At least 1 in 8 people suffer from a mental disorder worldwide (WHO, 2022). Depression and anxiety disorders are the most common difficulties (Global Health Data, 2022), with the prevalence rates of other disorders – such as post-traumatic stress disorder (PTSD), obsessive-compulsive disorder (OCD) and borderline personality disorder (BPD) also challengingly high.
The UK’s National Institute of Health and Care Excellence (NICE) guidelines recommends psychotherapies as evidence-based, first-line interventions for many of these conditions (Leicsenrin et al, 2022). The most common psychotherapies offered include Cognitive Behavioural Therapy (CBT), Dialectical Behavioural Therapy (DBT), Eye Movement Desensitisation and Reprocessing (EMDR), Family/Systemic and Psychodynamic Psychotherapy.
Hundreds of randomised controlled trials have studied the effects of psychotherapies, and meta-analyses have shown them to be effective for major depressive disorder (MDD), (Cuijpers et al, 2021), PTSD, (Mavranezouli et al, 2020) OCD (Reid et al, 2021) Anxiety Disorders (Papola et al, 2023) and BPD (Cristea et al, 2017). However, their absolute and relative outcomes of response and remission rates have been less well studied.
Cuijpers et al. (2024) set out to resolve this with a meta-analysis evaluating the comparable efficacy psychotherapies for eight mental disorders: MDD, PTSD, OCD, GAD, BPD, panic disorder, social anxiety disorder and specific phobia.
Psychotherapy is a key treatment for mental health disorders, helping millions of people worldwide. A new meta-analysis explores its effectiveness across eight major conditions.
Methods
A series of living systematic reviews from the Metapsy initiative (www.metapsy.org) were used to conduct a literature search across PubMed, PsycINFO, Embase, and the Cochrane Register of Controlled Trials.
The psychotherapies included CBT, BAT – behavioural activation therapy, IPT – interpersonal psychotherapy, PST – problem-solving therapy, SUP – supportive therapy, BT – behaviour therapy, ERP – exposure and response prevention, DBT, TF – trauma focused, EMDR, and NTF – non-trauma focused.
The primary outcome was the absolute measure of response rate for patients in therapy and control conditions. Response rate was calculated by using baseline and post-test means and standard deviations.
The meta-analysis included 441 studies. Several studies compared two or more psychotherapies with one control group, bumping up the total number of comparisons to 569 (ranging from 22 studies for BPD to 196 for MDD).
The number of included patients reached an impressive 33,881, with 19,769 in therapy and 14,112 in control conditions (ranging from 1,011 in the OCD dataset to 14,908 in the MDD dataset).
Across the eight mental health disorders, about half (51.8%) of the psychotherapies were delivered in an individual format, ranging from 36.4% for BPD to 80.5% for PTSD. Group therapies ranged from 9.2% for PTSD to 33.8% for social anxiety disorder, to not being used at all for BPD, specific phobia, and OCD. The mean number of sessions was 11.8 (± 8.5), ranging from 3.3 (± 3.1) for specific phobia to 53.6 (± 42.6) for BPD.
To evaluate the quality of the studies, the authors followed the Cochrane Risk of Bias (RoB) assessment tool. The proportion of studies with low risk of bias (RoB) ranged from 6.2% for panic disorder to 39.5% for MDD, although the definition of low RoB differed across datasets. In terms of incomplete data – the obvious example being when participants drop out of a study – the authors considered such individuals as ‘non-responders’ to the intervention. Such ‘intention to treat’ analysis stops the counter-bias of ignoring them from the results, and gives us more confidence in their results, which will inherently be more conservative. Study heterogeneity was moderate-to-high across all disorders, ranging from 65% for OCD to 82% for MDD and BPD.
What constitutes a ‘positive outcome’ in such work? A common marker, adopted by the authors is taking a ‘response rate’ (RR) of at least a 50% improvement in symptoms from when treatment started, for that individual. Of course, this means that RR can ‘look different’ between participants, and trials may utilise varying outcome measures.
The meta-analysis calculated the ‘Number Needed to Treat’ (NNT): how many individuals would need to be treated by a given intervention, for one to respond.
