Psychotherapy Works – An Inclusive and Affirming View to a Modern Mental Health Treatment

There is comprehensive meta-analytic evidence that psychotherapy works across the most common mental health conditions. Surprisingly – and perhaps more controversially – there is robust evidence that various psychotherapy orientations work well, and when intended to be therapeutic (i.e., bona fide therapies) are approximately equally effective. That is, the ongoing controversy between psychotherapy orientations may make the potential differences between orientations appear larger than can be empirically supported, which is particularly true for long-term follow-ups. In this commentary, we provide explicit definitions of psychotherapeutic treatments and their mental health outcomes. Furthermore, we provide evidence-based examples of studies comparing the lasting efficacy at follow-up of particular legitimate psychotherapies vis-à-vis other legitimate psychotherapies. Next, we provide examples of evidence-based psychotherapy principles and skills based on recently conducted meta-analytic summaries. Last, we discuss implications for therapeutic practices and training, and conclude with some more general guidance for future directions.

How Can Legitimate (Bona Fide) Psychotherapy Be Characterized and Why Is It Relevant to Identify Non Bona Fide Psychotherapy Conditions?

Psychotherapy is a highly collaborative and individualized mental health practice developed in (post-) modern societies (e.g., Elias, 1978; Huft, 2022). One of the main characteristics of psychotherapy is its collaborative nature. Furthermore, patients have a proactive role and are engaged in therapy. The key point is that psychotherapy is a socioculturally embedded mental health practice, with patients and society expecting a highly custom-tailored and collaborative talking cure to reduce target complains and fostering well-being and psychosocial functioning to proactively work on the clients´ sufferings.

One of the major challenges in psychotherapy research is that psychotherapy and its minimal standards are often not explicitly defined. We define the minimal standards of legitimate (bona fide) psychotherapy as follow (Wampold et al., 1997; Wampold & Imel, 2015): Psychotherapists with at least a master's degree provided treatment, and two of the following criteria are required: (a) treatment is generally recognized as legitimate psychotherapy, such as CBT or psychodynamic therapy, and therapists are not prohibited from using recognized therapeutic interventions, such as psychoeducation, empathy, a rationale for treatment, promotion of coping skills, or (b) the description of treatment includes a reference to a psychological mechanism (e.g., operant conditioning), or (c) a manual/guide is used, or (d) the treatment includes an active component that has appeared in the psychological literature. Careful consideration of the minimum standards of legitimate psychotherapy is a prerequisite for including only such trials in meta-analyses of psychotherapy comparisons to minimize research bias and triviality of findings.

With respect to non bona fide psychotherapy conditions, such conditions often are constructed to not be fully therapeutic (e.g., Westen et al., 2004; Munder et al., 2019; Flückiger et al., 2022). Non bona fide psychotherapy conditions often can be identified by looking at what interventions are excluded or even “banned”. Examples of non bona fide psychotherapy in the above-mentioned sense are: discussion group where the leaders were instructed to “not teach skills or differentially reinforce coping strategies” (Wetherell et al., 2003, p. 33), psychoeducation where “instructors were asked, as best as they could, not to teach any skills in a way that may enhance mindfulness” (Wong et al., 2016, p. 69), nondirective therapy where “direct suggestions, advice or coping methods were prohibited” (Borkovec & Costello, 1993, p. 613), three session contact control group to motivate patients “to wait for the start of the treatment” (Linden et al., 2005, p. 37).

In many cases such non bona fide “intent-to-fail”-interventions are used to demonstrate the relevance of specific components of the favored treatment (Westen et al., 2004). However, these conditions are not informative about the efficacy of the approach compared to legitimate psychotherapy and provide little information about specific ingredients.

How Can Psychotherapy Outcomes Be Defined?

WHO defines mental health as follows:

Consistent with the above-mentioned broad definition of the World Health Organization (WHO), we understand mental health as a lifelong process of development and adaptation. Most importantly, mental health has at least two constituents, including the reduction of suffering/symptoms as well as the promotion of well-being, personal values, and strengths (e.g., Schürmann-Vengels et al., 2022). In psychotherapy, collaborative efforts toward symptom reduction and well-being represent psychotherapeutic outcomes that result from careful therapeutic exploration (grosse Holtforth et al., 2004). The point relevant to this Special Issue on transtheoretical clinical training and practice is that psychotherapy outcomes go beyond the tailored ("primary") outcomes of a particular psychotherapy and should cover the broad spectrum of the WHO definition of mental health.

Does a Particular Bona Fide Psychotherapy Have a More Lasting Effect Than Another?

Although somewhat arbitrary in nature, a threshold for a clinically relevant treatment effect in anxiety and mood disorders has been estimated to be about d = .25 (Cuijpers et al., 2014). Comparative meta-analyses report small to negligible relative differences in efficacy if two or more bona fide psychotherapies are compared directly in an RCT design (Wampold & Imel, 2015).

