For whom should psychotherapy focus on problem coping? A machine learning algorithm for treatment personalization

Schwartz, Brian; Giesemann, Julia; Rubel, Julian; Deisenhofer, Anne‐Katharina; Lutz, Wolfgang

doi:10.1080/10503307.2021.1930242

Cited by 18 publications

(15 citation statements)

References 64 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…For example, research has consistently demonstrated that therapist performance differences and the quality of the psychotherapy process explain more outcome variance than the difference between delivering one EST over another (Wampold & Imel, 2015). Thus, perhaps future personalization efforts will be more fruitful if they focus on matching patients to personally well-suited providers (e.g., Constantino et al, 2021) or change processes (e.g., Gómez Penedo et al, 2021).…”

Section: Discussionmentioning

confidence: 99%

Replicating patient-level moderators of CBT and IPT’s comparative efficacy for depression.

Coyne¹,

Constantino²,

Ouimette³

et al. 2022

Psychotherapy

View full text Add to dashboard Cite

Although evidence-based psychotherapies, such as cognitive behavioral therapy (CBT) and interpersonal psychotherapy (IPT), produce comparable average outcomes, it is plausible that some patients who possess one or more specific characteristics may respond better to one over the other. Addressing this what works best for whom question, researchers have tested the moderating influence of patient characteristics on comparative treatment effects (viz. aptitude–treatment interactions [ATIs]). However, few ATIs have emerged or replicated, thereby providing little treatment–selection guidance. Informed by a systematic review of patient ATIs in trials that compared CBT versus IPT for depression (Bernecker et al., 2017), this study aimed to replicate (a) significant ATIs previously established in a single study; and (b) significant ATIs previously examined twice, with only one study demonstrating a moderating effect. Data derived from a trial in which adult outpatients with major depression were randomly assigned to 16 weeks of CBT (n = 41) or IPT (n = 39). Patient characteristics were measured at baseline, and patients rated their depression throughout treatment. Multilevel models revealed one ATI replication; for patients with more self-sacrificing interpersonal problems, CBT outperformed IPT; the reverse was true for patients with fewer such problems. Other moderators either failed to replicate or directionally contradicted prior research. Results help inform optimal treatment matching for some patients, which reflects a type of psychotherapy personalization. However, they also highlight limitations of traditional ATI research and suggest that different methods are needed to inform responsive personalization efforts more expansively and reliably.

show abstract

Section: Discussionmentioning

confidence: 99%

Replicating patient-level moderators of CBT and IPT’s comparative efficacy for depression.

Coyne¹,

Constantino²,

Ouimette³

et al. 2022

Psychotherapy

View full text Add to dashboard Cite

show abstract

“…(i.e., Go ´mez-Penedo et al, 2023;Go ´mez Penedo, Schwartz, et al, 2022;Rubel et al, 2017;C. Schwartz et al, 2018).…”

Section: Clinical Impact Statementmentioning

confidence: 99%

“…Besides the well-established process of the alliance (Flückiger et al, 2018), grounded on a meta-analysis on empirical findings, Grawe (1997) identified four additional processes of change: (a) problem actuation (i.e., patients experiencing their problems in session); (b) resource activation (i.e., patients experiencing themselves as someone with resources and strengths to cope with their problems); (c) mastery (i.e., the patients’ ability to cope with their problems); and (d) clarification (i.e., the patients’ understanding of the sources and consequences of his or her own problematic behaviors and experiences). From these four change processes, mastery and clarification have been the two with strong transtheoretical (e.g., Allemand & Flückiger, 2017) as well as empirical attention in recent years, with several studies showing that they are significantly associated with psychotherapy outcome (i.e., Gómez-Penedo et al, 2023; Gómez Penedo, Schwartz, et al, 2022; Rubel et al, 2017; C. Schwartz et al, 2018).…”

mentioning

confidence: 99%

“…Besides these initial efforts predicting alliance–outcome associations using ML, some studies also tried to predict clarification and mastery effects in psychotherapy (e.g., Gómez Penedo, Schwartz, et al, 2022; Lutz et al, 2019, 2022). Considering the high correlations between clarification and mastery when self-reported by patients ( r = .65; Flückiger et al, 2010), as one solution proposed in the literature (Rubel et al, 2017), these studies had merged mastery and clarification processes into a broader intrapersonal change process named problem-coping experiences (i.e., the degree in which the patient works on finding solutions to their problems or understanding their sources).…”

mentioning

confidence: 99%

“…Results showed that when therapists followed the algorithmic recommendations during the first 10 sessions, effect sizes were better than if they did not follow the respective recommendation (∼ d = 0.30). Additionally, Gómez Penedo, Schwartz, et al (2022) applied different ML algorithms using patients’ baseline characteristics to predict problem-coping experience effects on outcome during the first 10 sessions of therapy. The best-fitting algorithm in the training set presented adequate and stable results in the test set ( R 2 = .15, d = 0.84) 1…”

mentioning

confidence: 99%

See 2 more Smart Citations

Using machine learning algorithms to predict the effects of change processes in psychotherapy: Toward process-level treatment personalization.

Gómez Penedo,

Rubel,

Meglio

et al. 2023

Psychotherapy

Self Cite

View full text Add to dashboard Cite

This study aimed to develop and test algorithms to determine the individual relevance of two psychotherapeutic change processes (i.e., mastery and clarification) for outcome prediction. We measured process and outcome variables in a naturalistic outpatient sample treated with an integrative treatment for a variety of diagnoses (n = 608) during the first 10 sessions. We estimated individual within-patient effects of each therapist-evaluated process of change on patient-evaluated subsequent outcomes on a session-by-session basis. Using patients’ baseline characteristics, we trained machine learning algorithms on a randomly selected subsample (n = 407) to predict the effects of patients’ process variables on outcome. We subsequently tested the predictive capacity of the best algorithm for each process on a holdout subsample (n = 201). We found significant within-patient effects of therapist perceived mastery and clarification on subsequent outcome. In the holdout subsample, the best-performing algorithms resulted in significant but small-to-medium correlations between the predicted and observed relevance of therapist perceived mastery (r = .18) and clarification (r = .16). Using the algorithms to create criteria for individual recommendations, in the holdout sample, we identified patients for whom mastery (14%) or clarification (18%) were indicated. In the mastery-indicated group, a greater focus on mastery was moderately associated with better outcome (r = .33, d = .70), while in the clarification-indicated group, the focus was not related to outcome (r = −.05, d = .10). Results support the feasibility of performing individual predictions regarding mastery process relevance that can be useful for therapist feedback and treatment recommendations. However, results will need to be replicated with prospective experimental designs.

show abstract

Leveraging Novel Technologies and Artificial Intelligence to Advance Practice-Oriented Research

Atzil-Slonim,

Penedo,

Lutz

2023

Adm Policy Ment Health

View full text Add to dashboard Cite

For whom should psychotherapy focus on problem coping? A machine learning algorithm for treatment personalization

Cited by 18 publications

References 64 publications

Replicating patient-level moderators of CBT and IPT’s comparative efficacy for depression.

Replicating patient-level moderators of CBT and IPT’s comparative efficacy for depression.

Using machine learning algorithms to predict the effects of change processes in psychotherapy: Toward process-level treatment personalization.

Leveraging Novel Technologies and Artificial Intelligence to Advance Practice-Oriented Research

Contact Info

Product

Resources

About