Research Summary - 3
Profiling the Repeat Offender: Decoding Antimicrobial Treatment Trajectories in Dairy Cows
| Date/Time: |
8/29/2026
10:30
|
| Author: |
Xingfeiyang Liu |
| Clinic: |
Cornell University |
| City, State, ZIP: |
Ithaca, NY 14850 |
Xingfeiyang Liu, BS
1
;
Katherine J. Koebel, DVM
1
;
Michael Capel, DVM
2
;
Daryl Nydam, DVM, PhD
3
;
Renata Ivanek, DVM, MS, PhD
1
;
1Department of Population Medicine & Diagnostic Sciences, College of Veterinary Medicine, Cornell University, Ithaca NY
2Perry Veterinary Clinic, Perry NY
3Department of Public & Ecosystem Health, College of Veterinary Medicine, Cornell University, Ithaca NY
Introduction:
Despite the disproportionate labor demands, veterinary costs, and production losses associated with animals that experience repeated treatments, whether from the same or different ailments, the timing and spacing of recurrent treatment events of these “repeat offenders” remain poorly studied. The objective of this study was to analyze the timing, order, and frequency of multimorbidity to identify specific phenotypic trajectories.
Materials and methods:
A 1-year longitudinal dataset of antimicrobial treatment records was evaluated, comprising 4,226 dairy cattle with a total of 6108 events across n=5 conventional Holstein dairies of 1,000-5,000 head each. Health event metadata included the reason for treatment, the antimicrobial administered, age, and the animal’s fate (death or culling), if observed during the one-year observation period. Cattle were grouped by the total number of treatments received within 1 year. Agglomerative hierarchical clustering analysis, utilizing Gower’s distance and Ward's linkage, was performed to group animals based on chronological treatment sequence, age at treatment, and the median days between treatments. All analyses were performed in R.
Results:
Animals exhibiting multimorbidity received between 2-7 antibiotic treatments in a year. As the number of treatments per animal increased, progressively fewer animals exhibited that pattern, and those receiving multiple treatments within a year were more likely to have their fate recorded as culling or death. Youngstock experienced rapid-relapse cycles of under 20 days, driven by persistent pneumonia. In adults, transition failures (primarily reproductive diseases and systemic mastitis) caused the most rapid health declines, triggering secondary conditions like displaced abomasum and respiratory infections in as little as 5 days. Multimorbid mastitis trapped cows in monthly relapse cycles associated with culling. Conversely, intermittent lameness characterized the longest-surviving repeat offenders with initial treatment gaps exceeding 130 days. While this long gap may reflect delayed detection, the lameness itself frequently served as a precursor to subsequent mastitis.
Significance:
In this one-year longitudinal study, we find that multimorbidity in dairy cattle occurs in distinct patterns between calves and cows. Defining these treatment trajectories will provide producers with emerging clinical thresholds to determine when retreating a multimorbid animal is no longer economically viable. Future directions include survival analysis to mathematically forecast long-term culling risk and economic loss models to quantify the financial burden of each treatment pathway.
