Comparison of two major emergency department-based free-text chief-complaint coding systems

Area of Science:

• Public Health
• Epidemiology
• Emergency Medicine

Background:

• Emergency departments (EDs) utilize free-text chief complaints for syndromic surveillance.
• Varied descriptions and coding of chief complaints present a challenge for accurate data interpretation.

Purpose of the Study:

• To compare the agreement between two major ED-based free-text chief-complaint coding systems.
• To evaluate the consistency of syndrome coding from free-text chief complaints.

Main Methods:

• Compared the New York City Department of Health and Mental Hygiene (DOHMH) system with the Real-Time Outbreak Detection System Complaint Coder (CoCo).
• Processed chief-complaint data from 21,736 patients at an urban ED.
• Collapsed DOHMH syndromes to match CoCo categories for direct comparison.

Main Results:

• Achieved optimal overall agreement between DOHMH and CoCo syndrome coding (0.614 kappa).
• Individual syndrome agreement varied, with Rash (0.711 kappa) and Respiratory (0.594 kappa) showing the highest consistency.
• Gastrointestinal and Unknown syndromes demonstrated the lowest agreement (0.270 and 0.002 kappa, respectively).

Conclusions:

• Optimal overall agreement exists between the evaluated chief-complaint coding systems.
• Substantial differences in classification schemes highlight the need for a standardized approach.
• A consensus on chief-complaint classification is crucial for effective syndromic surveillance.