Skip to main navigation Skip to search Skip to main content

Incorporating Social Determinants of Health in Predicting Adverse Sepsis Outcomes

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Predictive modeling has become a valuable tool for identifying patients at high risk of adverse health outcomes. Advances in big data have further increased interest in Machine Learning (ML) techniques in healthcare, which provide flexible and robust predictive abilities using large datasets from electronic health records (EHRs). Recent research has focused on exploring the integration of community-level and individual-level social determinants of health (SDOH) into predictive models to improve health outcomes and resource use. Most sepsis prediction models have focused on traditional predictors like age, comorbidities, and illness severity. This study addresses this gap by developing ML models that integrate SDOH data from 3,230 sepsis admissions (May 2022-April 2024) at a regional medical system in Louisiana to predict in-hospital mortality, 30-day readmission, and length of stay (LOS). ML models, including Deep Neural Networks (DNNs), Random Forest (RF), and Logistic Regression (LR), were used for classification, while DNNs, RF, and XGBoost were used for regression. Correlation analysis was performed to check for multicollinearity. Data preprocessing included one-hot encoding of categorical variables, splitting the dataset into training and testing sets, scaling numerical variables, and oversampling the training set for classification models. After model training, classification performance was evaluated using accuracy, precision, recall, and area under the curve (AUC), whereas regression performance was assessed using mean squared error (MSE), root mean squared error (RMSE), mean absolute error (MAE), and R-squared. Such predictive models have potential applicability in clinical decision support systems to identify high-risk sepsis patients and guide timely, targeted interventions.

Original languageEnglish (US)
Title of host publicationIISE Annual Conference and Expo 2026
EditorsY. Xiang, D. Yu, R. Thiesing
PublisherInstitute of Industrial and Systems Engineers, IISE
Pages236-241
Number of pages6
ISBN (Electronic)9798331339197
DOIs
StatePublished - 2026
EventIISE Annual Conference and Expo 2026 - Arlington, United States
Duration: May 16 2026 → May 19 2026

Publication series

NameIISE Annual Conference and Expo 2026
Volume1

Conference

ConferenceIISE Annual Conference and Expo 2026
Country/TerritoryUnited States
CityArlington
Period5/16/26 → 5/19/26

Bibliographical note

Publisher Copyright:
© (2026) by Institute of Industrial & Systems Engineers (IISE) All rights reserved.

Keywords

  • health outcomes
  • machine learning
  • sepsis
  • Social determinants of health

Fingerprint

Dive into the research topics of 'Incorporating Social Determinants of Health in Predicting Adverse Sepsis Outcomes'. Together they form a unique fingerprint.

Cite this