North American Stroke Patient Data for Triage Score Evaluation

A two-state cohort study evaluates the Japan Urgent Stroke Triage (JUST) score for differentiating ischemic from hemorrhagic strokes in a North American population. The dataset contains clinical variables from adult patients transported by emergency medical services. Daria S. Nicke authored this study, which was last updated in April 2026.

Use Cases

• Predict hemorrhagic stroke using systolic blood pressure ≥165mmHg, altered mental status, headache, and nausea/vomiting as features.
• Compare model performance (AUC) between the JUST-7 score variables and an expanded model including sex, numbness, comorbidities, and AHA guidelines.
• Analyze odds ratios for neurological symptoms like altered mental status (OR 1.88) and headache (OR 1.82) associated with hemorrhagic stroke diagnosis.
• Validate the predictive power of a logistic regression model for stroke type differentiation using receiver operating characteristic (ROC) curve analysis.

Strengths

• Clinical variables are explicitly defined, including systolic blood pressure threshold, altered mental status, and nausea/vomiting.
• Model performance metrics like Area Under the Curve (AUC) values (0.68 to 0.72) and odds ratios with confidence intervals are provided.

Limitations

• The dataset is very small at 12.3 KB, indicating limited raw patient records or summary statistics only.
• Geographic scope is limited to a two-state cohort in North America, which may not be generalizable.
• Specific row count and individual patient-level data are unknown, restricting detailed analysis.

Provenance

Source

figshare, authored by Daria S. Nicke.

Collection Method

Data from a two-state cohort of adult acute stroke patients transported by emergency medical services (EMS).

Time Range

null

Freshness

null

Geography

North American population, specifically a two-state cohort.