Measurement-based matching of patients to psychotherapists' strengths.

Abstract

Treatment personalization has evolved into an important zeitgeist in psychotherapy research. To date, such efforts have principally embodied a unidirectional focus on personalizing interventions to the patient. For example, earlier work in this area attempted to determine whether, on average, certain patients with certain characteristics or needs would respond better to one treatment package versus others. To the extent such aggregate "Aptitude × Treatment interactions" emerged, they could help guide overarching treatment selection. More recently, and drawing on technological and statistical advancements (e.g., machine learning, dynamic modeling), predictive algorithms can help determine for which individual patients certain treatment packages (DeRubeis et al., 2014) or specific during-session interventions within them (Fisher & Boswell, 2016) confer the most advantage for clinical improvement. Again, such work can help guide treatment decisions, though now at multiple care points. Although the aforementioned innovations in personalized psychotherapy have been leading-edge, precision care need not remain unidirectional. Rather, it can be complemented by efforts to personalize treatment decisions to the therapist. Namely, we can harness therapist effectiveness data to help ensure that therapists treat the patients they are empirically most equipped to help and use the interventions with which they have had the most empirical success. Such threads have been the focus of our team's novel, evolving, and multimethod work on improving psychotherapy by leveraging therapists' own practice-based evidence. (PsycInfo Database Record (c) 2024 APA, all rights reserved).