Results
Results: response rates and NNTs
- As shown in the table, absolute Response Rates (again, a 50% symptom reduction) for psychotherapies were significant, though they varied between condition type, and are modest.
- Relative response rates were significant for all psychotherapies except those given in borderline personality disorder.
- In terms of NNTs, these ranged from 5.2 in GAD, through 4.8 in MDD and PTSD, to 2.4 for OCD.
| Mental health disorder | Absolute Response rate (95% CI) | Number needed to treat (95% CI) |
| Major depressive disorder | 0.42 (0.39 to 0.45) | 4.8 (4.1 to 5.8) |
| Post-traumatic stress disorder | 0.38 (0.33 to 0.43) | 4.8 (3.8 to 6.2) |
| Obsessive-compulsive disorder | 0.38 (0.30 to 0.47) | 2.4 (1.6 to 3.7) |
| Panic disorder | 0.38 (0.33 to 0.43) | 5.0 (3.7 to 7.3) |
| Generalised anxiety disorder | 0.36 (0.30 to 0.42) | 5.2 (3.7 to 7.8) |
| Social anxiety disorder | 0.32 (0.29 to 0.37) | 4.8 (3.8 to 6.1) |
| Specific phobia | 0.32 (0.23 to 0.42) | 4.6 (2.8 to 8.2) |
| Borderline personality disorder | 0.24 (0.15 to 0.36) | N/A |
This review found that “most patients receiving psychotherapy across all disorders do not show at least 50% symptom reduction. This means that clinicians often have to try several interventions or move to pharmacological or combined therapies to treat patients more effectively.”
Conclusions
This large meta-analysis provides a comprehensive, up-to-date, and nuanced understanding of treatment outcomes for a range of common psychotherapies, helping make informed decisions about mental health care. Overall, psychotherapies across eight mental disorders were effective, with at least 50% reduction in symptoms between initial treatment and measurement taken after the psychotherapeutic intervention. However, the absolute -response rates were modest, meaning the results were not as effective as we would have hoped in terms of understanding how many individuals might have benefited from a particular therapy. This is interesting, noting how – like all healthcare interventions – psychotherapies may not work in all instances or for all people.
A nuanced look at psychotherapy outcomes reveals both its effectiveness and the need for tailored mental health treatments.
Strengths and limitations
This is a large meta-analysis, impressive in its scope and scale, and provides updated figures on the effectiveness of a range of recommended psychological interventions. The researchers used ‘living systematic reviews’, an innovative approach as these is continuously updated as new evidence becomes available
Meta-analysis enabled the researchers to synthesise and combine data, increasing the sample size and enhancing the statistical power of the overall findings, making it easier to detect significant effects and draw more reliable conclusions. Furthermore, this enabled helpful comparisons to be made across different psychotherapies, and highlighted which approaches may be more beneficial for certain disorders, assisting healthcare providers in making more informed decisions about treatment options.
Although high levels of heterogeneity are common in systematic reviews and meta-analyses, there were a limited number of studies determined to have had low risk of bias. Generalisability of findings may be impacted as individual studies can differ in their populations and contexts.
Linked to this, much of the research included comes from North America and Europe, while we are mindful of variation in terms of sociodemographic, cultural, and clinical factors.
The work did not consider moderating factors such as therapist characteristics, and long-term effects such as whether any response was enduring. Fitting with this, we know that in those not showing significant symptom reduction, it is common in real-world practice that individuals might try several interventions or move to pharmacological or combined therapies to treat difficulties more effectively. Unfortunately, very little research on such sequential treatments has been conducted.
This large-scale meta-analysis updates psychotherapy effectiveness across disorders, but gaps remain in long-term impact, therapist influence, and global representation.
Implications for practice
This robust, large-scale research reaffirmed effectiveness of the psychotherapies evaluated, though more effective interventions are needed for individuals who do not respond to a first-line treatment.
