The Future of Enhanced Psychotherapy: Towards Precision Psychotherapy

The core idea of enhanced psychotherapy is to enrich established psychotherapy by complementing it with innovative treatment modules, approaches, or modes. Traditional psychotherapy in this sense comprises evidence-based talking therapies established in the different therapy schools provided in face-to-face settings. The overarching goal of this enhancement comprises the overall improvement of treatment efficacy and reaching a higher sustainability of treatment outcomes, especially for (core) symptoms and disorders which are hard to treat and for patients who profit less from traditional psychotherapy.

A specific aspect of enhanced psychotherapy is the development of treatment strategies towards increased personalization in the sense of precision psychotherapy [1] within a framework of precision mental health care [2] and precision medicine [3]. An important prerequisite for this strategy is a strong empirical backbone for the development, further refinement, and implementation of these new approaches. Importantly, the approach is twofold: new interventions or dissemination strategies are developed, and existing interventions are used but implemented in a more specific and precise way. This entails on the one hand that the development of novel interventions is closely informed by evidence on disorder or maintenance mechanisms, as it has previously been demanded for specific severe disorders [4], and on the other hand a close monitoring of treatment processes and outcomes by means of a data-informed and measurement-based approach [5] which allows precise evaluation, adaption, and tailoring of interventions to reach improved personalization as outlined above.

As a conceptual umbrella and strategy, enhanced psychotherapy can include a broad range of approaches, e.g., the use of digital technology such as virtual reality environments to facilitate exposure therapy, avatar therapy, wearable devices to track physiological responses, and computerized cognitive behavioural therapy, but also the combination of psychotherapy with computerized cognitive training approaches, targeting, for instance, dysfunctional appraisals of trauma, or the incorporation of neuromodulatory techniques. Enhancement of “traditional” psychotherapy can also entail microsocial interventions involving peers or innovative transgenerational treatment approaches in parent-child dynamics. Similarly, interventions that better incorporate the living environments of those affected (e.g., home treatment) and the inclusion of relevant living environments (e.g., school and the work environment) should also be better integrated in the context of enhancement. Table 1 gives an overview of examples of enhanced psychotherapy approaches, including an outlook on future directions.

While it is great to see such progress in the number of RCTs on the efficacy and effectiveness of psychotherapy for different disorders and treatment settings [25, 26], it is important to note that, as with most areas of medicine and psychology, more development is needed in certain areas. Nevertheless, we can be confident that with continued research and development we will be able to improve the effectiveness of psychotherapy even further. Treatment responses vary among patients [25], but with evidence-based therapies, there is hope for remission and relief from residual symptoms or relapse. We understand that access to these therapies can be difficult, especially for children, adolescents, and their families, as well as affected adults and their relatives. However, we are confident that with the appropriate interventions, we will be able to improve the quality of life of people with mental disorders.

When we look at health from a holistic bio-psycho-social perspective, as defined by the World Health Organization within the framework of One Health [27] we often find that therapeutic approaches only target a narrow concept of disorder and symptoms. The concept of One Health posits that human health is closely intertwined with the health of other animals and the environment that they inhabit [27]. However, we can take a more comprehensive approach to health by considering all aspects of a person’s well-being. By doing so, we can provide more effective and personalized treatments that address the underlying causes of health issues. The role of psychopharmacotherapy in improving mental health treatment effectiveness has been less prominent in recent years, but there have been interesting developments with agents with longer histories (e.g., ketamine) or use in different settings (e.g., psychedelics) [18, 28], although short- and long-term adverse effects are not yet fully understood [29]. There are also promising developments that allow early identification of treatment interventions, combined with targeted, mechanism-oriented interventions and rigorous treatment evaluation. These approaches include the use of digital approaches and artificial intelligence [30]. With these complementary advances, we can confidently say that the future of mental health treatment is looking brighter.

A patient starting an evidence-based treatment, even with an initially average level of effectiveness, might not necessarily attain a successful outcome [2]. Therefore, in recent years a new precision approach to mental health research has emerged using advanced statistical methods (machine learning) and novel designs to tailor and monitor interventions, especially for patients at risk for negative treatment outcomes [5]. Currently, this line of research is mostly in the pre-clinical phase, leveraging existing data sets to aid the development of data-driven diagnostic and prognostic tools that could help to guide treatment personalization. Such tools need to be prospectively tested in clinical trials to move the field forward. Given that traditional clinical trial designs focus on average treatment effects, a move towards personalization also requires clinical trial designs that enable us to better understand the heterogeneity of treatment response at an individual – rather than group – level. Here, we discuss some design options that could be effective in this regard, such as adaptive platform and pragmatic trials [31, 32], keeping also more general recommendations for trials on psychological interventions in mind [33].

Adaptive platform trials enable the concurrent evaluation of multiple interventions, allowing the incorporation of new treatments into the trial structure as they become accessible via intervention-specific subtrials. Such subtrials are either added or discontinued based on the results of interim analyses. This can result in a reduction of required patients in a study or patients assigned to the control group while still maintaining statistical power [34].

For example, the leapfrog design demands fewer participants compared to standard trial designs through sequential Bayesian analyses that early identify and eliminate ineffective treatments (e.g., [35]). Moreover, this design accommodates the integration of new treatment arms during an ongoing trial, ensuring adaptability to emerging research developments. This also allows a close connection between basic science, clinical translation, and implementation research [35].

