Purpose: Following its introduction in the early 1990s, cognitive behavioural therapy for psychosis (CBTp) has been evaluated in a large number of clinical trials and is now established as a recommended treatment in the UK National Health Service and elsewhere in the world. Meta-analyses, however, indicate modest effects compared to treatment as usual or comparison therapies such as supportive counselling. Here, we seek to identify factors impacting the effectiveness of CBTp, and avenues for future psychotherapy research that may improve outcomes.
Method: We outline two recent umbrella reviews and discuss factors likely to impact the effectiveness of CBTp.
Results: Modest effect sizes from meta-analyses mask heterogeneous outcomes, with some people benefiting and others possibly being harmed by therapy. Common factors such as the therapeutic alliance play an important role in determining outcomes but have been largely neglected by CBTp researchers. There is also the promise of improving outcomes by identifying and targeting the psychological mechanisms that either maintain psychotic symptoms (e.g. worry) or are causally implicated (e.g. trauma).
Conclusions: It is unlikely that everyone with psychosis will be equally responsive to the same therapeutic protocols. We need a new, personalised psychotherapy approach to CBTp research and practice, and can learn from research for anxiety and depression examining predictors of therapeutic response to inform treatment decisions. Precision psychological therapies informed by a combination of individual characteristics, common factors and a focus on specific mechanisms will require new research strategies and are likely to lead to improved outcomes for people with psychosis.
Keywords: AI; CBT; CBTp; common factors; machine learning; meta-analysis; outcomes; psychosis; therapeutic harm; umbrella review.
© 2023 The Authors. Psychology and Psychotherapy: Theory, Research and Practice published by John Wiley & Sons Ltd on behalf of The British Psychological Society.