Race and course associated risk factors for fatality in thoroughbreds

Abstract

The thoroughbred racing industry is one of ever increasing costs, euthanasia of a horse amounts to a significant loss of expenditure for owners and trainers, thus identifying factors which may contribute to fatal injuries is an area of major interest within the industry. The objective of this paper is to identify external causes of catastrophic limb injuries requiring euthanasia or resulting in death at the time of a race. External factors considered include among others, going, field size, number of hurdles, race number, number of fallers and distance. Internal factors are factors relating to a horse as an individual such as age, sex, previous runs, and previous injury. Basic fatality numbers compiled by the Irish Turf Club from 2015 were used and further investigated for the purposes of this paper. The data compiled formed a case control study with a total of 78 cases, 387 controls. Race courses across the country showed a large range of fatality rates per 1000 starts, varying between 0.35-11.35 per 1000 starts. Race type showed some association with fatality rates, flat racing had the lowest estimate coefficient at -3.66 (p=<0.0001, 0.92 fatalities per 1000 starts), hurdle compared to flat had a coefficient estimate of 0.91 (P=0.0279, 3.79 fatalities per 1000 starts), and steeple racing showed the largest fatality rate with a coefficient estimate of 1.61 (P=0.0002, 7.68 fatalities per 1000 starts). Number of runners per race returned results which also showed association with an increasing fatality rate, from 2 until 30 runners respectively the probabilities increase from 0.03-0.27 for flat racing, 0.07-0.48 for hurdle racing and 0.13-0.65 for steeple racing. Observational analysis of the circumstances surrounding the injury of case horses concluded that 84% of horses in steeplechase racing and 44% of horses in hurdle racing had collided to some extent with a fence or hurdle leading to the onset of symptoms of their injury. The number of falling horses per race demonstrated a positive association with fatality rates in each race type. Steeplechase races show the largest divergence with the coefficient estimates ranging from -2.47 for no fallers (P=<0.0001), to one faller with a coefficient estimate of 1.55 (P=0.0072), and two fallers or more with a coefficient estimate of 2.80 (P=<0.0001), the probabilities showed marked increases also from 0.08, 0.29 0.58 respectively. Hurdle showed significant trends also no faller had a coefficient estimate of -1.94 (P=<0.0001), one faller coefficient estimate of 0.92 (P=0.0249) and two or more fallers had a coefficient estimate of 1.38 (P=0.0398), the probabilities per category were 0.13, 0.27, and 0.36 respectively. In conclusion the rate of fatality per race course may vary on its geographical location, on the type of races run on the race course, on the ground surface type, on the ground moisture content and on its physical properties including corners and slope. Falling horses showed significant association with increasing fatality rates, factors which could be addressed to prevent this are variables such as amount of runners, ground, fence type and as previously mentioned the track layout. Further consideration from racecourses and racing regulatory bodies into racecourse design on tracks or race types with significantly higher fatality rates than others may be required to prevent such divergent fatality rates as mentioned above.