms, not caused by substance misuse, for >4 weeks and scored 4 or more on the Positive and Negative Syndrome Scale (PANSS). People detained in hospital under the Mental Health Act 1983 were excluded.

Embedded ImageIntervention:

Participants were randomly assigned to cognitive behavioural therapy or supportive counselling plus usual care compared with usual care alone (included antipsychotic medication). Within five weeks of admission, participants received 15–20 hours of supportive counselling or cognitive behavioural therapy from experienced therapists. Both therapies aimed to treat delusional behaviour, hallucinations, abnormal beliefs, and related distress and to develop coping strategies. Further treatment was provided after two weeks and one, two, and three months.

Embedded ImageOutcomes:

Schizophrenia symptoms (PANSS, PSYRATS); relapse and hospital readmission times (hospital admission data and case notes). Participants were monitored for 70 days after admission, with follow up 18 months after randomisation.

Embedded ImagePatient follow up:

73% interviewed, 99% assessed for readmission into hospital and 95% for relapse.


At 18 months both cognitive behavioural therapy and supportive counselling plus usual care significantly improved schizophrenia symptoms compared with usual care (p>0.005). There were no significant differences in hospital readmission or relapse rates among groups (see for table). There was no significant difference between cognitive behavioural therapy and supportive counselling in symptom improvement (p>0.725). There were no significant differences in medication use, type, and dosage among groups.


Cognitive behavioural therapy and supportive counselling significantly improve schizophrenia symptoms compared with usual care. However, neither of the therapies affected relapse or rehospitalisation rates in the long term.


Despite cognitive-behavioural therapy (CBT) being increasingly used to treat psychotic symptoms in medication resistant clients and chronic schizophrenia, use in first episode or “critical stages” has not been extensively researched.1 The Tarrier et al paper provides a welcome and valuable contribution to the literature on early intervention.

Although there was lack of differentiation between CBT and supportive counselling at follow up, both provided significant symptom reduction over usual treatment. The lack of difference in relapse or hospitalisation rates across the two treatments and control group was unexpected. Previous research shows an association between duration of untreated psychosis and outcome,2 and as this study focuses on a “critical” period, the ultimate course of the disorder may have been altered. This would potentially be reflected outside of this study’s 18 month follow up. The “concealment” of group allocation from both assessors and therapists maintained appropriate control groups and improved on comparison. Previous studies have not always had this methodological rigour, with resultant opaqueness concerning intervention and control group contrast. Long term, 18 month follow up is an important indicator of the “maintenance of change” associated with treatment in this study. A longer term examination of symptom change, relapse, and rehospitalisation rates as a result of CBT (and the surprisingly robust supportive counselling condition) would be beneficial. That said, this study highlights the importance of deploying psychological therapy during the phase of the illness in which it is most likely to be of benefit in minimising the possibly devastating, even fatal, effects of the disorder.

As treatment priorities vary over the course of the illness, a flexible integration of biological, social, and psychological interventions is necessary.3 As the supportive counselling condition was as effective at follow up as CBT, and was seen as “unstructured”, this may highlight the need for not allocating 100% of contact time to “pure” therapy, thereby sustaining engagement and promoting associated clinical gain. Importantly, the multicentre approach in this study begins to reflect the “real world” application of treatment in varying contexts. An increased number of booster sessions spread over contact time, and progressively implemented in a community setting, would be advantageous. This may improve on long term adaptation of the intervention and generalisation, with potential reduction in relapse rates.

Although primary prevention of schizophrenia and related disorders is at present out of reach, this article reinforces the importance of early intervention. Early focus on impairment of functioning is an essential clinical task. To do otherwise weakens subsequent change and reduces potential functional gain for the individual.


Supplementary materials

  • Web-only Table

    The table is available as a downloadable PDF (printer friendly file).

    If you do not have Adobe Reader installed on your computer,
    you can download this free-of-charge, please Click here


    Files in this Data Supplement:

    • [View PDF] - Table Main outcomes at 18 months (differences between groups not significant)


  • For correspondence: Nicholas Tarrier, Education and Research Building, Wythenshawe Hospital, Manchester, UK;

  • Sources of funding: the UK Medical Research Council, Northwest England Regional NHSE Office, Trent Regional NHSE Office and Manchester, Salford and Trafford, Liverpool, Sefton, St Helens and Knowsley and North Nottinghamshire Health Authorities.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.