The Future Of Cognitive Behavioral Therapy For Psychosis

Newman‐Taylor, K., & Bentall, R. (2023). Cognitive behavioural therapy for psychosis: The end of the line or time for a new approach?. Psychology and Psychotherapy: Theory, Research and Practice. https://doi.org/10.1111/papt.12498

Key Points

• This paper reviews research on cognitive behavioral therapy for psychosis (CBTp) and questions whether it should continue to be a frontline treatment given the modest effects found in studies.
• Meta-analyses show CBTp leads to small-medium reductions in symptoms compared to treatment-as-usual or active controls like counseling.
• Outcomes are heterogeneous – some people benefit, others show no change or get worse.
• Common factors like therapeutic alliance impact outcomes but are neglected.
• Personalized, formulation-based CBTp targeting specific mechanisms (e.g., trauma, worry) may improve outcomes.
• More research on moderators and mechanisms is needed to advance the field.

Rationale

Cognitive behavioral therapy for psychosis (CBTp) was introduced in the 1990s to target symptoms associated with psychosis.

Following encouraging early results, it became a recommended treatment in many Western countries.

However, subsequent meta-analyses of randomized trials indicate only small to medium effects for CBTp compared to treatment-as-usual and active comparison conditions like supportive counseling (Turner et al., 2014; McGlanaghy et al., 2021).

These findings raise questions about whether CBTp should continue to be promoted and implemented as a frontline intervention given the resource-intensive training and supervision it requires, especially in the context of ongoing resource limitations in mental health systems internationally (Burgess-Barr et al., 2023).

What’s the next step? This paper examines two recent umbrella reviews systematically summarizing CBTp research to consider whether the modest effects found indicate it is time to move on from CBTp or if refinements in how it is researched and delivered may improve outcomes and sustain the approach as an evidence-based treatment option.

Method

This paper outlines and discusses two recently published umbrella reviews on CBTp for psychosis and related conditions:

Solmi et al. (2023) reviewed 83 meta-analyses comprising data from 1,246 randomized controlled trials, including 84,925 participants.

The meta-analyses examined various psychosocial interventions, including CBTp, for adults with early psychosis and schizophrenia spectrum diagnoses, comparing them to treatment-as-usual and active control conditions.

Outcomes examined included symptom severity, functioning, recovery, etc. Thirteen of the meta-analyses focused specifically on CBTp.

Fusar-Poli et al. (2019) reviewed meta-analyses examining interventions aimed at reducing the risk of transition to psychosis in adolescents and young adults at clinical high risk. Most interventions were psychosocial, with several trials of CBTp.

By providing an overview of the available meta-analytic evidence, these umbrella reviews allow an assessment of the state of the evidence base regarding CBTp for psychosis and where there is room for improvement.

Sample

• The umbrella reviews synthesized data from RCT samples of people with early psychosis, schizophrenia-spectrum disorders, and clinical high risk of psychosis.
• The total number of participants across reviews was >100,000.

Statistical Analysis

• The umbrella reviews statistically analyzed data from meta-analyses reporting effect sizes, standard mean differences, risk ratios, and odds ratios.
• No additional statistical tests were performed in this conceptual paper.

Results

Solmi and colleagues found a small to medium effect size for CBTp on outcomes like symptoms and functioning compared to treatment-as-usual based on 13 meta-analyses.

The review by Fusar-Poli et al. found low to moderate-quality evidence that CBTp may reduce transition risk better than monitoring alone.

While these results generally support CBTp as having benefits over monitoring/treatment-as-usual, the effects tend to be modest compared to these controls and active comparison therapies like counseling.

Additionally, the average effects reported in meta-analyses may well mask a large degree of heterogeneity in outcomes.

Some people show strong benefits from CBTp on symptoms and functioning, whereas others show little change or even deterioration – so-called “therapeutic harm.” Unpacking for whom and under what conditions CBTp helps versus harms is critical.

Insight

These mixed findings raise questions about how CBTp is implemented and researched.

The modest effects found may reflect the fact that CBTp has been studied largely as a one-size-fits-all intervention, with little consideration for the mechanisms maintaining a person’s difficulties nor the common factors known to impact outcomes across therapies, like alliance.

