Developing and Validating a Prediction Model For Death or Critical Illness in Hospitalized Adults, an Opportunity for Human-Computer Collaboration

Abstract

Hospital early warning systems that use machine learning (ML) to predict clinical deterioration are increasingly being used to aid clinical decision-making. However, it is not known how ML predictions complement physician and nurse judgment. Our objective was to train and validate a ML model to predict patient deterioration and compare model predictions with real-world physician and nurse predictions.

Design: Retrospective and prospective cohort study.

Setting: Academic tertiary care hospital.

Patients: Adult general internal medicine hospitalizations.

Measurements and main results: We developed and validated a neural network model to predict in-hospital death and ICU admission in 23,528 hospitalizations between April 2011 and April 2019. We then compared model predictions with 3,374 prospectively collected predictions from nurses, residents, and attending physicians about their own patients in 960 hospitalizations between April 30, and August 28, 2019. ML model predictions achieved clinician-level accuracy for predicting ICU admission or death (ML median F1 score 0.32 [interquartile range (IQR) 0.30-0.34], AUC 0.77 [IQ 0.76-0.78]; clinicians median F1-score 0.33 [IQR 0.30-0.35], AUC 0.64 [IQR 0.63-0.66]). ML predictions were more accurate than clinicians for ICU admission. Of all ICU admissions and deaths, 36% occurred in hospitalizations where the model and clinicians disagreed. Combining human and model predictions detected 49% of clinical deterioration events, improving sensitivity by 16% compared with clinicians alone and 24% compared with the model alone while maintaining a positive predictive value of 33%, thus keeping false alarms at a clinically acceptable level.

Conclusions: ML models can complement clinician judgment to predict clinical deterioration in hospital. These findings demonstrate important opportunities for human-computer collaboration to improve prognostication and personalized medicine in hospital.

Keywords: artificial intelligence; clinical prediction; machine learning; mortality; prognosis.

Figure 1.

Accuracy of clinicians and the machine learning (ML) model in predicting ICU and death across 500 bootstrapped samples. Boxplots depict the distribution (horizontal bar denotes median, box is 25/75 percentile) of the F1-score of the ML model and clinicians across 500 bootstrapped samples. Different bootstrapped samples were used for each clinician type, resulting in slightly different estimates of ML model performance in each comparison. The ML model demonstrates superior F1-scores in predicting ICU admission, but the model was not more accurate than clinicians in predicting death. Combining model and clinician predictions leads to improvements in F1-scores, which is particularly important given that the value of identifying true positive cases outweighs identification of true negatives in an early warning system.