Wearable sensor to predict catastrophic injury risk in racehorses

(c) Donald Blais Dreamstime.com

Catastrophic musculoskeletal injury (CMI) remains one of the most serious welfare challenges in Thoroughbred racing. Although safety has improved in recent years, fatal limb injuries still occur on raceday and during training, often with little visible warning. A new study led by researchers at Washington State University (WSU) suggests that wearable sensor technology, combined with advanced data analysis, could help identify horses at highest risk before an injury occurs.

The device involved in the study is a small, lightweight sensor worn in the horse’s saddle cloth that functions as an inertial measurement unit (IMU), combined with global positioning system (GPS) technology. It measures acceleration and movement in multiple directions and records approximately 2,400 data points per second. This allows the system to capture minute details of stride mechanics - such as symmetry, impact force, and limb movement patterns - that are far too subtle and too fast to be detected by the human eye, especially at racing speeds.

An algorithm processes the stride data and compares it to patterns collected from thousands of sound, high-performing horses and from those that later suffered fatal musculoskeletal injuries. Based on this comparison, each horse is assigned a risk score from 1 to 6, where a score of 1 represents the lowest risk and a score of 6 indicates the highest risk of future fatal injury.

The researchers analysed stride data from 28,481 races involving 11,834 Thoroughbreds at 10 racetracks across the United States. These data were collected between July 2021 and May 2024. During this time, 74 horses were fatally injured. Statistical analysis revealed a relationship between risk score and the likelihood of injury. Horses that received the highest score of 6 represented only 0.4% of total race starts, yet they accounted for 4% of all fatalities. When compared to horses with a score of 1, horses assigned a 6 were 44 times more likely (44.6 times, to be precise) to suffer a fatal musculoskeletal injury.

The research team conducted logistic regression analyses to explore whether other factors affected injury risk. Interestingly, age was not found to be a significant predictor, meaning that older horses were not necessarily more at risk than younger ones. However, male horses (stallions and geldings) were more likely to suffer fatal injuries than females. Race distance also played a role; horses competing in shorter races were at greater risk, possibly due to higher speeds and greater peak forces on the limbs. In addition, fatal injuries occurred more frequently on dirt and turf tracks than on synthetic all-weather surfaces.

These findings align with previous research showing that surface type and speed can influence limb loading and injury rates. They also reinforce the idea that many racing injuries are not sudden or random events. Necropsy studies have previously shown that as many as 93% of horses that suffered catastrophic injury had evidence of a pre-existing bone abnormality, such as microfractures or maladaptive bone remodelling. This new sensor technology provides a way to detect the biomechanical consequences of those underlying problems before they reach a critical point.

Dr Warwick Bayly, lead researcher and professor in WSU’s College of Veterinary Medicine, believes this technology could be used as a preventative screening tool. Horses identified as high-risk could undergo detailed veterinary examination, including lameness assessments and advanced diagnostic imaging (such as MRI or CT scanning if appropriate). Trainers and veterinarians could then adjust training intensity, allow additional rest, or pursue treatment before a catastrophic breakdown happens.

The technology was originally tested at WSU’s Hitchcock Research Racetrack - the only dedicated university-based racetrack in the USA - before being expanded to major racing venues including Saratoga, Belmont, Keeneland, Churchill Downs and Emerald Downs. Its success is part of a wider effort to improve racehorse welfare. Since 2009, raceday fatality rates in the U.S. have dropped from 2 deaths per 1,000 starts to 1.32 in 2023. While this is significant progress, every fatality is devastating for the horse, the people involved, and the public perception of the industry.

By identifying at-risk horses early and intervening, this technology has the potential to reduce fatal injuries, making it a promising and ethically important step forward in the future of horse racing science.

For more details, see:

Mc Sweeney, Denise, Yuan Wang, Scott E. Palmer, Mikael Holmströem, Kevin D. Donohue, Kelly D. Farnsworth, Macarena G. Sanz, David H. Lambert, and Warwick M. Bayly. 