With respect to long-term outcomes, Kivlighan and co-authors (2015) for example examined the long-term follow-ups of bona fide psychodynamic psychotherapies compared with non-psychodynamic bona fide psychotherapies (k = 25). They hypothesized that psychodynamic psychotherapies would perform better over time than other treatments (assuming a positive and statistically significant slope effect from assessment post-therapy to follow-up). However, they found an effect of d_slope = .00 for disorder-specific targeted outcomes, nontargeted outcomes, and personality outcomes. These results suggest that the efficacy of psychodynamic therapies is generally not more sustainable but also not less sustainable after active treatment compared with other bona fide psychotherapies. Several other longitudinal multilevel meta-analyses on interventions for anxiety and depression did not show any significant increase or decrease in relative effects over follow-up time within studies, such as for cognitive interventions versus behavioral interventions (Podina et al., 2019), studies with additive components (Flückiger et al., 2015), and established cognitive behavioral therapy or augmented integrative cognitive behavioral therapy for generalized anxiety disorder under bona fide conditions (Flückiger et al., 2022; for all studies d_slope < .10). Again, these meta-analytic findings show that differential long-term outcomes for particular bona fide psychotherapy approaches usually are not likely.

The point relevant to the Special Issue for transtheoretical training and practice is that the enormous effort to specify psychotherapeutic effects through very specific approach explains surprisingly little on the long run. Do we therefore face a shambles? No, we think not! On the contrary, as we will show in the following sections there is considerable meta-analytic evidence for the outcome-relevance of multiple transtheoretical principles and skills.

Is There an Evidence Base for Transtheoretical Psychotherapy Principles and Skills?

Psychotherapy represents a cooperative course of action between therapist and patient during and between psychotherapy sessions. One of the most substantiated findings in psychotherapy research is that transtheoretical collaborative qualities are robustly linked to treatment outcomes across many psychotherapy conditions. Based on an international meta-analytic summary of 295 studies representing more than 30,000 psychotherapies, there is a moderate predictive association of 8% explained variance (r = .278; confidence interval .256 ≤ r ≤ .299; Flückiger et al., 2018) between alliance measured mostly once during therapy and therapy outcome (at the end of therapy) in face-to-face and internet-mediated treatments. These statistically significant results confirm those of previous meta-analyses. A comparable predictive power of the working alliance could also be meta-analytically confirmed in therapies with children and adolescents (Karver et al., 2018), in couple and family settings (Friedlander et al., 2018) and groups (Lo Coco et al., 2022).

There seem to be therapists who are comparatively more successful in building working alliances with their patients than others. These differences between therapists are relevant for treatment success, i.e., therapists who on average build better alliances also treat somewhat more successfully. These moderate effects have been meta-analytically confirmed (Del Re et al., 2021).

It is hypothesized that the alliance-outcome relationship manifests itself in particular because of the patients' intake characteristics (e.g., Feeley et al., 1999). However, based on 66 studies reporting both uncontrolled predictor models and predictor models controlled for the intake variables, there is no systematic evidence that the alliance can be fundamentally understood as an epiphenomenon of the intake variables, arguing for the collaborative conception during treatment (r = .25 vs. .22; Flückiger, Del Re, et al., 2020). Furthermore, an individual participant data analysis of 17 studies indicated that in the early phase of therapy, symptoms and alliance were reciprocally related to one other, often resulting in a positive upward spiral of higher alliance/lower symptoms that predicted higher alliances/lower symptoms in the subsequent sessions (Flückiger, Rubel, et al., 2020). Overall, the available evidence across hundreds of studies indicates that the above-mentioned sociocultural key principle of psychotherapy as a highly collaborative mental health treatment is robustly linked to psychotherapy outcomes. For other concepts that partly overlap with the alliance concepts comparable correlation patterns have been shown (for example empathy, goal agreement and group cohesion; Norcross & Lambert, 2018).

With respect to particular therapist skills, a recent APA interdivisional effort investigated 27 well-accepted basic psychotherapy skills and methods (Hill & Norcross, 2023). The results of the meta-analytic summaries revealed that some transtheoretical psychotherapy skills such systematic feedback in routine outcome monitoring (Barkham et al., 2023), emotion-regulation strategies (Iwakabe et al., 2023) or strength-based methods (Flückiger et al., 2023) were evaluated as “demonstrably effective” for post-treatment outcomes (published open access in the journal Psychotherapy Research; Hill & Norcross, 2023). Overall, findings from process-outcome research have the potential to moderately improve psychotherapy interventions across psychotherapy orientations.

Clinical and Training Implications for Future Training and Practice in Transtheoretical Psychotherapy

Based on the reviewed literature, the following transtheoretical, transdiagnostic and interdisciplinary implications can be summarized for future clinical practice and training.

Concluding Comment for Future Directions

Mental health is a human right and a shared responsibility of societies that cannot be entirely delegated to particular professions nor achieved by certain treatments. At the same time, carefully conducted mental health treatments are cultural achievements. Transtheoretical therapeutic factors such as collaborative qualities are relevant across orientations but also across professions and settings (Wampold & Flückiger, 2023). On the healthy side of mental health, there is openness, individuality, and basic human freedom in how we create our lives. Accepting co-responsibility for psychotherapeutic outcomes is not only a critical interpersonal skill of therapists, but also entails an attitude of trust in joint exploration of what individual well-being means.