Related Experiment Video
Updated: Oct 5, 2025
07:42
A Data-Driven Approach to Quantifying Immune States in Sepsis
Published on: February 7, 2025
330
Risk-standardized sepsis mortality map of the United States
Jiun-Ruey Hu1, Chia-Hung Yo2, Hsin-Ying Lee3
1Department of Internal Medicine, Yale School of Medicine, USA.
Digital Health
|January 31, 2022
Summary
Sepsis mortality rates vary significantly across US states. This study maps these disparities, highlighting areas needing improved sepsis care to reduce in-hospital deaths.
Area of Science:
- Public Health
- Healthcare Quality
- Medical Informatics
Background:
- Sepsis is a leading cause of in-hospital mortality in the US.
- Accurate sepsis mortality estimates are crucial for quality improvement initiatives.
- Risk-standardized mortality rates (RSMRs) are established for other conditions but not well-characterized for sepsis.
Purpose of the Study:
- To construct a US sepsis RSMR map.
- To identify and illustrate geographic disparities in sepsis care and outcomes.
Main Methods:
- Cross-sectional study of adult sepsis hospitalizations from the US Nationwide Inpatient Sample (2010-2011).
- Hospital-level RSMRs calculated using hierarchical logistic modeling.
- Risk adjustment employed Sepsis Risk Prediction Score and gradient-boosted decision trees (machine learning).
Main Results:
- Analysis included 1,739,033 adult sepsis hospitalizations.
- The median sepsis RSMR was 18.4% (interquartile range: 17.0-21.0) using the machine learning model.
- Highest RSMRs were in Wyoming, North Dakota, and Mississippi; lowest in Arizona, Colorado, and Michigan.
Conclusions:
- Significant state-level variation in sepsis RSMRs indicates opportunities for care improvement.
- This study presents the first map of state-level variation in sepsis RSMRs.