Thoroughbreds deemed to be most at risk by inertial measurement unit sensors suffered a fatal musculoskeletal injury at a higher rate than other racehorses

Journal of the American Veterinary Medical Association (2025) published online ahead of print 

https://doi.org/10.2460/javma.25.04.0268

Can Humans Really See Pain in Horses? What New Research Reveals About Equine Facial Expression

(c) Igordabari Dreamstime.com

Horses play an essential role in human society as working animals, companions, and elite
athletes in sport. Because they cannot verbally report discomfort, their welfare depends heavily on how accurately humans can recognise signs of pain and distress. While most people can easily recognise pain in other humans, far less is known about our ability to identify pain in non-human animals, including horses. A new study from Anglia Ruskin University (ARU) Writtle provides important insight into just how limited, and how variable, is our ability to correctly read pain in horse faces.

The study, published in the journal Anthrozoös, is the first to directly compare how well people can identify pain from both human and horse facial expressions. A total of 100 adult participants took part in the research. Of these, 70 had no previous experience of caring for horses, while 30 had some level of horse-care experience, ranging from recreational riders to more experienced horse handlers. Participants were shown 60 photographs in total: 30 human faces and 30 horse faces. For each image, they were asked to rate the level of pain, emotional arousal (how intense the emotion appeared), and valence (whether the expression seemed positive or negative).

To establish a baseline for “correct” answers in the horse photographs, ten equine behaviour professionals, with specialist knowledge of equine facial cues, also rated the same images. Their assessments were used to measure how accurate the general participants were when judging horse pain.

As expected, participants were far better at identifying pain in human faces than in horse faces. Overall, most people struggled to recognise when a horse was in pain. This finding is significant, as horses often suffer from conditions such as lameness, dental issues, gastric ulcers and musculoskeletal problems that may cause discomfort without obvious behavioural signs.

However, the study also revealed that experience makes a meaningful difference. Participants who had previous horse-care experience were significantly more accurate at recognising pain in horses than those without experience. Furthermore, the number of years of horse experience was a good predictor of accuracy: the more time a person had spent working with or caring for horses, the better they were at spotting subtle facial indicators of pain. These cues may include changes in ear position, tightening around the eyes, tension in the muzzle, and altered nostril shape; features that are measured in structured tools such as the Horse Grimace Scale.

Interestingly, the study also explored whether certain psychological traits influenced participants’ ability to recognise pain. The researchers measured empathy (both emotional and cognitive) and social anxiety. Contrary to what might be expected, empathy was not significantly related to how accurately participants identified pain in either humans or horses. In other words, being more empathetic did not necessarily make someone better at recognising equine discomfort.

Social anxiety, however, did have an effect, but only when participants were judging human faces. People with higher levels of social anxiety tended to be more sensitive to pain in human expressions. When it came to horses, though, this pattern did not hold. In fact, socially anxious individuals were more likely to incorrectly rate horses as being in pain when they were not. This suggests that human psychological traits can influence how we interpret animal behaviour, sometimes leading to misinterpretation.

Rosa Verwijs, Senior Lecturer in Equine Behaviour and Nutrition at ARU Writtle and a co-author of the study, explained that this difficulty in recognising pain is partly due to the horse’s evolutionary history. As prey animals, horses evolved to hide visible signs of weakness in order to avoid predation. This means that pain expressions in horses are often subtle and easy to miss, especially for people who lack experience or training.

This research carries important implications for equine welfare. If owners, riders, trainers and carers struggle to recognise pain, horses may continue to be worked, ridden, or raced while injured or unwell. In some cases, undetected discomfort may escalate into more serious conditions or lead to dangerous behaviour such as bucking, rearing or bolting. Improving knowledge of equine facial expressions and body language could therefore help people seek veterinary attention earlier and reduce suffering.

The researchers hope their findings will be used to support better education and training programmes for anyone who works with horses. For equine science students, this study highlights the importance of developing strong observational skills and scientific knowledge of equine behaviour. Being able to accurately read the subtle signs of discomfort in a horse is not just a useful skill,  it is a vital part of protecting and improving equine welfare.

As the equine industry continues to place greater emphasis on welfare, research such as this reminds us that understanding the horse’s silent signals is one of the most important responsibilities of everyone involved in their care.

For more details, see: 

Gregory, Nicola J., Mollie Trimmer, Trudi Dempsey, Rosa Verwijs, Gabriel Carreira Lencioni, and Rachel L. Moseley. 2025. 

Reading Pain in Horse and Human Faces: The Influence of Horse Experience, Social Anxiety, and Empathy.

