Data Sheet 1_Development and internal validation of a clinical prediction model for septic

Junyu Dong's research, last updated in May 2026, presents a clinical prediction model for septic shock in children with RSV bronchiolitis admitted to the PICU. The model was developed and internally validated using a retrospective cohort of 224 pediatric patients. It identifies four independent predictors: fungal co-infection, admission glucose, antithrombin III, and interleukin-6.

Use Cases

• Validating a septic shock prediction model based on the identified biomarkers (fungal co-infection, glucose, antithrombin III, interleukin-6).
• Benchmarking new pediatric risk stratification tools against the reported model performance metrics (AUC 0.892).
• Investigating the association between fungal co-infection and severe outcomes in pediatric respiratory infections.
• Developing a clinical decision support tool based on the web calculator mentioned in the description.

Strengths

• Model performance was rigorously evaluated using 5-fold cross-validation and bootstrap resampling.
• The final model demonstrated excellent discrimination with an AUC of 0.892 and good calibration metrics.
• The dataset includes 224 pediatric patient records from a PICU setting.

Limitations

• The sample size is modest (224 patients), and the authors note external validation in larger cohorts is essential.
• Column-level documentation is absent; field semantics must be inferred from the research paper.
• Row count for the underlying data is unknown, which may limit suitability assessment.

Provenance

Source

figshare, author Junyu Dong

Collection Method

Retrospective cohort study of pediatric patients admitted to PICU with RSV bronchiolitis.

Freshness

Last updated 2026-05-15 05:48:07; freshness should be verified.