Predictors of remission from body dysmorphic disorder after internet-delivered cognitive behavior therapy: a machine learning approach

Abstract

Background: Previous attempts to identify predictors of treatment outcomes in body dysmorphic disorder (BDD) have yielded inconsistent findings. One way to increase precision and clinical utility could be to use machine learning methods, which can incorporate multiple non-linear associations in prediction models. Methods: This study used a random forests machine learning approach to test if it is possible to reliably predict remission from BDD in a sample of 88 individuals that had received internet-delivered cognitive behavioral therapy for BDD. The random forest models were compared to traditional logistic regression analyses. Results: Random forests correctly identified 78% of participants as remitters or non-remitters at post-treatment. The accuracy of prediction was lower in subsequent follow-ups (68, 66 and 61% correctly classified at 3-, 12-and 24month follow-ups, respectively). Depressive symptoms, treatment credibility, working alliance, and initial severity of BDD were among the most important predictors at the beginning of treatment. By contrast, the logistic regression models did not identify consistent and strong predictors of remission from BDD. Conclusions: The results provide initial support for the clinical utility of machine learning approaches in the prediction of outcomes of patients with BDD. Trial registration: ClinicalTrials.gov ID: NCT02010619.

The promise of machine learning in predicting treatment outcomes in psychiatry

“…In a study which included, in addition to demographic and clinical data, therapy‐related predictors of treatment credibility and working alliance, assessed at week 2 of treatment, Flygare et al 82 used a random forest algorithm to predict remission from body dysmorphic disorder after iCBT in a sample of 88 patients, comparing the results to logistic regression. Random forests achieved a prediction accuracy of 78% at post‐treatment, with lower accuracy in subsequent follow‐ups.…”

Internet-Based Treatment of Body Dysmorphic Disorder

“…Again, this result could be interpreted in terms of the possible skepticism toward psychological help that may persist even among individuals who give the online treatment a try. As Flygare et al (2020) showed, credibility was a predictor of remission in their online intervention for BDD. Future studies may need to focus on increasing credibility and expectancy and on setting realistic standards for change.…”

“…For eating disorders, results on iCBT are less clear and high attrition is observed (Linardon et al, 2020). In the context of BDD, three studies including one randomized controlled trial (RCT) and different reevaluations of the data have tested the so-called BDD-NET program (Drüge et al, 2022; Enander et al, 2014; Enander et al, 2016; Enander et al, 2019; Flygare et al, 2020), an eight-module iCBT program based on the model of Wilhelm et al (2013). The RCT found large between-group effect sizes in symptom severity in favor of the BDD-NET program compared with supportive control treatment (after the 3-month treatment, d = 0.95; follow-up after 6 months, d = 0.87) (Enander et al, 2016).…”

Predictors of pharmacotherapy outcomes for body dysmorphic disorder: a machine learning approach

“…That notwithstanding, successful machine learning studies have been accomplished using smaller sample sizes (e.g. Flygare et al, 2020), and approaches were adopted in the current study to optimize performance, given the sample size. Specifically, SVM algorithms have utility in smaller sample sizes (Boehmke & Greenwell, 2019), and 10-fold cross-validation was employed to prevent biasing the model training on any single subsample of the dataset.…”