Assessment and Prediction of Clinical Outcomes for ICU-Admitted Patients Diagnosed with Hepatitis: Integrating Sociodemographic and Comorbidity Data
Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Background
Hepatitis, a disease characterized by inflammation of the liver, is a leading global health challenge that contributes to over 1.3 million deaths annually, with hepatitis B and C accounting for many of these fatalities. Intensive care unit (ICU) management of patients is particularly challenging due to the complex clinical care and resource demands. Despite advancements in ICU predictive analytics, limited research has specifically addressed hepatitis patients, creating a gap in optimizing care for this population.
Methods
This study focuses on predicting length of stay (LoS) and discharge outcomes and discharge location for ICU-admitted hepatitis patients using machine learning (ML) models. Leveraging data from the MIMIC-IV database, which includes around 94,500 ICU patient records, this study uses sociodemographic details, clinical characteristics, and resource utilization metrics to develop predictive models such as Random Forest, Logistic Regression, Gradient Boosting Machines, and Generalized Additive Model with Negative Binomial Regression.
Results
The ML models identified medications, procedures, comorbidities, age, and race as key predictors. Total LoS emerged as an important factor in predicting discharge outcomes and location.
Conclusion
This study demonstrates the value of machine learning models for predicting clinical outcomes for hepatitis patients, including length of stay and discharge status. The results underscore the influence of factors like race and age, revealing disparities that must be addressed in predictive care strategies. While the models show promise, challenges such as variability in prolonged stays and limited multi-class prediction accuracy point to the need for ongoing refinement and research.
What is already known on this topic?
Hepatitis B and C cause significant global mortality and often lead to critical illness requiring ICU care. ICU management for hepatitis patients is complex, and prolonged ICU stays are associated with increased costs and higher mortality risks. Although machine learning has been applied in ICU settings, few studies have focused on predictive modeling specifically for hepatitis patients.
What this study adds?
This study applies machine learning models to predict ICU length of stay, discharge outcomes, and discharge locations among hepatitis patients using a large, real-world dataset. It identifies key clinical and sociodemographic predictors, such as ICU medications, procedures, comorbidities, age, and race. The findings reveal that race is a consistent predictor across all models, highlighting underlying disparities in critical care outcomes.
How might this study affect research, practice or policy?
These predictive models can support more informed clinical decision-making, enhance ICU resource planning, and improve patient outcomes by identifying high-risk individuals early. The study also emphasize the need to address racial disparities in critical care outcomes through targeted research and policy changes.
