Finding the best trade-off between performance and interpretability in predicting hospital length of stay using structured and unstructured data.

OBJECTIVE: This study aims to develop high-performing Machine Learning and Deep Learning models in predicting hospital length of stay (LOS) while enhancing interpretability. We compare performance and interpretability of models trained only on structured tabular data with models trained only on unstructured clinical text data, and on mixed data. METHODS: The structured data was used to train fourteen classical Machine Learning models including advanced ensemble trees, neural networks and k-nearest neighbors. The unstructured data was used to fine-tune a pre-trained Bio Clinical BERT Transformer Deep Learning model. The structured and unstructured data were then merged into a tabular dataset after vectorization of the clinical text and a dimensional reduction through Latent Dirichlet Allocation. The study used the free and publicly available Medical Information Mart for Intensive Care (MIMIC) III database, on the open AutoML Library AutoGluon. Performance is evaluated with respect to two types of random classifiers, used as baselines. RESULTS: The best model from structured data demonstrates high performance (ROC AUC = 0.944, PRC AUC = 0.655) with limited interpretability, where the most important predictors of prolonged LOS are the level of blood urea nitrogen and of platelets. The Transformer model displays a good but lower performance (ROC AUC = 0.842, PRC AUC = 0.375) with a richer array of interpretability by providing more specific in-hospital factors including procedures, conditions, and medical history. The best model trained on mixed data satisfies both a high level of performance (ROC AUC = 0.963, PRC AUC = 0.746) and a much larger scope in interpretability including pathologies of the intestine, the colon, and the blood; infectious diseases, respiratory problems, procedures involving sedation and intubation, and vascular surgery. CONCLUSIONS: Our results outperform most of the state-of-the-art models in LOS prediction both in terms of performance and of interpretability. Data fusion between structured and unstructured text data may significantly improve performance and interpretability.

Machine-learning prediction for hospital length of stay using a French medico-administrative database.

INTRODUCTION: Prolonged Hospital Length of Stay (PLOS) is an indicator of deteriorated efficiency in Quality of Care. One goal of public health management is to reduce PLOS by identifying its most relevant predictors. The objective of this study is to explore Machine Learning (ML) models that best predict PLOS. METHODS: Our dataset was collected from the French Medico-Administrative database (PMSI) as a retrospective cohort study of all discharges in the year 2015 from a large university hospital in France (APHM). The study outcomes were LOS transformed into a binary variable (long vs. short LOS) according to the 90th percentile (14 days). Logistic regression (LR), classification and regression trees (CART), random forest (RF), gradient boosting (GB) and neural networks (NN) were applied to the collected data. The predictive performance of the models was evaluated using the area under the ROC curve (AUC). RESULTS: Our analysis included 73,182 hospitalizations, of which 7,341 (10.0%) led to PLOS. The GB classifier was the most performant model with the highest AUC (0.810), superior to all the other models (all p-values <0.0001). The performance of the RF, GB and NN models (AUC ranged from 0.808 to 0.810) was superior to that of the LR model (AUC = 0.795); all p-values <0.0001. In contrast, LR was superior to CART (AUC = 0.786), p < 0.0001. The variable most predictive of the PLOS was the destination of the patient after hospitalization to other institutions. The typical clinical profile of these patients (17.5% of the sample) was the elderly patient, admitted in emergency, for a trauma, a neurological or a cardiovascular pathology, more often institutionalized, with more comorbidities notably mental health problems, dementia and hemiplegia. DISCUSSION: The integration of ML, particularly the GB algorithm, may be useful for health-care professionals and bed managers to better identify patients at risk of PLOS. These findings underscore the need to strengthen hospitals through targeted allocation to meet the needs of an aging population.

Machine-learning prediction of unplanned 30-day rehospitalization using the French hospital medico-administrative database.

Predicting unplanned rehospitalizations has traditionally employed logistic regression models. Machine learning (ML) methods have been introduced in health service research and may improve the prediction of health outcomes. The objective of this work was to develop a ML model to predict 30-day all-cause rehospitalizations based on the French hospital medico-administrative database.This was a retrospective cohort study of all discharges in the year 2015 from acute-care inpatient hospitalizations in a tertiary-care university center comprising 4 French hospitals. The study endpoint was unplanned 30-day all-cause rehospitalization. Logistic regression (LR), classification and regression trees (CART), random forest (RF), gradient boosting (GB), and neural networks (NN) were applied to the collected data. The predictive performance of the models was evaluated using the H-measure and the area under the ROC curve (AUC).Our analysis included 118,650 hospitalizations, of which 4127 (3.5%) led to rehospitalizations via emergency departments. The RF model was the most performant model according to the H-measure (0.29) and the AUC (0.79). The performances of the RF, GB and NN models (H-measures ranged from 0.18 to 0. 29, AUC ranged from 0.74 to 0.79) were better than those of the LR model (H-measure = 0.18, AUC = 0.74); all P values <.001. In contrast, LR was superior to CART (H-measure = 0.16, AUC = 0.70), P < .0001.The use of ML may be an alternative to regression models to predict health outcomes. The integration of ML, particularly the RF algorithm, in the prediction of unplanned rehospitalization may help health service providers target patients at high risk of rehospitalizations and propose effective interventions at the hospital level.