Impact of case formulation in cognitive behaviour therapy for psychosis

https://doi.org/10.1016/S0005-7967(02)00033-5Get rights and content

Abstract

Case formulation (CF) is the ‘first principle’ in cognitive behavior therapy (CBT), and the cornerstone of CT for psychosis. Experiment 1 (E1) assesses impact of CF in CBT for psychosis on perception of therapeutic relationship. Experiment 2 assesses the impact of CF on strength of delusional and self-evaluative beliefs (E2). Both E1 and E2 assess impact of CF on symptoms of anxiety and depression. CF had little or no impact on these variables. Descriptive data suggested there were other benefits for certain participants. Therapists’ perception of the therapeutic relationship did improve following CF. Six clients found CF, at least in part, a negative experience.

Section snippets

Experiment 1: Introduction

Case formulation (CF) is ‘a hypothesis about the causes, precipitants and maintaining influences of patients’ psychological, interpersonal and behavioral problems’ (Eels, 1997, p.1). It has been described as the ‘first principle of cognitive therapy’ (Beck, 1995: and see Persons, 1989) and as especially important with complex cases. CF is seen as crucial to successful outcome in CBT for psychosis. In Experiment 1 (E1), we assess two specific empirical claims made for the impact of CF in CBT for

Participants and therapists

All participants were referred for CBT for psychosis, because of drug-resistant distressing positive symptoms of at least 6 months duration. Sixteen people were provided with written information about the studies. Fifteen gave informed consent, and met DSM IV criteria (American Psychiatric Association, 1994) for diagnoses of paranoid schizophrenia (12), schizo-affective disorder (2) and delusional disorder (1). Two dropped out of therapy prior to formulation. 13 participants (7 men, 6 women,

Results

Thirteen clients completed the four time points. Data from the HAq-P and HAq-T each showed good internal reliability (Cronbach’s Alpha=0.93); descriptive statistics for these measures, and the HADS, appear in Table 1. A Friedman 2-way ANOVA for related samples revealed a significant interaction between time and total scores on the HAq-P (Chi squared=10.7, df=3, p<0.05) and HAq-T (Chi squared=10.9, df=3, p<0.05). Following criticism of the HAq, analyses were repeated having removed those four

Discussion

The present experiment collected empirical data to operationalise and test two hypotheses about the impact of CF in CBT for psychosis—that it improves alliance, and that it reduces distress. The data from 13 participants did not support either hypothesis. Ratings of the alliance were positive—as in other research using CBT for psychosis (see Svensson & Hansson, 1999). Client feedback suggested that CF did increase understanding and optimism (n=9). Four people reported mixed (i.e. negative and

Experiment 2: Introduction

Experiment 1 found that CF did not impact on therapeutic alliance, as measured by the HAq, or distress, as measured by the HADS. In Experiment 2 (E2) we broaden the focus to assess the impact of CF on the three main outcomes in CBT for psychosis—distress (again using HADS), distressing secondary delusions, and negative beliefs about the self (Chadwick, Birchwood and Trower, 1996, Fowler, Garety and Kuipers, 1995).

Participants

Four people with distressing auditory hallucinations and related secondary paranoid delusions participated in a multiple-baseline design. Beth (names and details have been altered), a 20-year-old, single woman resided on a secure ward following three suicide attempts and regular self-harm, both driven by auditory hallucinations and related paranoid beliefs. Her Beliefs About Voices Questionnaire scores (BAVQ-R: Chadwick, Lees, & Birchwood, 2000) scores were omnipotence (18), malevolence (18),

Belief conviction in delusions and negative self-evaluations

Figure 1 shows belief conviction ratings at each session during baseline. For seven of the eight beliefs, scores during CF were no lower than during baseline. For Beth, the first two delusional conviction ratings during CF were lower than any during baseline: conviction rose over the final two scores in this phase to 100%. Conviction in the NSE fell when these beliefs were challenged (phase 3); in three cases conviction fell by 40% or more, for one by 10%. For two participants, delusional

Discussion

Experiment 2 found that CF when delivered over four separate sessions did not have a significant impact on strength of delusions, or negative self-evaluations. This is perhaps surprising given that CF portrayed both sets of beliefs as reactions to and ways of making sense of key life events and experience (Maher, 1988, Chadwick and Lowe, 1990). For all four people, both sets of beliefs weakened during subsequent direct challenge. These findings suggest that CF alone does not reduce belief

General discussion

Taken separately and together the two experiments found no evidence that CF has a direct impact on any of the main targets of CBT for psychosis (Fowler, Garety and Kuipers, 1995, Chadwick, Birchwood and Trower, 1996)—namely, therapeutic relationship (n=13), delusions or self-evaluations (n=4) and distress (n=17).

It remains true that CF might have a range of other direct benefits not researched in these experiments. Also, CF might have impacted on those dependent variables used in our research,

References (21)

  • G.T. Wilson

    Manual based treatments: the clinical application of research findings

    Behaviour Research and Therapy

    (1996)
  • Diagnostic and Statistical Manual of Mental Disorders (DSM IV)

    (1994)
  • L.B. Alexander et al.

    The Penn Helping Alliance Scales

  • M. Bassler et al.

    The Helping Alliance Questionnaire (HAQ) by Luborsky—possibilities for evaluation of the inpatient psychotherapy treatment process

    Psychotherpeut

    (1995)
  • A.T. Beck

    Beyond belief: a theory of modes, personality and psychopathology

  • P. Chadwick et al.

    Cognitive therapy for delusions, voices and paranoia

    (1996)
  • P.D. Chadwick et al.

    Beliefs about voices questionnaire—revised (BAVQ-R)

    British Journal of Psychiatry

    (2000)
  • P. Chadwick et al.

    Measurement and modification of delusional beliefs

    Journal of Consulting and Clinical Psychology

    (1990)
  • T. Eells

    Handbook of psychotherapy case formulation

    (1997)
  • D. Fowler et al.

    Cognitive behavioural therapy for people with psychosis: a clinical handbook

    (1995)
There are more references available in the full text version of this article.

Cited by (106)

View all citing articles on Scopus
View full text