It sounds completely backwards, the idea that you might actually increase health risks by postponing training and competition until a horse is four or older. It goes against the ages-old and widely held belief that you cause damage by initiating work before a horse's skeleton matures. Yet research conducted from the 1980s through the present day has steadily been debunking the old theories, replacing supposition and circumstantial evidence with hard scientific fact.

In short, the newly understood reality is that done correctly, exercising and even competing a 2-year-old horse could be beneficial by strengthening the horse's musculoskeletal system and decreasing his long-term risk of injury. We'll look at the research supporting this theory (most of which relates to racehorses), discuss how the information might apply to other types of riding, and touch on the unknowns that still remain.

The Challenge of Remodeling

The underlying theme that carries through the research done on bone development and exercise is that bone--particularly the cannon bone--must remodel, or change, in order to develop the shape necessary to endure the rigors of racing. According to Rick Arthur, DVM, an American Association of Equine Practitioners (AAEP) past president and a private Southern California racetrack practitioner for 27 years, the cannon bone of a young horse which has not exercised is almost perfectly round (see page 38). But in order to withstand the stresses of racing, the cannon bone must become more elliptical in shape--thicker on the front and inside. The challenge, Arthur adds, is getting the bone to remodel from one shape to the other without tripping the remodeling process over into a clinical problem such as bucked shins or cortical stress fractures (fractures in the outer tubular layer of the bone; see page 37).

Left to nature, this challenge isn't so great. The bone would remodel itself over time, responding gradually to stresses by building strong, dense lamellar bone (or "normal" bone) where it's needed, according to David Nunamaker, VMD. An orthopedic surgeon and head of the University of Pennsylvania's New Bolton Center, Nunamaker has been a major contributor to our understanding of the effects of exercise on bone development.

The problem lies with more of Nunamaker's discoveries. First, he found that if stress is applied to the bone suddenly, then periosteal fiber bone is formed instead of lamellar bone, as it's laid down more quickly than the stronger lamellar bone. Beneath the periosteal fiber bone, the underlying lamellar bone that doesn't remodel is prone to saucer fractures from cyclic fatigue later in the horse's life. The microfractures of bucked shins occur in the fiber bone and the underlying lamellar bone when the initial resorption phase of remodeling due to training stress results in bone that becomes temporarily weaker before it is made stronger. While a horse will recover from bucked shins with rest, Nunamaker's research shows that horses which bucked their shins tend to be the ones which get saucer fractures in this area later in their careers.

Nunamaker's research also showed that while natural remodeling (with pasture turnout or casual exercise) reshapes the bone over time, this new shape is not adapted for the higher stresses of racing. Even the long, slow workouts traditionally used to condition racehorses don't give a horse the bone shape he needs.

In one study, Nunamaker put two yearlings out to pasture with no riding, and put another two yearlings into traditional race training with no speed workouts. His results showed no difference in the bone shape of the four horses: The leg bones of horses undergoing long, slow gallops didn't change bone shape any more than in those left out to pasture, and they certainly weren't shaped correctly to handle the concussion and fatigue of racing at speed. In other words, traditional race training wasn't doing the job of preparing young horses' bones for racing.

Obviously, some new tactics needed to be tried--something that would put enough stress on the bone to cause it to remodel for speed, without adding too much stress too soon and causing lameness and periosteal fiber bone formation. Nunamaker devised a follow-up study that would examine the effects of speed work.

He looked at four groups of 2-year-olds. Group one was left at pasture. Groups two and three trained at separate tracks, but both took part in long, slow workout programs. Group four's workouts included short speed bursts three days per week (up to four furlongs at 13 seconds per furlong). The results showed that the pastured horses had undergone little bone remodeling, and both traditionally trained groups developed notable amounts of weaker periosteal bone.

However, the cannon bones of the 2-year-olds in the fourth group (which were trained at speed) showed almost no periosteal bone growth. Instead, they showed an increase in lamellar bone development. In fact, their bone shape became very similar to that of most 4-year-olds which had been training and racing for two years. The message was pretty clear: The speed work had added enough stress to prepare the bone for racing, without overdoing things to the point of causing bucked shins and periosteal bone growth.

The reason that speed work is effective, surmised Nunamaker, has to do with loading the front of the bone in compression with fast work rather than tension (stretching) as occurs during slow or little work. "So, horses on long, slow works remodel their bones for training, while horses that breeze more often remodel their bones for racing," he explained.

