Equine syndromic surveillance in Colorado using veterinary laboratory testing order data.

Area of Science:

• Veterinary epidemiology
• Animal health surveillance
• Syndromic surveillance

Background:

• The Risk Identification Unit (RIU) of the Center for Epidemiology and Animal Health (CEAH) is enhancing national livestock health surveillance.
• Initiatives include expanding monitored species, health issues, and data sources.
• A new system uses weekly syndromic counts of laboratory test orders.

Purpose of the Study:

• To build a robust syndromic surveillance system for livestock using laboratory test data.
• To refine syndrome groups and data filtering to minimize alerting bias.
• To identify optimal statistical detection methods for specific laboratory data characteristics.

Main Methods:

• Analyzed 12 years of equine laboratory test records.
• Developed syndrome groups based on veterinary expertise and literature, refined through data analysis and consultation.
• Excluded regulatory, teaching hospital, and research tests.
• Evaluated various statistical alerting algorithms (e.g., C2, CuSUM, EWMA) using a testbed simulation.

Main Results:

• Established seven equine syndrome groups: abortion/reproductive, diarrhea/GI, necropsy, neurological, respiratory, systemic fungal, and tickborne.
• Data transformation and filtering significantly reduced test counts (e.g., >80% for diarrhea/GI).
• Derived optimal methods, parameters, and thresholds for each syndrome based on performance requirements.

Conclusions:

• Understanding laboratory data sources and workflow is critical for accurate syndrome surveillance.
• Syndrome group formation requires careful consideration of veterinary expertise and laboratory processes.
• Tailored statistical methods and parameters are essential for effective disease monitoring.