Systematic Review of Sepsis-Associated Brain Dysfunction Risk Prediction Models

A systematic review of 12 studies containing 24 risk prediction models for sepsis-associated brain dysfunction, published on figshare in March 2026. The review, conducted by Wen Shen, analyzed models predicting sepsis-associated encephalopathy (SAE) or sepsis-associated delirium (SAD) from databases including CNKI, PubMed, and Web of Science up to April 2025. It extracted model characteristics using the CHARMS checklist and assessed methodological quality with the PROBAST tool.

Use Cases

• Benchmarking new sepsis-associated brain dysfunction models based on the review of 24 existing models.
• Assessing methodological quality in clinical prediction studies based on the PROBAST criteria applied in the review.
• Identifying sources of heterogeneity in model development based on the reported variations in outcome definitions and variable selection.
• Designing validation strategies for clinical models based on the reported use of internal and external validation in the included studies.

Strengths

• Includes 12 studies and 24 distinct risk prediction models for analysis.
• Methodological quality was assessed using the standardized PROBAST tool.
• Calibration was reported in 10 of the 12 included studies.

Limitations

• Row count and column-level documentation are absent; field semantics must be inferred after download.
• All 12 included studies were assessed as having a high risk of bias, predominantly in the analysis domain.
• The 3.2 MB file size suggests a small dataset, potentially limiting the scope of the underlying data.

Provenance

Source

figshare

Collection Method

Systematic literature review from nine databases including CNKI, PubMed, and Web of Science.

Time Range

Database inception to April 2025

Freshness

Last updated 2026-03-18 07:35:20