Machine Learning for Psychotherapy Outcome Predictions

Prioritizing the right patients and providing personalized treatment in a timely manner is crucial to improve access to healthcare. In psychotherapy at least 1 in 3 patients drop out of treatment with therapeutic alliance and patient motivation among the common predictors. Recommendations include strengthening the patient-therapist bond through developing common goals and checking in on progress and treatment path. Using a sample of 10,363 mental health from the USA, we used machine learning to develop a clinicalfeedback support tool to encourage patient-therapist goal alignment. A gradient boosted decision tree was trained on pre-treatment patient-reported data to provide predictions of early treatment dropout, treatment duration, and symptom outcomes conditional on different treatment durations in out-of-sample patients. The models improved performance versus baseline predictions. The resulting decision support tool could assist in the collaborative selection of treatment goals, appropriate treatment intensity, and optimal allocation of resources. Results are discussed in the context of explainable AI and the ethical implications of predictive modeling in this context.