In the Stratified Medicine Approaches for Treatment Selection (SMART) Mental Health Prediction Tournament, 13 teams from around the world will compete to see who can build the best predictive model for anxiety and depression treatment response. Each team will be provided with the same large, anonymized mental health treatment outcome dataset from the UK’s national health system. A separate, held-out test sample will be used to determine the winner. This will provide a level playing field for evaluating each model’s efficacy, but it will also allow participants and, ultimately, the field, to understand the reasons for the advantages and disadvantages of each strategy, under different conditions. Head-to-head comparisons of the best approaches for selecting mental health treatments will, we believe, yield knowledge that can be used to maximize the efficiency of mental health care delivery in the future.
The first aim of the tournament will be to learn more about different methodological approaches to building predictive models for treatment selection in mental health. We also hope to contribute to the conversation about barriers to implementation of precision medicine approaches in real-world clinical settings, informed in part by work we will undertake with key stakeholders (service-users and clinicians) as part of this project.
The second aim is to produce an algorithm that could be used to inform treatment decisions in the UK’s IAPT (Improving Access to Psychological Therapies) services, with the goal of enhancing efficiency, efficacy, and service.