Nunamaker's studies also led him to one more conclusion: Once a horse's bone had remodeled appropriately--yielding a shape adapted for the stresses placed upon it--it would actually stay that way for life. "At this point," Nunamaker said, "after the bone has changed shape, you could change his training or racing program any way you want (except to stop his exercise entirely), because his bone won't change back again."

How Much Speed?

With short speed work seeming to be the answer to effective bone remodeling, the next question became how much speed work would be the right amount to stress the bone just enough, but not too much? Unfortunately, there is still no definitive answer to that--and probably never will be, since so much depends on the individual.

All the same, one of Nunamaker's colleagues, John Fisher, DVM, a veterinarian and racehorse trainer, set out to see if he could develop an optimal program to use on the horses in his barn. The schedule he came up with starts with basic conditioning. Once a horse could comfortably go one mile with furlongs at 18-20 seconds each, Fisher began introducing speed work twice a week, using a pace of one furlong (one-eighth-mile) per 15 seconds, and starting with a one-furlong distance. Every three to four weeks, he increased the distance by another furlong. By the end of three months, the horse was doing three-eighths-mile bursts at speed, but only once every five days. On non-speed days, the horse was either rested or galloped, allowing time for the stimulated bone to remodel.

Over the years, Fisher has found this program effective in his barn. Shin problems have dropped from being practically the norm to being a rarity, yet he's seen no increase in injuries to his horses' ankles and knees. Arthur, too, has seen a notable difference at the tracks where he works. "When I started practicing more than 25 years ago, it was rare to have a horse that did not have bucked shins," he says. "Nowadays, it's almost the opposite, because most people manage (training) in a more modern way, adjusting exercise to match the remodeling process. And it's not that different, it just takes more time--and time is money." In fact, he adds, the one trainer he knows who routinely has horses with bucked shins "is 80 years old and has not changed the way he trains in 60 years. He believes they have to buck (their shins)."

The important point here is not simply the speed work or the exact distances, but the fact that the exercise program is progressive--increasing the stress on the bone just enough to stimulate remodeling, allowing time for changes to occur, then gradually increasing the stress. On the other hand, training regimens that rapidly increase stress levels over a short time--such as some of those designed to bring a young horse to racing speed in just a few weeks for 2-year-old in-training sales--has the opposite effect, increasing the risk of injury.

The Resilience of Youth

Even if, as the research indicates, these exercise regimens are effective in building stronger bone, the question still remains whether it is necessary, much less a good thing, to put young horses through these programs. The answer isn't exactly clear-cut, but there is evidence that it might indeed be safer to apply these stresses to younger rather than older horses.

For one thing, a young, growing horse's skeletal structure is basically programmed for rapid remodeling, according to Jeffrey P. Watkins, DVM, MS, Dipl. ACVS, professor and chief of surgery for the Department of Large Animal Medicine and Surgery at Texas A&M University's College of Veterinary Medicine.

Furthermore, the horse's skeleton is probably more resilient at two, allowing the horse to overcome any injury faster, says C. Wayne McIlwraith, BVSc, PhD, Dipl. ACVS, the Barbara Cox Anthony Endowed University Chair and director of the Equine Orthopaedic Research Center at Colorado State University. "I think the bone may become more brittle (with age) and thus more vulnerable to problems," McIlwraith notes.

So it seems likely that a younger horse would be able to remodel his bone more easily and rapidly, with fewer risks, than horses which did not start a training/conditioning program until they were older.

Research done by Roger Smith, BVet Med, PhD; and Allen Goodship, BVSc, PhD, MRCVS, at England's Royal Veterinary College, further supports this idea. Smith and Goodship have noted that before the age of two, a horse's tendons grow and adapt in response to weight-bearing stresses. Exercise before two years of age, then--such as semi-weekly sprints that build gradually from one furlong to one mile--might accelerate this process, they say. But once the horse is older than two, tendon changes tend to be degenerative, not adaptive.

"We have found that some components of tendon matrix are present in the early stages of development, but decline in later life; one such protein is cartilage oligomeric matrix protein, which we feel is likely to play a part in the ability of some tendons to respond to exercise," explains Goodship. "The superficial digital flexor tendon sustains a high incidence of injury on both performance and leisure horses--it does not respond to exercise in terms of becoming larger or stronger in the older horse and indeed may sustain damage more readily in older horses. We are suggesting that when exercise is imposed on a gradual basis in the younger horse, the matrix of this tendon is not damaged and remains normal. It seems sensible to begin to condition the musculoskeletal system gradually during its development rather than wait until the muscles, bones, and tendons have formed and matured and then impose a sudden training schedule in the young adult horse." From this, Smith and Goodship also theorize that exercising and racing young horses might actually decrease the risk of tendonitis when the horse is older.