The first author of this blog, a practicing clinical psychologist, often sees individuals with co-occurring disorders, such as an individual experiencing anxiety and depression. Different therapies tend to address each condition or problem type, and it is not uncommon for individuals to experience a range of psychotherapies before one starts to make a difference. Many mental health conditions are multifaceted, and one can benefit from a combination of emotional, cognitive, and behavioural based interventions. Different therapies can also target different aspects of an individuals’ difficulty, such as trauma, anxiety, depression, or relationship problems, and similarly, individuals’ needs can evolve as treatment progresses. Initial therapy might have to focus on crisis stabilisation, while later sessions might have to address deeper psychological issues or skill-building. A combination of psychotherapies must therefore need to adapt to the individuals’ changing needs.
Not only should therapists have flexibility to the patient’s unique needs via an integrative approach (Wachtel, 2018) but, such adaptability is likely to improve therapeutic alliance, and we know that safe and secure relationships between a therapist and patient are one of the main predictors of successful treatment (Boswell et al., 2013).
Overall, some emerging research indicates that integrative therapies can be as effective, if not more so, than single-modality approaches. (Cook, Schwartz and Kaslow, 2017). More effective treatments and interventions for those who do not respond to one or a first-line treatment are needed in order to meet the individual needs of patients who experience complex mental health difficulties, and we are certainly seeing a lot more complexity coming through the clinical door nowadays.
Clinicians may wish to adopt a flexible, integrative approach, combining therapies and adapting to evolving patient needs, to improve outcomes for those with complex mental health conditions.
Acknowledgements
Thanks to Dr Derek Tracy for reviewing drafts of this blog and providing constructive feedback.
Statement of interests
No conflicts of interests to declare.
Links
Primary paper
Other references
Boswell, J.F., Kraus, D.R., Miller, S.D. and Lambert, M.J. (2013). Implementing routine outcome monitoring in clinical practice: Benefits, challenges, and solutions. Psychotherapy Research, 25(1), pp.6–19. https://doi.org/10.1080/10503307.2013.817696
Cook, S.C., Schwartz, A.C. and Kaslow, N.J. (2017). Evidence-Based Psychotherapy: Advantages and Challenges. Neurotherapeutics, [online] 14(3), pp.537–545. https://doi.org/10.1007/s13311-017-0549-4
Citrome, L. (2010). Relative vs. absolute measures of benefit and risk: what’s the difference? Acta Psychiatrica Scandinavica, 121(2), pp.94–102. https://doi.org/10.1111/j.1600-0447.2009.01449.x
Cristea IA, Gentili C, Cotet CD et al. Efficacy of psychotherapy for borderline personality disorder: a systematic review and meta-analysis. JAMA Psychiatry 2017;74:319-28.
Cuijpers P, Quero S, Noma H et al. Psychotherapies for depression: a network meta-analysis covering efficacy, acceptability and long-term outcomes of all main treatment types. World Psychiatry 2021;20:283-93.
Leichsenring F, Steinert C, Rabung S et al. The efficacy of psychotherapies and pharmacotherapies for mental disorders in adults: an umbrella review and meta-analytic evaluation of recent meta-analyses. World Psychiatry 2022;21: 133-45.
Published guidance, NICE advice and quality standards | Guidance | NICE
Mavranezouli I, Megnin-Viggars O, Daly C et al. Psychological treatments for post-traumatic stress disorder in adults: a network meta-analysis. Psychol Med 2020;50:542-55
Mulrow, C.D. (1994). Systematic Reviews: Rationale for systematic reviews. BMJ, 309(6954), pp.597–599. https://doi.org/10.1136/bmj.309.6954.597
Papola P, Miguel C, Mazzaglia M et al. Psychotherapies for generalized anxiety disorders in adults: systematic review and network meta-analysis of randomized-controlled trials. JAMA Psychiatry 2023;81:250-9.
Reid JE, Laws KR, Drummond L et al. Cognitive behavioural therapy with exposure and response prevention in the treatment of obsessive-compulsive disorder: a systematic review and meta-analysis of randomised controlled trials. Compr Psychiatry 2021;106:152223.
Wachtel, P.L. (2018). Pathways to progress for integrative psychotherapy: Perspectives on practice and research. Journal of Psychotherapy Integration, 28(2), pp.202–212. https://doi.org/10.1037/int0000089.
World Health Organization. World mental health report: transforming mental health for all. Geneva: World Health Organization, 2022.