 Anthrozoös, (2025) October, 1–18. 

https://doi.org/10.1080/08927936.2025.2551433

Early allergen exposure protects Icelandic horses from developing midge-related allergy, Cornell study finds

(c) Oleksandr Korzhenko Dreamstime.com

A long-term study of Icelandic horses housed at Cornell University has provided strong evidence
that early-life exposure to an allergen can prevent the development of allergy later on. The findings, published in Frontiers in Immunology, also shed light on similar processes in humans and lend support to long-standing theories about how the immune system learns to tolerate environmental challenges.

The research focused on hypersensitivity to Culicoides midges; tiny blood-feeding insects whose saliva contains proteins capable of triggering intense allergic reactions in horses. Affected animals may suffer from severe itchiness, hair loss, dermatitis, and, in extreme cases, weight loss. Icelandic horses are particularly vulnerable once they leave their native environment because Culicoides species do not exist in Iceland. As a result, horses exported later in life often encounter midges for the first time abroad and may develop what is commonly known as “sweet itch.”

Lead researcher Dr. Bettina Wagner, James Law Professor of Immunology at Cornell’s College of Veterinary Medicine, had long been intrigued by anecdotal reports from breeders. “The rumour was that adult Icelandic horses exported to Europe developed Culicoides allergy far more often than those bred there,” she explained. Later epidemiological studies confirmed these observations, prompting Wagner and her team to investigate when and why allergic reactions occur.

Their 13-year study followed several cohorts of related Icelandic horses living together under identical environmental conditions at Cornell. The crucial difference among groups lay in the timing of first exposure to Culicoides and whether foals received allergen-specific maternal antibodies. Since Culicoides do not exist in Iceland, the imported horses provided a uniquely “naïve” population, free from prior exposure.

The results were striking. Horses experiencing Culicoides bites for the first time in adulthood showedmthe highest rates of hypersensitivity. 62.5% went on to develop the characteristic eczema-like symptoms. When first exposed during adolescence, only 21.4% developed allergy. Yet none of the horses exposed from birth ever showed signs of Culicoides hypersensitivity, despite having a genetic predisposition: all full-siblings had at least one allergic parent.

These findings strongly support the idea that early exposure to an allergen can “train” the developing immune system to tolerate it. On the Cornell Chronicle website, Wagner explains that during early development, immune responses are more flexible and more capable of learning to recognise otherwise reactive substances as harmless. “Training of the immune system to develop tolerance against allergens early in life creates a specific ability to adjust to those allergens and recognise them as harmless,” she said. “The earlier it happens, the more successful it is.”

In humans, similar concepts underlie the well-known hygiene hypothesis - the observation that children raised in environments rich in microbes and animal exposures tend to have fewer allergies than those raised in more sanitized conditions. Because horses share many features of immune function with humans, the study contributes important evidence to understanding how early-life environmental contact shapes lifelong immune health.

The study also tested whether maternal antibodies, specifically allergen-specific Immunoglobulin E (IgE) and Immunoglobulin G (IgG), played a role in determining allergy outcomes. Previous observations suggested that allergic mothers might pass IgE to their offspring, increasing the likelihood of allergy. However, the Cornell study found no such effect. Foals born to mares carrying Culicoides-specific IgE did not show higher allergy rates, nor did maternal antibodies appear to protect against or predispose foals to hypersensitivity.

By following the horses for at least nine years of repeated seasonal exposure, the researchers were able to firmly establish that early-life exposure, not maternal immunity, was the determining factor in whether hypersensitivity developed.

The work was led by first author Dr. Elisabeth Simonin, a postdoctoral researcher in Wagner’s laboratory, with collaborators from the University of Iceland, Cornell technicians, and veterinary students. The study was supported by the Harry M. Zweig Memorial Fund for Equine Research, which funded both the importation of the Icelandic herd and their long-term maintenance.

For equine scientists, these results highlight the importance of developmental timing in immune system training and raise questions about how management practices, especially for horses moved between environments, may influence long-term health. More broadly, the research offers a powerful example of how equine models can advance understanding of allergy mechanisms relevant to multiple species, including humans.

