Can new approaches to synthesising evidence help achieve a consensus in psychotherapy research?

Abstract

Objective

The recent debate around the College’s Clinical Practice Guidelines on mood disorders have highlighted differences in opinion on interpreting evidence from randomised control trials (RCTs) for psychodynamic psychotherapy. This paper discusses new techniques of synthesising research evidence (e.g., umbrella reviews) that may help minimise disagreements in the interpretation of RCTs and foster greater consensus on treatment guidelines.

Conclusions

Findings from the latest umbrella review suggest that psychodynamic therapy is an evidence-based approach, among several, for common mental disorders.

The recent debate around the College’s Clinical Practice Guidelines (CPGs) on mood disorders have highlighted differences in opinion on interpreting evidence from randomised control trials (RCTs) for psychodynamic psychotherapy.^1–4 Of course, there are challenges in applying RCTs to psychotherapy, although these are not unique to this field and are shared with other non-pharmacological treatments (NPTs).^5–7 Examples of NPTs include surgery, physiotherapy, occupational therapy and rehabilitation.^5,6 All these are complex interventions with issues around coherence, compatibility, and comparability meaning it may be difficult to assess which component was responsible for any effect. Unlike trials of medication, the training and skills of the person delivering the treatment are also important, as well as their adherence to proper technique, which should be monitored over the course of the trial. ⁸

In comparison to pharmacotherapy, it is much more difficult to blind participants, therapists, and evaluators in NPT. ⁶ Design of the appropriate control condition, including the choice of the right controls, is also important but often overlooked. ⁷ For instance, while the actual treatment is often carefully planned and carried out by trained staff, the control group may receive either no treatment or just usual care.^7,8 This means the therapy’s apparent benefits might come from spending more time with the person delivering the treatment rather than the intervention itself. Ideally the control group should therefore receive an active treatment delivered with the same enthusiasm and number of sessions. ⁹ The therapist also needs to be sufficiently trained and to follow therapy guidelines (i.e. manualisation), with checks on whether this occurred (i.e. fidelity).^7,8

In spite of these challenges, there has also been growing evidence from systematic reviews and meta-analyses (SRMAs) of RCTs for psychotherapy. ¹⁰ This includes psychodynamic approaches, ¹¹ particularly when delivered as short-term therapy.^12,13 However, given the volume of literature, clinicians are often reliant on summaries of evidence such as CPGs rather than individual SRMAs. Traditionally, these have taken the form of narrative reviews. ¹⁴ However, synthesising evidence even from SRMAs of randomised controlled trials may be subject to differences of interpretation. ¹⁴ As a result, more systematic techniques have been developed to assist with the literature search, study selection, data extraction, statistical analysis (if appropriate), the grading of evidence, and interpretation of findings.^14–16 Examples include meta-reviews, umbrella reviews and overviews of systematic reviews. ¹⁴ Exact definitions vary but share common themes of methodological reproducibility and transparency. The aim is to provide a critical and comprehensive summary of the evidence from existing systematic reviews, including, where possible, a synthesis of the results.^14,15

This approach has recently been applied to psychodynamic therapy as an empirically supported treatment for depressive, anxiety, personality (PDs), and somatic symptom disorders (SSDs). ¹⁷ In this case, it was an umbrella review restricted to meta-analyses of randomised controlled trials (RCTs) with a focus on papers that were published in the previous 2 years. There were comparisons of psychodynamic therapy against all types of control (inactive and active), active controls (e.g. treatment as usual or supportive therapy) and active therapies (e.g. pharmacotherapy or another form of psychotherapy). ¹⁷ Included studies were rated by at least two authors for treatment effect, bias, inconsistency (i.e. heterogeneity), indirectness (i.e. lack of generalisability), imprecision (i.e., results were only from one or two small studies), publication bias, treatment fidelity, and quality of evidence. Assessment tools included the Critical Appraisal Checklist for Systematic Reviews and Research Syntheses developed by the Joanna Briggs Institute and the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) system.^16,18

Eleven SRMAs were identified. ¹⁷ Psychodynamic therapy resulted in better outcomes than both inactive and active controls and was equally effective as other active therapies in all four disorders. ¹⁷ Any differences to other active treatments were small and not clinically significant. Evidence was of high quality for depression and SSDs, and moderate for anxiety and PDs. Results were consistent across studies, and there were no major methodological issues except possible publication bias in the case of SSDs. Where there were limitations in specific research areas, such as risk of bias and imprecision, these were no worse than in other evidence-based psychotherapies. There was also some preliminary evidence that psychodynamic therapy was cost-effective in all the disorders. ¹⁷

It has to be acknowledged that umbrella reviews have drawbacks and may not be suitable for every clinical decision or situation. Results depend on the quality of included SRMAs, which can vary, and might not capture nuances specific to individual studies. Additionally, heterogeneity among studies, and differences in methodologies, can pose challenges in drawing clear conclusions.

Other factors in determining the choice of therapy

There are, of course, additional factors to RCT evidence in determining treatment choice such as therapist availability and patient preference. ⁸ This is explicitly recognised in the most recent NICE guidelines published in 2022 in the role of psychodynamic psychotherapy for the treatment of depression. ¹⁹ NICE considered that there was evidence for the effectiveness of short-term psychodynamic psychotherapy in the treatment of depression but rated this approach lower than alternatives because it did not appear, on average, to be as cost-effective. However, the NICE guidelines also acknowledged that treatment choice should be based on individual preferences, and it was therefore reasonable for people to request alternatives to NICE’s recommendations for first-line treatment. ¹⁹ Importantly, the NICE guidelines were published after the most recent umbrella review of psychodynamic psychotherapy as previously discussed. ¹⁷

Conclusions

In conclusion, new techniques of synthesising research evidence may help minimise disagreements in the interpretation of RCTs and foster greater consensus on treatment guidelines. In particular, these advances offer the possibility of greater standardisation in applying results to clinical practice. Using these techniques, psychodynamic therapy appears to be an evidence-based approach, among several, for common mental disorders. ¹⁷ Although psychotherapy can be subject to rivalry between different techniques, patients should therefore have access to a choice of treatments that have been proven effective through research. ¹⁷ This matters for patient care because not every person responds in the same way to any given type of therapy, as shown by modest success rates across all evidence-based treatments. ¹⁷ There are flow charts that can help in the choice of the most appropriate psychotherapeutic intervention based on what is available and the best fit between the patient, therapist, and therapy type. ²⁰