Reusing health records from farm animal practices at scale: A potential complementary method of surveillance.

Farm animal EHRs combined with text mining enable real-time disease and antimicrobial surveillance at scale.

This pilot study evaluated the utility of electronic health records (EHRs) from four Welsh farm animal veterinary practices as a surveillance tool for disease and antimicrobial use patterns. Over one year (February 2024–January 2025), 32,799 records were collected and analyzed using text mining and topic modeling techniques. Antimicrobial prescription rates were substantial: 32.6% in cattle and 63.8% in sheep records. Tetracyclines, macrolides, penicillins, and penicillin–aminoglycoside combinations were most frequently prescribed in both species. Notably, no category A antimicrobials were recorded, and category B antimicrobials represented only 0.12% and 0.04% of cattle and sheep records respectively. Text mining successfully identified key disease syndromes including mastitis, joint ill, lameness, and pneumonia. A limitation identified was that some records described multiple animals with different diagnoses, complicating treatment-to-syndrome attribution. The study demonstrates that EHRs at scale and in real-time offer a complementary surveillance approach for farm animal disease monitoring, with artificial intelligence and text mining potentially providing efficient, novel insights into antimicrobial stewardship patterns on farms.

This summary was distilled by AI and may occasionally misinterpret data. Confirm critical details with the primary literature before clinical application.