For more details, see:

Early allergen introduction overrides allergy predisposition in offspring of horses with Culicoides hypersensitivity
Frontiers in Immunology (2025)
Vol 16 . 1654693
https://doi.org/10.3389/fimmu.2025.1654693

Study shows “Bute” can temporarily disrupt assisted breeding in horses

(c) Zuzana Tillerová Dreamstime.com

 Research from Texas A&M University has found that phenylbutazone (“bute”), one of the most
commonly used pain-relief medications in horses, can interfere with artificial breeding techniques, especially at the crucial stages of egg maturation, fertilisation, and early embryo development.

This finding is particularly important for mares involved in assisted reproductive technologies (ART), such as intracytoplasmic sperm injection (ICSI) or embryo transfer. Many donor mares, especially older mares retired from competition or horses working through chronic lameness, may be receiving long-term bute to manage pain. 

Phenylbutazone is a non-steroidal anti-inflammatory drug (NSAID) widely used to manage lameness. While highly effective for pain, the Texas A&M team discovered that it can temporarily reduce the ability of a mare’s egg cells (oocytes) to develop properly in laboratory conditions.

In horses, assisted reproduction is more difficult than in humans. Mature oocytes must be collected, cultured for around 30 hours, and then fertilised by injecting an individual sperm directly into the egg, a process known as ICSI. If the egg fails to mature or the early embryo fails to grow, it cannot be used.

The research team tested the effect of a 10-day course of oral bute on oocyte quality. Oocytes were collected using transvaginal ultrasound-guided aspiration and evaluated shortly after treatment, then again one month and two-and-a-half months later. Cumulus–oocyte complexes (COCs), the small clusters of egg cells and surrounding support cells, were matured in vitro and fertilised using ICSI.

Key findings included:

Immediately after treatment (3 days post-bute):

• Oocytes from treated mares were less likely to be recovered and mature normally.
• Cleavage (early embryo division) rates were significantly lower.
• No fertilised oocytes from bute-treated mares developed into blastocysts, whereas 28% of those from untreated mares did.

Later in recovery (33 and 77 days post-treatment):

• Differences between groups disappeared.
• Oocytes collected after these intervals matured and developed at normal rates.

This means that bute’s negative effect on reproduction is real but temporary. A waiting period of at least one month after stopping bute appears to allow oocyte quality to return to normal.

This study suggests that:

• Egg collections performed shortly after a course of bute may fail, wasting both time and significant financial investment.
• Planning ART procedures around medication withdrawal could improve success rates.
• A minimum 30-day washout period may be advisable before oocyte aspiration.

Future research will explore whether other NSAIDs offer safer alternatives that do not interfere with reproduction.

For more details, see:

Luisa Ramírez-Agámez, Camilo Hernández-Avilés, Canaan M. Whitfield-Cargile, Michelle C. Coleman, Charles C. Love,

Treatment of mares with the non-steroidal anti-inflammatory drug (NSAID) phenylbutazone transiently affects in vitro maturation of equine oocytes and blastocyst development after Intracytoplasmic Sperm Injection (ICSI),

Theriogenology (2024), Vol 223, Pp 53-58

https://doi.org/10.1016/j.theriogenology.2024.04.017

Study finds social contact improves welfare and behaviour in stabled horses

(c) Abby Koriaty Dreamstime

New research has found that even limited social interaction can improve the welfare and
emotional state of horses kept in individual stables. The study showed that horses allowed to spend short periods of time with another horse displayed fewer stress-related behaviours, were calmer around humans, and appeared to perceive their environment more positively.

The research, carried out in France, involved 20 adult saddle-bred sport horses living in individual stalls. Over four months, half of the horses were given one hour of daily social contact in a “meeting box,” where they could interact freely with another horse. The remaining horses stayed fully isolated. Researchers observed all horses in their stalls, during grooming sessions with humans, and in a judgement bias test designed to assess emotional state.

In their stables, horses that had daily social contact displayed fewer stereotypic behaviours such as repetitive movements or abnormal oral actions compared with isolated horses. They also showed lower levels of aggressiveness, apathy, and hypervigilance, all indicators of compromised welfare.

During grooming, the socially housed horses were less tense and more relaxed, showing fewer negative reactions toward handlers. They also spent less time on alert, suggesting that regular interaction with other horses not only improved their general mood but also strengthened their trust in humans.

The results of the judgement bias test supported these behavioural observations. Horses that had daily social contact displayed a more optimistic bias, meaning they interpreted ambiguous stimuli more positively than their isolated counterparts. In welfare science, such optimism is considered a sign of a positive emotional state.

