Department of Anesthesiology, Washington University in Saint Louis, United States of America; Institute for Informatics, School of Medicine, Washington University in Saint Louis, United States of America; Deparment of Computer Science and Engineering, McKelvey School of Engineering, Washington University in Saint Louis, United States of America.
Deparment of Computer Science and Engineering, McKelvey School of Engineering, Washington University in Saint Louis, United States of America.
J Affect Disord. 2022 Jul 1;308:89-97. doi: 10.1016/j.jad.2022.04.015. Epub 2022 Apr 7.
Psychotherapy is a standard depression treatment; however, determining a patient's prognosis with therapy relies on clinical judgment that is subject to trial-and-error and provider variability.
To develop machine learning (ML) algorithms to predict depression remission for patients undergoing 6 months of problem-solving therapy (PST).
Using data from the treatment arm of 2 randomized trials, ML models were trained and validated on ENGAGE-2 (ClinicalTrials.gov, #NCT03841682) and tested on RAINBOW (ClinicalTrials.gov, #NCT02246413) for predictions at baseline and at 2-months. Primary outcome was depression remission using the Depression Symptom Checklist (SCL-20) score < 0.5 at 6 months. Predictor variables included baseline characteristics (sociodemographic, behavioral, clinical, psychosocial) and intervention engagement through 2-months.
Of the 26 candidate variables, 8 for baseline and 11 for 2-months were predictive of depression remission, and used to train the models. The best-performing model predicted remission with an accuracy significantly greater than chance in internal validation using the ENGAGE-2 cohort, at baseline [72.6% (SD = 3.6%), p < 0.0001] and at 2-months [72.3% (5.1%), p < 0.0001], and in external validation with the RAINBOW cohort at baseline [58.3% (0%), p < 0.0001] and at 2-months [62.3% (0%), p < 0.0001]. Model-agnostic explanations highlighted key predictors of depression remission at the cohort and patient levels, including female sex, lower self-reported sleep disturbance, lower sleep-related impairment, and lower negative problem orientation.
ML models using clinical and patient-reported data can predict depression remission for patients undergoing PST, affording opportunities for prospective identification of likely responders, and for developing personalized early treatment optimization along the patient care trajectory.
心理疗法是抑郁症的标准治疗方法；然而，确定患者的治疗预后依赖于临床判断，而这种判断存在试错和提供者差异。
开发机器学习（ML）算法来预测接受 6 个月问题解决治疗（PST）的患者的抑郁缓解情况。
使用来自 2 项随机试验治疗臂的数据，在 ENGAGE-2（ClinicalTrials.gov，#NCT03841682）上训练和验证 ML 模型，并在 RAINBOW（ClinicalTrials.gov，#NCT02246413）上进行测试，以在基线和 2 个月时进行预测。主要结果是使用抑郁症状清单（SCL-20）评分<0.5 在 6 个月时达到抑郁缓解。预测变量包括基线特征（社会人口学、行为、临床、心理社会）和 2 个月时的干预参与度。
在 26 个候选变量中，有 8 个基线变量和 11 个 2 个月变量可预测抑郁缓解，并用于训练模型。表现最佳的模型在使用 ENGAGE-2 队列进行内部验证时，在基线时[72.6%（SD=3.6%），p<0.0001]和在 2 个月时[72.3%（5.1%），p<0.0001]，以及在使用 RAINBOW 队列进行外部验证时，预测缓解的准确率明显高于机会水平，在基线时[58.3%（0%），p<0.0001]和在 2 个月时[62.3%（0%），p<0.0001]。模型不可知的解释突出了队列和患者水平上抑郁缓解的关键预测因素，包括女性、自我报告的睡眠障碍程度较低、睡眠相关损害程度较低以及消极问题取向程度较低。
使用临床和患者报告数据的 ML 模型可以预测接受 PST 的患者的抑郁缓解情况，为前瞻性识别可能的应答者提供了机会，并为沿着患者护理轨迹开发个性化的早期治疗优化提供了机会。
