Developing a diagnosis-based severity classification system for use in emergency medical services for children

Area of Science:

• Health Services Research
• Emergency Medicine
• Medical Informatics

Background:

• Adequate risk adjustment methodologies are crucial for advancing emergency medicine health services research.
• Current methods may not fully capture patient severity in emergency departments (EDs).

Purpose of the Study:

• To develop and validate a consensus-derived, diagnosis-based severity classification system (SCS) for ED visits.
• To assess the association between the SCS and ED resource utilization.
• To establish the SCS as a potential risk adjustment tool for health services research.

Main Methods:

• A panel of experts assigned severity scores (1-5) to 3,041 International Classification of Diseases (ICD), 9th revision, diagnosis codes.
• SCS scores were determined by the highest severity diagnosis code for each ED visit.
• The association between SCS scores and ED resource use was evaluated across three datasets: PCDP, NHAMCS, and Connecticut state ED data.

Main Results:

• A significant association was found between the five-level SCS and all six measures of ED resource use.
• These measures included triage category, disposition, ED resource use, CPT E&M codes, ED length of stay, and ED charges.
• The associations were consistent across the three evaluated datasets.

Conclusions:

• The SCS is a valid tool, demonstrating a strong correlation with actual ED resource utilization.
• The system's reliance on easily accessible ICD-9 diagnosis codes makes it practical for health services research.
• The SCS can serve as a valuable risk adjustment tool in emergency medicine research.