The findings highlight how social needs remain fundamental to equine welfare, even for horses managed in modern livery or competition environments. Horses are inherently social animals, and complete isolation can lead to behavioural problems, reduced welfare, and a less positive relationship with humans.

Overall, the findings indicate that even temporary but regular opportunities for social interaction can enhance the emotional wellbeing of individually housed horses. 

The researchers concluded that providing horses with the chance to engage in natural social behaviours helps reduce stress, prevent the development of stereotypies, and promote a more positive perception of their environment.

These results underline the importance of social needs in horse management. For a highly social species such as Equus caballus, complete isolation can have significant welfare costs. Allowing horses to interact, even for short periods, appears to improve their mental state, behaviour, and responses to human handling - key elements in maintaining both welfare and performance.

This suggests practical applications for yards and training centres where full group turnout may not always be possible.

By meeting horses’ natural social needs, owners and managers can help reduce stress-related behaviours, improve handling responses, and support a more positive emotional outlook.

In summary, horses that are given opportunities for daily social contact appear not only calmer and more optimistic but also more comfortable in their interactions with humans. The study adds to growing evidence that social enrichment, even in small doses, plays a crucial role in promoting both welfare and performance in stabled horses.

 

For more details, see:

 

A. Flamand, L. Robinet, A. Raskin, M. Braconnier, A. Bouhamidi, G. Derolez, C. Lochin, C. Helleu, O. Petit,

The social dimension of equine welfare: social contact positively affects the emotional state of stalled horses,

Animal Behaviour,  (2025) Vol 221,123055.

https://doi.org/10.1016/j.anbehav.2024.123055

UConn study links poor coat condition to higher parasite loads

A student works with a horse in the UConn stables
(c) Jason Sheldon / UConn)

Poor coat condition in horses has long been associated with heavy parasite burdens. A new
study from the University of Connecticut provides fresh evidence supporting this link and introduces a new tool to help assess it.

Led by Associate Professor Jenifer Nadeau in UConn’s Department of Animal Science, researchers developed a skin and coat condition scoring system to evaluate horses alongside their parasite levels. The work, published in the Journal of Equine Veterinary Science, builds on ideas that suggest that outward appearance may reflect an animal’s parasite resistance.

In the study, UConn’s herd of Morgan and Thoroughbred horses was regularly evaluated by students as part of the university’s hands-on equine science program. Faecal egg counts were collected to measure strongyle parasite loads, while separate teams of students, blinded to those results, scored each horse’s skin and coat condition.

The findings showed a clear trend: horses with higher parasite loads were significantly less likely to have good hair lustre. Although factors such as breed, age, housing, and season also influenced coat condition, parasite burden remained a strong predictor. Thoroughbreds, for example, tended to score lower than Morgans across several skin and coat categories.

The new scoring system may provide horse owners, managers, and even animal control agencies with a practical tool to help identify horses struggling with parasite-related issues or neglect. It can complement existing faecal testing by offering a quick visual assessment that may signal when further investigation is needed.

Nadeau plans to continue investigating how parasite levels change throughout a horse’s life and how coat condition can be used as an additional indicator of health. The study reinforces a message well known to equine professionals: paying close attention to coat quality can provide valuable clues about what’s happening inside the horse.

For more details, see:

N. DeBel, J. Johnson, A. Simoneau, B. Sweeney, G. Stearns, M. Amalaradjou, E. Gibbs, T. Moore, J. Nadeau,

The effect of parasitic load on skin and hair coat condition in Morgan and Thoroughbred horses,

Journal of Equine Veterinary Science, (2025) Vol 148, 105546

 

https://doi.org/10.1016/j.jevs.2025.105546

//////

Study on Deworming Practices

 The University of Connecticut Equine Extension Program has announced a study to better understand how horse owners and equine professionals manage deworming. Through a brief 10-minute survey, the team aims to gather real-world insight into current parasite-control practices and the industry’s overall awareness of effective deworming strategies.

Participation is voluntary, and all responses will remain confidential and used solely for educational purposes. The survey asks for basic information about respondents’ horses and their deworming experiences. Once started, answers cannot be revisited, and participants may stop at any time.

Those interested in contributing to this research can complete the survey at: https://s.uconn.edu/deworming-survey