Prediction of post-traumatic stress disorder in family members of ICU patients: a machine learning approach

• Published in: Intensive care medicine Vol. 50; no. 1; pp. 114 - 124
• Main Authors: Dupont, Thibault, Kentish-Barnes, Nancy, Pochard, Frédéric, Duchesnay, Edouard, Azoulay, Elie
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.01.2024; Springer Nature B.V
• Subjects: Algorithms; Anesthesiology; Critical Care Medicine; Deep learning; Emergency Medicine; Intensive; Intensive care; Learning algorithms; Machine learning; Medicine; Medicine & Public Health; Mental disorders; Original; Pain Medicine; Patients; Pediatrics; Performance prediction; Pneumology/Respiratory System; Post traumatic stress disorder; Probability learning; Psychological stress; Sensitivity analysis; Supervised learning; Post-ICU burden; Algorithm; Mechanical ventilation; Machine learning; Prediction
• ISSN: 0342-4642; 1432-1238
• DOI: 10.1007/s00134-023-07288-1

Purpose Post-traumatic stress disorder (PTSD) can affect family members of patients admitted to the intensive care unit (ICU). Easily accessible patient’s and relative’s information may help develop accurate risk stratification tools to direct relatives at higher risk of PTSD toward appropriate management. Methods PTSD was measured 90 days after ICU discharge using validated instruments (Impact of Event Scale and Impact of Event Scale-Revised) in 2374 family members. Various supervised machine learning approaches were used to predict PTSD in family members and evaluated on an independent held-out test dataset. To better understand variables’ contributions to PTSD predicted probability, we used machine learning interpretability methods on the best predictive algorithm. Results Non-linear ensemble learning tree-based methods showed better predictive performances (Random Forest—area under curve, AUC = 0.73 [0.68–0.77] and XGBoost—AUC = 0.73 [0.69–0.78]) than regularized linear models, kernel-based models, or deep learning models. In the best performing algorithm, most important features that positively contributed to PTSD’s predicted probability were all non-modifiable factors, namely, lower patient’s age, longer duration of ICU stay, relative’s female sex, lower relative’s age, relative being a spouse/child, and patient’s death in ICU. A sensitivity analysis in bereaved relatives did not alter the algorithm’s predictive performance. Conclusion We propose a machine learning-based approach to predict PTSD in relatives of ICU patients at an individual level. In this model, PTSD is mostly influenced by non-modifiable factors.