Formulation-based CBT aims to map out an individualized understanding of a person’s problems to guide treatment, but in practice, research trials have mainly tested manualized protocols.

Additionally, while psychosis has complex roots, including genetic, biological, and psychosocial factors, CBTp primarily targets symptom maintenance processes like reasoning biases, behaviors, and coping strategies.

An alternative perspective suggests we need to advance CBTp research and practice towards more personalized, mechanism-focused precision therapies.

There is untapped potential to improve outcomes by assessing baseline characteristics, forming strong therapeutic alliances, and selecting and monitoring specific treatment techniques based on an individual’s formulation.

For example, targeting underlying causal processes like trauma or proximal symptom drivers like worry holds promise. Adapting CBTp based on cultural factors also remains an underexplored avenue (Rathod et al., 2016).

While CBTp has strengths as an accessible, symptoms-focused therapy, modifications informed by common factors, mechanisms, and individual differences may be needed to fulfill its promise as a leading psychosocial intervention for psychosis.

If findings continue indicating modest effects on average, changes to clinical recommendations regarding CBTp as a frontline treatment may also be warranted.

Strengths

• Comprehensively summarizes CBTp research, identifying consistent modest effects.
• Outlines factors limiting effectiveness based on psychotherapy science more broadly.
• Makes specific, testable recommendations to improve outcomes.

Limitations

• The umbrella reviews have methodological restrictions typical of meta-analyses.
• The discussion overlooks strengths of the evidence base, like symptom specificity.
• Conceptual ideas proposed require empirical testing before clinical implementation.

Implications

• Supports continued provision of CBTp but indicates room for improvement.
• Suggests research and clinical efforts should focus on precision approaches.
• If findings replicate, changes to clinical guidelines regarding CBTp may be warranted.

Conclusion

CBTp for psychosis currently generates small to medium improvements in symptoms on average but with a range of outcomes from deterioration to remission.

Abandoning CBTp risks losing gains made in establishing psychological approaches in psychosis treatment.

However, continuing to deliver manualized CBTp as usual also seems unlikely to advance outcomes.

An exciting middle ground supported by the current evidence base involves refining this therapy to create more personalized, mechanism-focused precision protocols.

Matching specific cognitive and behavioral techniques to baseline assessments and formulations holds promise for improving real-world outcomes.

To achieve reliable benefits from CBTp and sustain it as an evidence-based treatment, renewed research strategies targeting moderators and mechanisms are imperative.

Reference

Primary reference

Newman‐Taylor, K., & Bentall, R. (2023). Cognitive behavioural therapy for psychosis: The end of the line or time for a new approach?. Psychology and Psychotherapy: Theory, Research and Practice. https://doi.org/10.1111/papt.12498

Other references

Bentall, R. P., Jackson, H. F., & Pilgrim, D. (1988). Abandoning the concept of ‘schizophrenia’: Some implications of validity arguments for psychological research into psychotic phenomena. British Journal of Clinical Psychology, 27(4), 303-324. https://doi.org/10.1111/j.2044-8260.1988.tb00795.x

Bighelli, I., Rodolico, A., García-Mieres, H., Pitschel-Walz, G., Hansen, W.-P., Schneider-Thoma, J., Siafis, S., Wu, H., Wang, D., Salanti, G., Furukawa, T. A., Barbui, C., & Leucht, S. (2021). Psychosocial and psychological interventions for relapse prevention in schizophrenia: A systematic review and network meta-analysis. Lancet Psychiatry, 8, 969–980. https://doi.org/10.1002/wps.20577

Fusar-Poli, P., Davies, C., Solmi, M., Brondino, N., De Micheli, A., Kotlicka-Antczak, M., Shin, J. I., & Radua, J. (2019). Preventive treatments for psychosis: Umbrella review (just the evidence). Frontiers in Psychiatry, 10, 764. https://doi.org/10.1002/wps.20446

Galletly, C., Castle, D., Dark, F., Humberstone, V., Jablensky, A., Killackey, E., Kulkarni, J., McGorry, P., Nielssen, O., & Tran, N. (2016). Royal Australian and New Zealand College of Psychiatrists clinical practice guidelines for the management of schizophrenia and related disorders. Australian and New Zealand Journal of Psychiatry, 50, 410–472. https://doi.org/10.1177/0004867416641195

