Electronic Letters to:
|
|
eLetters: Electronic letters published:
-
![[Read eLetter]](/icons/misc/sectionDOWN.gif)
Risk of harm with CBT in Early Psychosis
- Karl H Marlowe (26 October 2004)
-
Re: Risk of harm with CBT in Early Psychosis
- Anthony P Morrison, Paul French, Lara Walford, Shôn W. Lewis, Aoiffe Kilcommons, Joanne Green, Sophie Parker, and Richard P. Bentall (23 November 2004)
-
Interpreting numbers needed to treat – cognitive therapy to prevent transition into psychosis.
- Manjiri V Lele, Adhiraj Joglekar, Specialist Registrar, Child & Adolescent Psychiatry, Windmill Lodge, West London Mental Health NHS Trust, Southall, UB1 3EU. (3 February 2005)
|
|
|||
|
Karl H Marlowe, Consultant Psychiatrist Early Psychosis Intervention Team (EPIT), Counties Manukau, Auckland, New Zealand.
Send letter to journal:
Karl.Marlowe{at}middlemore.co.nz Karl H Marlowe
|
Sir, The last Cochrane systematic review of early intervention for those with psychosis included CBT, Family Therapy and Medication and reported no significant decrease in the development of psychosis at 12 months follow up (Marshal & Lockwood, 2004). The implications of the recent CBT prevention study (Morrison et al, 2004) needs to be realistically interpreted, with this background. Firstly, there was the exclusion of 2 people from the Cognitive Therapy arm, after the trial had begun, which would have led to a non- significant result. This should have been acknowledged in the abstract, as an abstract has the most impact with service planners. The second issue is that after 6 months of Cognitive Therapy, there was a decrease in the development of psychosis compared to the control arm, however, there was similar distress for both groups. A cognitive therapy for psychosis has an aim of decreasing the distress of psychosis as well as the formulation of an explanatory model for that psychosis. It may be that a reframed and normalised explanatory language was taught to the high risk individuals, and this led to the decreased identification of symptoms at 12 months and the masking of a psychotic episode. This would not ultimately lead to a decrease in distressing psychosis, but to a later identification of psychosis and a possible delay in pharmacological treatment. The possible risk of harm or hazard was ignored, with a clear bias against the use of medication expressed by the authors in the discussion. Furthermore, the editorial comment alluded to the suggestion of a premature publication (Tyre,P, 2004), but it is the implication of harm which needs to be explicitly stated. References: Marshal, M & Lockwood, A. (2004) Early Intervention for Psychosis (Cochrane Review). Cochrane Library, Issue 3. Oxford: update Software. Morrison, A. P., French, P., Walford, L., et al. (2004) Cognitive therapy for the prevention of psychosis in people at ultra-high risk: Randomised controlled trial. British Journal of Psychiatry, 185, 291-297. Tyrer, P. (2004) From the Editor's desk. British Journal of Psychiatry, 185, 360. |
|||
|
|
|||
|
Anthony P Morrison, Reader in Clinical Psychology University of Manchester, Paul French, Lara Walford, Shôn W. Lewis, Aoiffe Kilcommons, Joanne Green, Sophie Parker, and Richard P. Bentall
Send letter to journal:
tony.morrison{at}psy.man.ac.uk Anthony P Morrison, et al.
