Pulled-up race outcomes as welfare-relevant surveillance indicators in United States Steeplechase horses, 2023-2025.

Overview

• This study analyzed pulled-up race outcomes in United States steeplechase horses from 2023 to 2025 to determine whether being pulled up during a race is associated with disruptions in future participation.
• The research found that horses pulled up during races were more likely to have extended breaks, including missing future starts, suggesting pulled-up outcomes might be useful as indicators for welfare monitoring.

Background and Purpose

• Pulled-up outcomes occur when a horse stops or is stopped during a race before finishing; these are routinely recorded but not specific to any one cause.
• Potential reasons for being pulled up include fatigue, poor performance, unsafe footing, rider’s decision, suspected injury, or physiological compromise.
• The study aimed to determine whether these pulled-up outcomes correlate with future disruptions in racing participation, which may indicate underlying welfare issues.
• Understanding this association could help use pulled-up race outcomes as surveillance indicators for horse welfare in steeplechase racing.

Methods

• Data Source: Retrospective analysis of National Steeplechase Association Race Results Reports from 2023 through 2025 at the horse-start level.
• Sample: 3,070 race starts by 710 different horses.
• Key outcome measures included:
  • Time to the next recorded race start after a pulled-up event.
  • Binary indicators for extended breaks defined as ≥60 days interruption and ≥183 days absence.
  • Whether the horse had any future recorded race start after the pulled-up event.
• Statistical Analyses:
  • Cox proportional hazards models stratified by race type and accounting for clustering by horse, to analyze time-to-event data.
  • Generalized estimating equations used for binary outcomes with horse clustering.
  • Sensitivity analyses excluded November starts since steeplechase races generally occur between March and November, to check seasonal effects.

Results

• Out of all starts, 367 (11.95%) were pulled up, involving 277 horses (39% of horses had at least one pulled-up event).
• After a pulled-up start:
  • 37.6% of those horses never had a later recorded race start.
  • Among horses who returned to racing, the median time to next start was 42 days compared to 35 days after non-pulled-up starts.
  • Considering all starts, the difference in median time extended to 148 days (pulled-up) versus 45 days (non-pulled-up), indicating longer disruptions.
• Statistical associations found:
  • Pulled-up status related to a significantly lower rate of returning to racing (Hazard Ratio 0.589, meaning about 41% reduced hazard of return compared to non-pulled-up horses).
  • Odds of experiencing a ≥60-day interruption were 2.10 times higher after being pulled up.
  • Odds of experiencing a ≥183-day absence were 2.36 times higher after being pulled up.
  • These associations persisted even when excluding November races, suggesting seasonality did not drive the findings.

Interpretation

• Being pulled up during a steeplechase race strongly correlates with disruptions in continued racing participation.
• Because pulled-up outcomes are nonspecific, the exact reasons for these disruptions are unknown and could include injury or other welfare concerns.
• Despite their limitations, pulled-up race outcomes can serve as useful indicators to prioritize horses needing veterinary follow-up or more detailed clinical evaluation.
• However, they should not replace direct veterinary examinations or clinical data for diagnosing or managing health or performance problems.

Potential Implications

• Race organizers and steeplechase associations could use pulled-up outcomes as part of welfare surveillance systems to monitor horse health trends longitudinally.
• These indicators could help target resources for veterinary care or rest periods for horses at risk of injury or prolonged absence due to underlying issues.
• Future research could focus on linking pulled-up events with clinical findings to improve specificity of welfare monitoring.