A Comparison of Models Predicting One-Year Mortality at Time of Admission

• Published in: Journal of pain and symptom management 2022-03, Vol.63 (3), p.e287-e293
• Main Authors: Pierce, Robert P., Raithel, Seth, Brandt, Lea, Clary, Kevin W., Craig, Kevin
• Format: Article
• Language: English
• Subjects: advance care planning; Advance directives; clinical decision support; Data; Death & dying; End of life care; End of life decisions; Health risks; Hospice care; Hospital Mortality; Hospitalization; Humans; Logistic Models; Machine Learning; Medical prognosis; Mortality; Palliative care; random forest; Retrospective Studies; ROC Curve; Thresholds
• ISSN: 0885-3924; 1873-6513
• DOI: 10.1016/j.jpainsymman.2021.11.006

Hospitalization provides an opportunity to address end-of-life care (EoLC) preferences if patients at risk of death can be accurately identified while in the hospital. The modified Hospital One-Year Mortality Risk (mHOMR) uses demographic and admission data in a logistic regression algorithm to identify patients at risk of death one year from admission. This project sought to validate mHOMR and identify superior models. The mHOMR model was validated using historical data from an academic health system. Alternative logistic regression and random forest (RF) models were developed using the same variables. Receiver operating characteristic (ROC) and precision recall curves were developed, and sensitivity, specificity, and positive and negative predictive values were compared over a range of model thresholds. The RF model demonstrated higher area under the ROC curve (0.950, 95% CI 0.947 – 0.954) as compared to the logistic regression models (0.818 [95% CI 0.812 – 0.825] and 0.841 [95% CI 0.836 – 0.847]). Area under the precision recall curve was higher with the random forest model compared to the logistic regression models (0.863 vs. 0.458 and 0.494, respectively). Across a range of thresholds, the RF model demonstrated superior sensitivity, equivalent specificity, and higher positive and negative predictive values. A machine learning RF model, using common demographic and utilization data available on hospital admission, identified inpatients at risk of death more effectively than logistic regression models using the same variables. Machine learning models have promise for identifying admitted patients with elevated one-year mortality risk, increasing opportunities to prompt discussion of EoLC preferences.