Prospective evaluation of a clinical decision support system in psychological therapy.

Abstract

Objective: Thus far, most applications in precision mental health have not been evaluated prospectively. This article presents the results of a prospective randomized-controlled trial investigating the effects of a digital decision support and feedback system, which includes two components of patient-specific recommendations: (a) a clinical strategy recommendation and (b) adaptive recommendations for patients at risk for treatment failure. Method: Therapist-patient dyads (N = 538) in a cognitive behavioral therapy outpatient clinic were randomized to either having access to a decision support system (intervention group; n = 335) or not (treatment as usual; n = 203). First, treatment strategy recommendations (problem-solving, motivationoriented, or a mix of both strategies) for the first 10 sessions were evaluated. Second, the effect of psychometric feedback enhanced with clinical problem-solving tools on treatment outcome was investigated. Results: The prospective evaluation showed a differential effect size of about 0.3 when therapists followed the recommended treatment strategy in the first 10 sessions. Moreover, the linear mixed models revealed therapist symptom awareness and therapist attitude and confidence as significant predictors of an outcome as well as therapist-rated usefulness of feedback as a significant moderator of the feedback-outcome and the not on track-outcome associations. However, no main effects were found for feedback. Conclusions: The results demonstrate the importance of prospective studies and the highquality implementation of digital decision support tools in clinical practice. Therapists seem to be able to learn from such systems and incorporate them into their clinical practice to enhance patient outcomes, but

Predicting patients who will drop out of out-patient psychotherapy using machine learning algorithms

“…Over recent years, machine-learning approaches in particular have had a large impact on prediction modelling and on the most recent debate about the implementation of personalised or precision medicine concepts in mental health. 10,11 Machine learning has been applied in various prediction contexts, [12][13][14][15] taking advantage of the ability to capture non-linear relationships. 16 Nevertheless, machine learning does not always have an advantage over more traditional methods, 17 indicating that personalised medical care faces serious challenges that cannot be addressed through algorithmic complexity alone.…”

Evidence-based tailoring of treatment to patients, providers, and processes: Introduction to the special issue.

“…As noted, the process of treatment can also be continually tailored to fit the patient's characteristics and needs based on data collected routinely throughout the course of the treatment. The study by Lutz et al (2021) demonstrated how such data-based feedback systems can provide a powerful tool for such continued tailoring. Moreover, not only can the packages and parameters of treatment be personalized for each patient but so too can the therapeutic content.…”

“…With a look toward the future, these aims might eventually be integrated to create an evidence-based decision tree. In this process, the individual-specific data would be fed into the system to support both pretreatment decisions in choosing the therapist and treatment, as well as the continual tailoring of treatment strategies to the patient and their evolving needs (Lutz et al, 2021).…”

“…For example, Bruijniks et al (2021) used patient-reported depressive symptom severity as the outcome measure, and Minor et al (2021) assessed psychotic symptoms. At the same time, many of the studies measured outcome beyond symptom reduction, including functioning (Boswell et al, 2021; Coyne et al, 2021; Lutz et al, 2021), affect (Minor et al, 2021), interpersonal relationships (Boswell et al, 2021; Coyne et al, 2021), quality of life (Boswell et al, 2021; Coyne et al, 2021), and general well-being (Lutz et al, 2021; Minor et al, 2021). Other relevant measures, especially for new interventions, include feasibility, acceptability, and dropout rate (Langer et al, 2021; Minor et al, 2021).…”

“…When the tailoring process is based on the patient’s ongoing clinical status throughout treatment, then in-treatment progress and symptom severity measures are used. For example, Lutz et al (2021) based their in-treatment tailoring process on indices like whether the patient was not on track (based on whether they crossed a risk level of symptoms, above what was predicted for them), and whether the therapists were aware of their patients’ emotional and psychological state.…”