This is particularly important in light of another Goodship finding: There is a direct relationship between bone and tendon strength, such that horses having inherently strong bones also have inherently strong tendons and vice versa.

McIlwraith adds that, "There is no question that if the muscle tone is slack and the horse is not fit, they are more vulnerable to injury. Muscle fitness is important for a sound overall musculoskeletal system."

Another study conducted by leading researchers in Great Britain seems to further support the idea that starting racing at age two could be beneficial. This particular study found that horses which began racing at age four were twice as likely to die of catastrophic injury as horses which began racing at age two. It's important to note, though, that there was no clear answer as to why this was the case.

Likewise, Sue Stover, DVM, PhD, and her co-workers at the University of California, Davis showed (as part of the California racetrack post-mortem program) that 4-year-olds were most commonly represented with catastrophic injuries.

Another group of researchers is seeking to prove that early conditioning will prevent injury. The group, which calls itself the Global Equine Research Alliance (GERA), is comprised of researchers from four universities in four countries: Drs. McIlwraith and Chris Kawcak, DVM, PhD, from Colorado State University; Elwyn Firth, BVSc, MS, PhD, Dipl. ACVS, and Chris Rogers, BSc, Dipl. Agr. Sci., M. Agr. Sci., PhD, from Massey University in New Zealand; Goodship and Smith from the Royal Veterinary College in England; and Dr. A.B. Barneveld and Rene van Weeren, DVM, PhD, Dipl. ECVS, Dipl. RNVA, from the University of Utrecht in The Netherlands.

GERA researchers are working with a group of 33 foals they bred. When the foals were three weeks of age, half were started on a canter exercise program. The other half were left at pasture. These treatment and control protocols continued until the foals reached 18 months of age, explains McIlwraith. At that time, all were broken into race training with the same training regime. While the researchers won't be ready to present their preliminary data until July 2004, McIlwraith shares some general comments: "If you ask, 'Can we tell the difference between the conditioned group and the others?' Yes, the muscle is better developed. As far as proving our hypothesis (about bone), it's too early to say."

Even if all the research finds that it's not only okay but better for the long-term soundness of the horse to start training at two--or even as a foal--it's not a good idea to set one particular age as the best time to begin exercise, says Arthur. "You can't pick a particular age, because what's appropriate for one horse is not appropriate for another," he states. "You have to look at the individual and make individual decisions. That's where horsemanship comes in."

It's also important to note that a horse must undergo appropriate bone remodeling regardless of age. As McIlwraith says, "If a horse is put into training as a three- or four-year-old, he goes through the same problems that we see in the baby." For instance, he continues, "Years ago, when Arabian racing was getting popular in California, people were training these horses at five and six, and they were still getting the same problems as younger horses did. It's a matter of the bone responding to stimulus and saying, 'I need to be stronger.' "

Therefore, whether a horse is two or four when he enters training, he must embark on an exercise process that encourages gradual remodeling, allowing the bone to strengthen without causing problems.

Training vs. Competition

If we can now assume that, in general, it is not harmful to train a juvenile horse, does the same apply to competing one? It seems so, says Arthur. He points again to the studies showing that younger horses have fewer catastrophic injuries per start than older horses. In addition, research done by Craig Bailey, BVSc, MACVSc, PhD, resident in small animal surgery at the University of Sydney (US) and colleagues at US found that horses first raced at age two had longer racing careers than those first raced at three or older. (Again, the reasons behind this difference are not known and could be related to the older horses having had training injuries or simply a lack of talent.)

Still, cautions Arthur, you can't simply dump even a properly conditioned young horse into the competitive fray without care. You need to take several factors into account. Footing, for example, makes a difference--not only the difference between good footing and bad footing from one track or show ring to the next, but also the difference between the type of footing a horse has been trained on and the type of footing he will compete on.

