Acute Physiology Score and Sepsis in ICU Non-Surviving IBD Patients

A dataset of 190 adult patients with inflammatory bowel disease who died during ICU admission, derived from the eICU-CRD database. The data was used to analyze the association between Acute Physiology Score (APS) and sepsis, employing weighted logistic regression and XGBoost. The dataset was created by Chenyang Li and last updated on 2026-06-03.

Use Cases

• Modeling sepsis risk based on Acute Physiology Score and other clinical variables.
• Analyzing the correlation between APS and biomarkers like anion gap, lactate, and bicarbonate.
• Investigating subgroup differences in sepsis prediction for elderly and male ICU patients.
• Training machine learning models for mortality risk stratification in ICU settings.

Strengths

• Dataset includes 190 patient records with a defined outcome (sepsis vs. non-sepsis).
• Analysis includes specific statistical results, such as an OR of 4.856 for the highest APS quartile.
• The dataset is associated with a peer-reviewed study methodology, including XGBoost and SHAP analysis.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The study sample is small (190 patients) and focused exclusively on non-survivors, limiting generalizability.

Provenance

Source

eICU-CRD database

Collection Method

Retrospective cohort study

Freshness

Last updated 2026-06-03 04:48:51; freshness should be verified.