Related Experiment Video
Updated: Feb 19, 2026
05:28
Evaluation of a Reliable Biomarker in a Cecal Ligation and Puncture-Induced Mouse Model of Sepsis
Published on: December 9, 2022
4.5K
Risk prediction models for sepsis-associated encephalopathy: a systematic evaluation and meta-analysis
Ting Ting He1, Tuo Quan Jiao1, Xue Mei An2,3
1School of Nursing, Chengdu University of Traditional Chinese Medicine, Chengdu, China.
Peerj
|February 18, 2026
Summary
Risk prediction models for sepsis-associated encephalopathy (SAE) show fair discrimination but have a high risk of bias. Further research is needed to improve their clinical applicability.
Area of Science:
- Critical care medicine
- Neurology
- Epidemiology
Background:
- The development of risk prediction models for sepsis-associated encephalopathy (SAE) is growing.
- The clinical utility and quality of these models are currently uncertain.
Purpose of the Study:
- To systematically review and assess the quality of published studies on SAE risk prediction models.
- To evaluate the risk of bias and applicability of existing SAE prediction models.
Main Methods:
- A comprehensive systematic search of multiple databases (PubMed, Embase, etc.) was performed.
- Data extraction and risk of bias assessment using the Prediction model Risk Of Bias Assessment Tool (PROBAST) were conducted by two independent reviewers.
- Observational studies focusing on SAE risk prediction models were included.
Main Results:
- Ten studies involving 1,994 participants were included, with SAE incidence ranging from 15.16% to 63.3%.
- Age and Sequential Organ Failure Assessment (SOFA) score were the most common predictors, showing significant association with SAE.
- The pooled area under the receiver operating characteristic curve (AUC) for validated models was 0.83, indicating fair discrimination, but all studies exhibited a high risk of bias.
Conclusions:
- While SAE prediction models demonstrate fair discriminative ability, all reviewed studies were found to have a high risk of bias.
- Limitations in data sources and analysis contributed to the high risk of bias, questioning the current clinical applicability of these models.