Jauhar, S., Laws, K. R., & McKenna, P. J. (2019). CBT for schizophrenia: A critical viewpoint. Psychological Medicine, 49(8), 1233–1236. https://doi.org/10.1017/S0033291718004166

Leichsenring, F., Abbass, A., Heim, N., Keefe, J. R., Kisely, S., Luyten, P., Rabung, S., & Steinert, C. (2023). The status of psychodynamic psychotherapy as an empirically supported treatment for common mental disorders—An umbrella review based on updated criteria. World Psychiatry, 22(2), 286–304. https://doi.org/10.1002/wps.21104

McGlanaghy, E., Turner, D., Davis, G. A., Sharpe, H., Dougall, N., Morris, P., Prentice, W., & Hutton, P. (2021). A network meta-analysis of psychological interventions for schizophrenia and psychosis: Impact on symptoms. Schizophrenia Research, 228, 447–459. https://doi.org/10.1016/j.schres.2020.12.036

National Institute for Health and Care Excellence. (2014). Psychosis and schizophrenia: Treatment and Management. (Clinical guideline 178). http://guidance.nice.org.uk/CG178

Norman, R., Lecomte, T., Addington, D., & Anderson, E. (2017). Canadian treatment guidelines on psychosocial treatment of schizophrenia in adults. The Canadian Journal of Psychiatry, 62(9), 617–623. https://doi.org/10.1177/0706743717719894

Rathod, S., Garner, C., Griffiths, A., Dimitrov, B. D., Newman-Taylor, K., Woodfine, C., Hansen, L., Tabraham, P., Ward, K., Asher, C., Phiri, P., Naeem, F., North, P., Munshi, T., & Kingdon, D. (2016). Protocol for a multicentre study to assess feasibility, acceptability, effectiveness and direct costs of TRIumPH (treatment and recovery In PsycHosis): Integrated care pathway for psychosis. British Medical Journal Open, 6(12), e012751.

Solmi, M., Croatto, G., Piva, G., Rosson, S., Fusar-Poli, P., Rubio, J. M., Carvalho, A. F., Vieta, E., Arango, C., DeTore, N. R., Eberlin, E. S., Mueser, K. T., & Correll, C. U. (2023). Efficacy and acceptability of psychosocial interventions in schizophrenia: Systematic overview and quality appraisal of the meta-analytic evidence. Molecular Psychiatry, 28, 354–368.

Tong, Z., Li, F., Ogawa, Y., Watanabe, N., & Furukawa, T. A. (2018). Quality of randomized controlled trials of new generation antidepressants and antipsychotics identified in the China National Knowledge Infrastructure (CNKI): A literature and telephone interview study. BMC Medical Research Methodology, 18, 96.

Turner, D. T., van der Gaag, M., Karyotaki, E., & Cuipers, P. (2014). Psychological interventions for psychosis: A meta-analysis of comparative outcome studies. American Journal of Psychiatry, 171, 523–538. https://doi.org/10.1176/appi.ajp.2013.13081159

Varese, F., Smeets, F., Drukker, M., Lieverse, R., Lataster, T., Viechtbauer, W., Read, J., van Os, J., & Bentall, R. P. (2012). Childhood adversities increase the risk of psychosis: A meta-analysis of patient-control, prospective-and cross-sectional cohort studies. Schizophrenia Bulletin, 38(4), 661–671. https://doi.org/10.1093/schbul/sbs050

Wu, T., Li, Y., Bian, Z., Liu, G., & Moher, D. (2009). Randomized trials published in some Chinese journals: How many are randomized? Trials, 10, 46. https://doi.org/10.1186/1745-6215-10-46

Learning check

1. What are the potential benefits of taking a more personalized, mechanistic approach to CBTp? What are the challenges?
2. How might we better measure whether CBTp is truly helping or harming someone?
3. If the common factors like alliance drive outcomes, should we focus more research efforts there than specific techniques? Why or why not?
4. Could other therapeutic approaches like psychodynamic or humanistic therapies potentially help with mechanisms like trauma and attachment in psychosis? What might they offer?
5. What innovations in research methods might allow for more rapid testing of precision psychotherapy protocols? What are the obstacles to testing more personalized approaches?