|
Sir We welcome Dr Marlowe’s response to our paper, and would like to respond to the issues that he identified. The Cochrane review to which he refers examined more traditional approaches to early intervention (i.e. from first episode onwards) rather than a preventative approach in people at high risk, so we are unsure of the relevance of this. Within the manuscript we clearly acknowledge that there were several methodological limitations, including the exclusion of 2 participants, but we were unable to incorporate these in the abstract as he suggests due to abstract length limitations imposed by the journal (indeed, we were asked to further reduce the abstract at the proofs stage). We agree that cognitive therapy for psychosis (and the prevention of psychosis) has an aim of decreasing the distress of psychotic experiences as well as the formulation of an explanatory model for a person’s difficulties. We also agree that a reframed and normalised explanatory language may be developed by the service users; however, it is unlikely that this would lead to a masking of a psychotic episode. Rather, it is intended to reduce the potential for catastrophic appraisals of psychotic experiences, which are very clearly implicated in the experience of distress (Chadwick & Birchwood, 1994) and the development of normalising appraisals is at the heart of cognitive therapy for established psychosis (Morrison, Renton, Dunn, Williams, & Bentall, 2003) and the prevention of psychosis alike (French & Morrison, 2004). Even if such a masking were to occur, the assumption that this could cause harm clearly demonstrates a bias against the use of psychosocial interventions, as it suggests that only pharmacological treatments can reduce the potential harm that may result from an untreated psychotic episode, when there is evidence that psychological treatment is also important in this respect (de Haan, Linszen, Lenior, de Win & Gorsira, 2003). We are accused of being biased against using antipsychotic medication; we certainly are in a population who are yet to develop a psychotic disorder, for the ethical reasons outlined within our paper and elsewhere (Bentall & Morrison, 2002). Finally, it is suggested that we avoid explicitly stating the possibility of harm arising from such an intervention; however, we clearly highlight the possibility of harm resulting from stigmatisation. References Bentall, R. P., & Morrison, A. P. (2002). More harm than good: The case against using antipsychotic drugs to prevent severe mental illness. Journal of Mental Health, 11, 351-365. Chadwick, P., & Birchwood, M. (1994). The omnipotence of voices: A cognitive approach to auditory hallucinations. British Journal of Psychiatry, 164, 190-201. de Haan L, Linszen DH, Lenior ME, de Win ED, Gorsira R. (2003) Duration of untreated psychosis and outcome of schizophrenia: delay in intensive psychosocial treatment versus delay in treatment with antipsychotic medication. Schizophrenia Bulletin, 29(2):341-8. French, P., & Morrison, A. P. (2004). Cognitive therapy for people at high-risk of psychosis. London: Wiley. Morrison, A. P., Renton, J. C., Dunn, H., Williams, S., & Bentall, R. P. (2003). Cognitive Therapy for Psychosis: a Formulation- based Approach. London: Psychology Press. |
|||
|
|
|||
|
Manjiri V Lele, Senior House Officer Regional Secure Unit, West London Mental Health NHS Trust, Southall, UB1 3EU, Adhiraj Joglekar, Specialist Registrar, Child & Adolescent Psychiatry, Windmill Lodge, West London Mental Health NHS Trust, Southall, UB1 3EU.
Send letter to journal:
lelemanjiri{at}hotmail.com Manjiri V Lele, et al.
|
The numbers needed to treat (NNT) is the average number of patients a clinician needs to treat with a particular therapy to prevent one bad outcome. It is a translation into clinical terms of the absolute risk reduction derived from a trial. NNT is a useful way of reporting the results, but it is important that the uncertainty in the estimated number is accounted for by using confidence intervals (Altman, 1998). Morrison et al (2004) use point estimates of the NNT to demonstrate the benefit of cognitive therapy for the prevention of psychosis in people at ultra-high risk. Their NNT to prevent transition into psychosis are impressive (the NNT for preventing someone from meeting DSM–IV criteria for a psychotic disorder is 5, the NNT for preventing prescription of antipsychotic medication is 5 and the NNT for PANSS-defined transition is 6). Unfortunately, when we calculated the confidence intervals (CI), the results failed to evoke the same confidence. The CI for the first two of the above NNT is 2.5 to 117.8 and 2.2 to 22.7 respectively. The CI for the PANSS-defined transition is -39.9 to 2.8. Interpretation of the latter is more complex. Such difficulties arise when the CI for the absolute risk reduction includes zero. Based on suggestions made by Altman (1998) we can interpret the above CI in two parts: a) we would require 3 to infinite number of patients to receive cognitive therapy to prevent one patients’ transition to psychosis b) the negative number needed to treat indicates that the treatment has a harmful effect. In this case treating anywhere between 40 to infinite number of patients would result in one additional patient being harmed by the therapy. Our argument should not discount the results of Morrison et al (2004). We hope that adequately powered studies that address methodological limitations of their work are awaited before prematurely justifying use of prodromal interventions as done by Singh & Fisher (2005). The number needed to treat has as much potential to distort and mislead as to enlighten and caution is imperative while interpreting them. References: Altman D.G. (1998) Confidence intervals for the number needed to treat British Medical Journal; 317:1309-1312 Morrison, A.P., French. P., Lewis. S.W., et al (2004) Cognitive therapy for the prevention of psychosis in people at ultra-high risk. The British Journal of Psychiatry 185: 291-297 Singh, S.P & Fisher, H.L. (2005) Early intervention in psychosis: obstacles and opportunities. Advances in Psychiatric Treatment (2005) 11: 71-78 Declaration of interest None |
|||