Disentangling psychological interventions for mental disorders into a taxonomy of active ingredients

Project Information

DECOMPOSE

Grant agreement ID: 101042701

DOI

10.3030/101042701

EC signature date 31 March 2022

Start date 1 November 2022

End date 31 October 2027

Funded under

• European Research Council (ERC)

Total cost

€ 1 497 500,00

EU contribution

€ 1 497 500,00

Coordinated by

UNIVERSITA DEGLI STUDI DI PADOVA
Italy

Periodic Reporting for period 1 - DECOMPOSE (Disentangling psychological interventions for mental disorders into a taxonomy of active ingredients)

Reporting period: 2022-11-01 to 2025-04-30

Summary of the context and overall objectives of the project

Psychological interventions, alone or combined with other treatments, improve symptoms of mental disorders, including severe ones like psychotic or borderline personality. However, despite tremendous efforts, fundamental questions remained unanswered. For example, what are the mechanisms through which psychotherapies impact symptoms or other important outcomes, or what types of patients would benefit more from some interventions rather than from others. In DECOMPOSE, we believe that a root cause for this lack of progress lies within the very nature of psychotherapies, as complex interventions composed of multiple, perhaps interacting, parts. Psychotherapies are developed and studied as packages, integrating a diversity of elements, often combined without clear or evidence-based rules. Randomized trials, the gold standard method to evaluate treatments, are conducted to assess whether these packages are effective. Drugs, the other main treatment for mental disorders, are similarly evaluated. But unlike for drugs, we know very little about the ‘ingredients’ that compose a psychotherapy treatment package. To address this knowledge gap, DECOMPOSE proposes a radical shift from approaching interventions as brands and packages towards active ingredients.
DECOMPOSE will employ a systematic, comprehensive and reproducible approach for decoding, classifying and evaluating the active ingredients of psychological interventions. The goals are to break down psychotherapies into their constituent elements, such as techniques or processes, and use these to create a taxonomy of psychotherapy components. The taxonomy, in conjunction with advanced statistical methodologies called component network meta-analyses, will be used to reevaluate which elements contained in psychotherapy packages are more effective and for which types of patients. Finally, the findings will be used to inform a clinical decision support system to assist clinicians in “assembling” and “dismantling” interventions out of components, which may be necessary in situations of limited resources or time. By changing the lens from treatment packages to active ingredients, DECOMPOSE could revolutionize psychotherapy research. It promises to fine-tune effectiveness, inaugurating new paths for research into mechanisms and treatment personalization, supporting mechanistic intervention development and expediting dissemination of effective psychological treatments to reduce the global burden of mental disorders.

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

There are countless psychological interventions developed for mental disorders. Therefore, in the first step of DECOMPOSE, we needed to assemble a collection of interventions for which to identify active ingredients. We chose to focus on the best interventions, also called “evidence-based”- the ones shown to be effective in randomized controlled trials, the gold standard design to evaluating treatments. To assemble a large collection of evidence-based psychological interventions, we used network meta-analyses as our starting point. Network meta-analyses estimate treatment effects by systematically assembling large collections of trials, which compare a variety of interventions for the same condition. From large network meta-analyses for six distinct mental disorders (schizophrenia and other psychotic disorders, borderline personality disorder, substance use disorders, bipolar disorder, anorexia nervosa and bulimia nervosa), we assembled a cohort of 260 randomized trials. A total of 401 psychological interventions had been tested in these trials and were selected for dismantling. In the subsequent step of DECOMPOSE, we searched for detailed intervention descriptions that we could use to parse out constituent elements. These descriptions are usually in the form or protocols or treatment manuals. We used a multipronged approach, including identifying published protocols, searching trial registries, contacting authors and, as a last step, looking for commercially available protocols or manuals. Identifying detailed intervention descriptions proved challenging. However, we were able to retrieve protocols or manuals for 80% of the psychological interventions selected. In the subsequent step, teams of two researchers, working independently, read intervention descriptions from protocols or manuals and dismantled them into constituent elements. The teams worked iteratively, focusing on groups of five interventions, and then checking consensus. For each component, uniform labels and definitions were discussed between team members. These were then used for the extraction from the next group of 5 interventions. Once this phase is finalized, the complete list of components identified for each disorder, together with the standardized and agreed-upon definitions will be integrated in a taxonomy of active ingredients.

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

In the process of searching for detailed intervention descriptions, DECOMPOSE produced a large-scale, unprecedented examination of the accessibility and public availability of treatment protocols and manuals for evidence-based psychological interventions. Treatment protocols and manuals are equivalent of the drug package insert or label in drug regulation, which specify ingredients, clinical applications, conditions of use and counterindications. Using a multipronged approach (published protocol, trial registries, author contact, commercial availability), we quantified what proportion of protocols or manuals we were able to retrieve, as well as what proportion was publicly available. We identified no other similar investigation for any mental disorder, much less one of this scale. Overall, for a fifth of the psychological interventions, we could not retrieve detailed description to allow identification of components. Less than a quarter of the protocols and manuals were publicly available. Relying only on open-access resources and sharing from authors, a researcher would be able to access about one third of trial protocols or manuals. Our investigation is singular in its breadth and systematic approach and the findings will likely spur a lot of discussion. Key actors in the field such as funders, editors, clinical guideline panels, policy makers, and others might reconsider the evaluation of psychological treatments generally, so as to ensure the content of an intervention proved effective can also be known, as a minimum for research purposes. Additionally, our approach can provide a blueprint for examining access to detailed intervention descriptions for other mental disorders, thereby filling the substantial gap in the literature and informing the identification of intervention components for other disorders. Without access to protocols or manuals, the content of psychological interventions remains a black box. Moreover, improving dissemination and implementation of evidence-based treatments requires publicly available treatment manuals.

Logo