Ranking hospitals' burn care capacity using cluster analysis on open government data.

Area of Science:

• Healthcare Management
• Emergency Medicine
• Data Science

Background:

• National emergency response plans (ERPs) for burn mass casualty incidents (BMCIs) are vital for effective patient management.
• A standardized hospital burn care capacity ranking model is essential for optimizing ERPs.
• Taiwan lacked a specific model for ranking hospital burn care capacity.

Purpose of the Study:

• To develop a novel ranking model for classifying hospital burn care capacity in Taiwan.
• To support the creation of a national BMCI ERP by providing essential data for patient distribution.
• To establish a framework for future healthcare capacity assessments.

Main Methods:

• Utilized open government data from 535 hospitals, focusing on 42 selected variables.
• Employed hierarchical cluster analysis (Ward's method) and tree-based model analysis.
• Incorporated expert panel consultations and both internal and external validation methods.

Main Results:

• Identified four distinct clusters representing different levels of burn care capacity.
• The refined level of emergency care responsibility hospital was the most significant factor in hospital classification.
• Cluster analysis results were validated through Kruskal-Wallis tests and external questionnaire responses.

Conclusions:

• Open government data and cluster analysis effectively created a hospital burn care capacity ranking model for Taiwan.
• The developed model is suitable for informing BMCI ERP development.
• This methodology can serve as a template for ranking healthcare capacity or quality in other fields using open data.