General body language may be more reliable as an indicator of mild pain than the equine pain face.

What if you could tell your horse were lame, even mildly, just by watching his behaviour at rest?

According to Swedish researchers, it could be possible. At least in their experimental conditions, they could detect lameness by paying attention to horses’ resting body language. By observing your horse’s posture, head position, location in a box stall, focus, and interactive behaviour—and possibly facial expressions as well—you might be able to pick up on signs of low-grade lameness before you even take the horse out for a lameness check.

Their findings are published in the open access journal Animals.

“Horses cannot self-report, and their pain behaviour is probably what comes closest to a self-report,” said Katrina Ask, DVM, PhD candidate at the Swedish University of Agricultural Sciences in Uppsala. “We just have to learn to understand what they communicate to us.”

In their new study, Ask and her fellow researchers found that horses can give clear behavioural signs that their legs hurt, even when that pain is minor, mainly through the way they stand, where they stand, and how they react to sights and sounds while standing.

In eight horses with mild pain in a single hindlimb, they saw that the horses tended to stand rigidly, with a low, flat head carriage, she said. They also generally stood in the back of a box stall or in the middle of the stall while facing backwards, and were less interactive—meaning they paid less attention to humans and to sights and sounds around them.

The pain-related posture they saw has already been described in two science-based pain scales: the Equine Pain Scale (EPS), suggested in 2015 by Danish researchers, and the Composite Pain Scale (CPS), developed in 2008 by French and Canadian researchers. The EPS has also outlined the stall-position criteria, and both EPS and CPS have described the changes in interactive behaviour.

The EPS and CPS are two of four recognised pain scales described specifically for orthopaedic pain in horses, Ask said. The other two—the 2015 Equine Utrecht University Scale of Facial Assessment of Pain (EQUUS-FAP) and the 2014Horse Grimace Scale (HGS)—both focus on facial expressions of pain.

While each pain scale has its merit in different kinds of equine pain, Ask and her fellow researchers aimed to find out which aspects of each pain scale could point specifically to mild orthopaedic pain in horses at rest.

To do so, they injected one hock joint with a biological inflammatory to induce mild to moderate short-term arthritis in each of eight healthy, sound horses of an average age of 14 years old. They had three trained observers watch the horses’ behaviours at rest both before and after pain induction. Within days, all the horses had made a full recovery, Ask said.

They found that full-bodied behaviour like posture, stall-standing position, and interaction levels were more consistently graded by the observers and pointed to discomfort, Ask said. Although the horses also clearly showed facial expressions of their pain, the trained observers didn’t always agree with what they saw, making it more difficult to come to a consensus.

As such, the more general body language was more reliable as an indicator of mild pain, according to Ask.

“In our study, we saw that facial expressions were important but apparently more difficult to assess precisely than body behaviour, since the observers disagreed more often on what they saw,” said Ask. “Body behaviours seemed easier to assess and were in this study more associated to orthopaedic pain. Whether this is due to the fact that only trained observers can reliably assess facial behaviours, or due to the presence of specific orthopaedic pain is something we are working on.

“With only a little training, assessment of orthopaedic pain would therefore benefit from including body behaviours also during undisturbed rest in the box stall, and not only during movement as in the lameness exam,” she continued. “In pain scales based only on facial expressions, the facial expressions may not be recognised by the untrained observer or during situations with many external inputs (for instance during stress) in horses with mild orthopaedic pain, as pain of higher intensity may.”

The findings are particularly important for application in the home stable, according to Ask. Because horses are less likely to hide signs of pain at home, where they feel more comfortable, owners could be at the front line of recognizing these subtle indications of mild pain.

“We believe that our research should be transformed into education – and we think that the owner is the closest to monitor changes in their horse’s behaviours and facial expressions over a long time,” said Ask.

“Pain assessment can be challenging for clinicians since they often observe the horse at the hospital, where the horse may be stressed and nervous,” she said. “These are affects that can mask the pain assessment. By increasing the time for pain assessment and register both facial expressions and behaviour, the quality of the assessment may improve.”

The ethics of inducing mild, temporary pain in horses for research

Are you concerned about the eight horses who underwent mild inflammatory pain in their hocks for this study? While research sometimes leads to discomfort for a few individuals, national ethics boards oversee scientific study procedures and weigh the costs and benefits for both human and animal subjects. The Swedish ethics board doesn’t take animal research lightly, according to Ask. The horses in her study were scrupulously cared for, and their mild pain contributed to a much larger good for horses worldwide.

“The number of horses, the amount of pain, and the chance for a normal life after the study are all thoroughly considered before approval [of the study by the ethics board],” Ask told Horses and People.

“The reason for inducing orthopaedic pain is that there is no gold standard for whether a horse experiences pain or not,” she continued. “In this way, we achieve a standardised and controlled pain experience which can allow us to use the smallest possible number of horses. We chose a pain model with a temporary and fully reversible pain type. All horses returned to soundness again within 24-72 hours.

Humane endpoints were included, meaning that if the lameness degree exceeded 3 on a scale of 5, analgesic treatment had to be provided. In this way, we could closely monitor the change in pain intensity over time to understand how facial expressions and behaviours change with pain intensity. We could also obtain information on what behaviours and facial expressions that were normal for each horse – their baseline.

“The data collected can be used for several studies and will rescue the many horses suffering from unidentified pain, and also those horses being regarded as ‘stupid’ or non-collaborative, being punished for trying to cope with their painful orthopaedic diseases,” Ask added. “It’s a utilitarian ethical framework, where the few suffer a little for the great benefit of the many.”

This study is published in Animals. It is titled: Identification of Body Behaviors and Facial Expressions Associated with Induced Orthopedic Pain in Four Equine Pain Scales by Katrina Ask, Marie Rhodin, Lena-Mari Tamminen, Elin Hernlund and Pia Haubro Andersen. You can read it here.

Download the Equine Pain Face poster here. 

Learn to recognise subtle signs of pain here.

Read our report on the Horse Grimace Scale research.