Emergency Department Triage & Patient Flow Data
Build and test patient-flow analytics without exposing patient information.
What you can build
- ED crowding and patient-flow dashboards
- ATS performance and waiting-time analysis
- Forecasting, segmentation and machine-learning experiments
- SQL, Python, Power BI and Tableau training
- Data-pipeline and clinical-software QA
Designed analytical signals
The generator models interpretable relationships between arrival mode, urgency, occupancy, waiting time, length of stay and disposition. Higher occupancy produces longer modelled waits and stays. Ambulance and police arrivals skew toward more urgent ATS categories, while bed block is concentrated in admission pathways.
Schema preview
| Field | Meaning |
|---|---|
| ats_triage_category | Australasian Triage Scale category 1–5 |
| department_occupancy_pct_at_arrival | Modelled ED occupancy at presentation |
| time_to_clinician_minutes | Minutes from arrival to clinician |
| length_of_stay_minutes | Total modelled ED stay |
| disposition | Discharge, admission or other outcome |
| bed_block_flag | Whether admitted flow was delayed |
Files supplied
Complete CSV, data dictionary, README, licence, methodology and limitations, summary statistics, SQL practice queries, Python analysis notebook, offline dashboard and case study in the supported package.
Responsible-use statement
Every patient, clinician and facility is fictional. The dataset contains no personal or health information from real people and is not clinically validated. It is designed for research, education, analytics, AI/ML development and software testing—not patient care or real-facility benchmarking.