Fatality rate drops again; at lowest level since data has ben collected
The rate of fatal injury in 2021 on all surfaces at North American racetracks declined to its lowest number since the industry began collecting data on the rates in 2009, according to statistics released by The Jockey Club on Tuesday.
Though the 2021 rate of 1.39 was only a slight decline from the 2020 rate of 1.41, the rate last year was a 30.5 percent decline from the 2.00 rate in 2009. The rate represents the number of fatalities per 1,000 starts. In addition, the overall rate has declined 17.3 percent since 2018, according to the data.
Thoroughbred fatalities are one of the most pressing issues facing the sport in an era when concerns over animal welfare, especially in industries in which animals are used for entertainment, have become more broadly and intensely felt in countries where racing is conducted. In the U.S., those concerns led to an existential crisis in 2019, when a spate of fatal injuries at Santa Anita Park in Southern California garnered widespread media attention.
The U.S. racing industry responded to the crisis with a renewed emphasis on safety measures and an increased scrutiny of medication use and veterinary practices. Most racing officials involved in the effort believe the measures adopted by the industry have led to a decline in the fatality rate, though quantifying the impact is complicated.
“The downward trends in the [Equine Injury Database] data are very encouraging,” said James Gagliano, the president and chief operating officer of The Jockey Club, which maintains the database.
By surface, the fatality rate on artificial surfaces, such as those in use at Golden Gate Fields, Presque Isle Downs, and Turfway Park, were the lowest, at 0.73 fatalities per 1,000 starts. The rate for dirt races in 2021 was 1.51, while the rate for turf races was 1.25.
Overall since 2009, the fatality rate for artificial surfaces has declined 51 percent, and the 0.73 rate last year was the lowest rate on record for any surface. The rate on dirt surfaces has declined 28.1 percent, and the rate for turf has declined 35.6 percent.
Last year’s data was marked by an unusual spike in the fatality rate for 2-year-olds, a category of racing that is singled out for criticism by many opponents of the sport. The 2020 rate was 1.69 per 1,000 starts, well outside the statistical norm for the category since 2009, but in 2021, the rate fell to 0.98 per 1,000 starts, the lowest on record.
Some equine physiology experts have speculated that the disruptions caused by the pandemic in 2020 may have contributed to the increase in the overall fatality rate for 2-year-olds due to uncharacteristic gaps in their training schedules. Though excessive training can lead to bone damage, regular gallops and training routines build stronger bone mass, especially in young horses.
The fatality rate for 3-year-old horses in 2021 was 1.52 per 1,000 starts, statistically even with the rate in 2020 of 1.57.
“The trends we discovered in the 2020 data show evidence that interrupting a 2-year-old’s prep year may have a detrimental effect, but it does not carry over to the 3-year-old year,” said Dr. Tim Parkin, a veterinary epidemiologist at the University of Glasgow who has analyzed the data in the EID since its inception. “We will continue to investigate the data to further help prevent injuries and make the sport safer.”
By distance, the fatality rate has consistently been higher for races shorter than six furlongs, but last year, the rate for so-called “middle distance” races – six to eight furlongs – was the highest, at 1.46, compared to 1.35 in the shortest races. Racetracks are increasingly holding sprints on turf, which have a lower fatality rate than dirt races.
Races longer than eight furlongs continued to show the lowest fatality rate by distance, at 1.19.
The EID registers fatalities for any horse that dies within 72 hours of a race. Although participation in the project is voluntary, The Jockey Club said that “approximately 99 percent of all starts” in 2021 were made at tracks that log their data with the EID.
Participating racetracks also have the option to allow their fatality data to be publicly available. Among those tracks that have elected to divulge the data, the fatality rate was 1.15, compared to 1.54 for the tracks that do not publish.
In addition, those tracks that are accredited by the National Thoroughbred Racing Association had a much lower fatality rate, 1.24, then those tracks that are not accredited. Of the tracks participating in the EID, 22 are accredited, while 58 are not, according to The Jockey Club.

