Psychological Medicine

Original Article

Prognostic indices with brief and standard CBT for panic disorder: I. Predictors of outcome

MICHAEL G. T. DOWa1 c1, JUSTIN A. KENARDYa2, DEREK W. JOHNSTONa3, MICHELLE G. NEWMANa4, C. BARR TAYLORa5 and AILEEN THOMSONa6

a1 NHS Fife, Department of Clinical Psychology, Stratheden Hospital, Cupar, Fife, UK

a2 Centre for National Research on Disability and Rehabilitation Medicine, and School of Psychology, University of Queensland, Brisbane, Queensland, Australia

a3 Department of Psychology, Kings College, University of Aberdeen, UK

a4 Department of Psychology, Pennsylvania State University, PA, USA

a5 Department of Psychiatry and Behavioral Science, Stanford University School of Medicine, Stanford, CA, USA

a6 Health Psychology Department, Gloucestershire Royal Hospital, Gloucester, UK

ABSTRACT

Background Although the effectiveness of cognitive behavioural therapy (CBT) in the management of panic disorder (PD) is now well established, there have been few studies of predictors of outcome with this patient group using clinical effectiveness trial data, a hypothesis-testing model, and a dependent measure of clinically significant change.

Method The data for this study came from a randomized controlled trial of three forms of CBT delivery for PD with and without agoraphobia (two 6-week CBT programmes, one of which was computer assisted, and one therapist-directed 12-week CBT programme), comprising a total of 186 patients across two sites. Based on previous related research, five hypothesized predictors of post-treatment and follow-up outcome were identified and examined, using a series of bivariate and multivariate analyses.

Results The results in general supported the hypotheses. Strength of blood/injury fears, age of initial onset of panic symptoms, co-morbid social anxieties and degree of agoraphobic avoidance were predictive of both measures of post-treatment outcome. Degree of residual social difficulties and the continued use of anxiolytics at post-treatment were also shown to predict poor outcome at the 6-month follow-up. However, strength of continuing dysfunctional agoraphobic cognitions by the end of active treatment did not predict outcome at follow-up for the sample as a whole.

Conclusions The identification of consistent predictors of outcome with CBT has many clinical and research benefits. As CBT, however, is being delivered increasingly in a variety of brief formats, further research is required to identify moderators of response to these ‘non-standard’ treatment formats.

(Online publication May 10 2007)

Correspondence

c1 Address for correspondence: Dr Michael G. T. Dow, Department of Psychology, University of Stirling, Stirling FK9 4LA, UK. (Email: mgtd1@stir.ac.uk)

Metrics