The sequential multiple assignment randomized trial (SMART) design is another type of adaptive clinical trial design that allows the testing of dynamic treatment strategies and evaluates not only the effectiveness of individual treatments but also the sequencing and timing of treatments [36]. This design might be particularly valuable in developing adaptive intervention strategies for patients at risk for negative outcomes in the psychological therapies where the optimal course of interventions potentially needs to be adapted based on a patient’s (negative) response during treatment [5]. Such adaptive trial designs have great potential to enable the development of data-driven personalization and decision-support tools. Once such tools are developed, the next phase involves testing their effectiveness in real-world settings. Pragmatic trials (as RCTs implemented in routine care settings) can help to test the clinical and cost-effectiveness of personalization or decision-support tools in subsamples within larger naturalistic samples (e.g., see [7] and can move the field much closer to real-world implementation of data-driven psychotherapy [1]). Such studies would also allow, if successfully conducted, an appropriate generalization to the target patient group in clinical practice as well as potential limits of applicability for other clinical groups.

Another promising pathway to achieve personalization is modular therapy [6, 37, 38]. Modular approaches provide clinicians with an evidence-based toolbox to integrate treatment modules systematically as independent but combinable sets of functional units. By tailoring module selection and application to the specific characteristics and needs of each patient, modular therapy promises higher acceptance by patients and therapists as well as better treatment outcomes.

Up to now, little is known about how interventions should be structured in modular or enhanced psychotherapies. Different from most practising clinicians who follow eclectic treatment approaches according to their expertise and intuition, empirically based rules are needed to arrive at replicable treatment standards. In the future, a combination of theory- and data-based algorithms may be optimal, with machine learning analyses of actuarial data to build replicable processes and outputs. Further challenges are how to combine and integrate modules from different approaches which represent different therapeutic attitudes (such as a learning stance in CBASP, a discovering stance in mentalization, a spiritual stance in mindfulness, and a well-being stance in well-being therapy) during the same therapy or even the same session. Similarly, teaching and supervising numerous possible modules from numerous different therapy approaches may be a challenge. Last, but not least, while all these approaches are highly promising as distinct pathways to achieve personalization, clear evidence that data-driven or modular approaches to personalization are more effective than standard therapy is still pending.

In 2024, the German Centre for Mental Health (DZPG, https://www.dzpg.org/en/) has been launched [39]. Funded by the Federal Ministry of Education and Research (BMBF) and the ministries of the German federal states, the DZPG will finally have an annual budget of about €30 million and is expected to significantly improve the implementation of research in the field of mental health and the prevention and treatment of mental disorders. A particular focus will be on accelerating the translation of research findings into routine care. One important goal is to improve psychotherapeutic approaches across the entire lifespan. By pooling expertise and networking of clinical institutions, (large) data sets will be generated and shared.

Importantly, the DZPG also holds a strong Patient and Public Involvement stance, ensuring that individuals affected by mental health disorders are structurally and practically involved in research and dissemination across all phases as experts by experience [40]. Hence, enhanced psychotherapy research in the DZPG involves experts by experience as stakeholders to ensure participatory advancement of innovative treatment approaches, i.e., by establishing lived experience councils as part of clinical studies [41]. Another important long-term strategy of the DZPG, which is also aimed at ensuring progress towards a more data-driven precision psychotherapy, is to rely on computational models and approaches, implementing tools and analysis strategies for the integration of multiple data sources (e.g., PROMS [42], expert ratings, ecological momentary assessments, and biomarkers) [43]. To this end, an infrastructure for psychotherapy is being established within the DZPG, which will provide services for conducting psychotherapy studies for all sites involved in the DZPG (Berlin-Potsdam, Bochum-Marburg, Halle-Jena-Magdeburg, Mannheim-Heidelberg-Ulm, Tübingen, and Munich-Augsburg).

The Psychotherapy Research Infrastructure will provide a durable digital research, training, and data infrastructure that will enable harmonized assessments, long-term outcome data collection, and systematic coordination and quality management of psychotherapy research across the age range from infancy to older adults and across all disciplines and settings relevant to psychotherapy. It systematically links rapid exchange with basic research, the development and testing of existing and novel psychotherapeutic, psychosocial, and enhancement strategies under controlled research conditions, and the collection of large data sets under routine clinical conditions. The above clinical research designs and field experiments (e.g., testing new interventions vs. standard routine care) can be conducted rapidly and with large samples (e.g., 20 clinics vs. 20 others; including 1,000–5,000 patients), systematically addressing typical problems such as recruitment issues, lack of standardized assessments, and replicability. The coordinated approach optimizes the use of ecologically valid routine care data and accelerates efficient clinical translation.

The newly founded German Centre for Mental Health (DZPG) has identified as one of its main flagship projects the advancement of psychotherapy across the entire lifespan. Collaboratively, DZPG research will work towards innovative approaches in the sense of “enhanced psychotherapy,” enriching traditional established psychotherapy to reach better treatment efficacy and sustainability.

Enhanced psychotherapy comprises two strategies: the development and dissemination of novel mechanism-informed interventions and the more specific and precise implementation of established interventions, both resulting in a stronger precision psychotherapy approach. DZPG will draw back on long-term resources and infrastructures and a strong collaborative network in the field of enhanced psychotherapy, including a strong Patient and Public Involvement and computational approaches for data integration, which will ultimately enable accelerated translation of mental health care.

The authors have no conflicts of interest to declare.

Parts of the editorial were supported by the German Centre for Mental Health (DZPG), Bundesministerium für Bildung und Forschung (01EE2301G).

Stephan Zipfel, Wolfgang Lutz, Silvia Schneider, Elisabeth Schramm, Jaime Delgadillo, and Katrin E. Giel were involved in drafting and critically revising this manuscript.