Fisher and Bill Moyer, DVM, head of the large animal medicine and surgery facility at Texas A&M's College of Veterinary Medicine, for instance, did research that showed if a horse was going to be raced on dirt, then he needed to train on dirt. The more forgiving surface of a wood chip track simply didn't create enough stress to strengthen the bone for racing on the harder dirt surfaces.

Along the same lines, Tim Parkins, BSc, BVSc, PhD, of Liverpool University found that the use of sand gallops was linked to increased fracture risk. And James Wood, BSc, BVetMed, MSc, PhD, DLSHTM, MRCVS, of England's Animal Health Trust, also found that fractures were more likely to occur in horses which trained over wood chip or sand gallops.

Competition places more stress on the musculoskeletal system than even a good training program. Part of that might be the distances horses are asked to cover. For example, Wood's research found that the risk of fatal injuries increased with the distance raced. By shortening races a half-mile, he saw a 35% reduction in risk per start.

A similar problem pops up in some non-racing disciplines. For instance, in the Western performance sports, such as reining and cutting, a horse often does multiple go-rounds to complete a single event, notes Jerry Black, DVM, immediate past president of the AAEP and senior partner of Pioneer Equine Hospital in Northern California. Between the go-rounds and the practice sessions, he says, "The horse is often being worked eight to 10 times in just a few days."

It's even worse when a horse competes in multiple divisions, such as Open with the trainer and Non Pro with the owner. "I think we're finding that these young horses cannot handle that mentally and physically," says Black. "It's too much pressure."

Mind Over Matter

That brings up another aspect of competing juveniles: It's not just their bodies that you need to worry about. There is also considerable mental stress placed on a horse in competition--from the demands of the contest itself, to the trailer rides, the strange barns, and the different sights, sounds, and horses to which the youngster is suddenly exposed. Not every young horse is up to handling that pressure, especially on a week-after-week basis. Indeed, notes Arthur, even the training environment itself can cause stress beyond the musculoskeletal system. "You have a young horse in a new environment doing things that he doesn't always understand," he says. "Some horses won't care and most will want to do more, but there are some horses that don't tolerate it--it's too psychologically stressful for them."

"You need to be smart about it," says Black. "Limit the number of shows you go to, just like racing, where trainers don't send their 2-year-olds out every weekend."

Indeed, Parkins learned that the fewer days a horse is allowed off between races, the higher his risk of fracture.

For both mental and physical reasons, Black recommends giving the horse a break not just from competition, but also from the rigid training routine. "Basic trail riding, letting them relax and not be under the pressure of specific training, is good for their minds and goes a long way in helping them have longer careers," he says. His training program starts with 30-45 days of basic saddle training in December of the horse's yearling year or January of his 2-year-old year. That's followed by 60 to 90 days of pasture turnout before the horse returns for serious training.

Nunamaker's research suggests that while layoff might be good for the mind, time between training phases might not be the best thing for bone development because the layoff initiates the first phase of remodeling to match the decreased work load. This resorption phase, as discussed earlier, initially weakens the bone. When training resumes a short time later, it's with weakened bone, which can cause more problems (i.e., bucked shins) than if the horse were kept in continuous training.

The Universal Elements

While all of the research to date is reassuring news to those who've long held that training and competing young horses does no harm, one point is important to remember: The data aren't universal.

"The only place where fast, short speed work has been shown to be beneficial is in bucked shins of young Thoroughbred racehorses," says McIlwraith. "Extrapolating (beyond that) has risks."

Still, Nunamaker's findings indicate that the same general idea should be applied--and adapted--to fit the particular stress of each performance endeavor. "Depending on what the training is like, you can just about change the bone in any direction you want," he says. Speed training works for racing, he explains, because racing involves running at speed, which places the leg bone at a different angle of impact than jogging or galloping. Standardbreds, he notes, rarely get bucked shins because they already train at the same speeds at which they race.

Black agrees that there is no research in other disciplines as extensive as what we have on the racing industry. However, he adds, "All performance horses need to be stressed a bit."

Arthur agrees, saying, "Whatever sport the horse is used for, his body has to adapt to the stress of that job. You can't take a horse from the pasture to barrel racing full bore without problems."

Think about it from a human point of view and it makes a lot of sense, he adds. "If you were a baseball star who wanted to take up track or football, you would have to retrain for the new sport," adds Arthur. "The whole body has to adapt."

Still, the same basic conditioning principles should apply across the board, he acknowledges: Take the horse's ultimate job, then scale it back to the basics and gradually build the stress level so that it remodels the bone over time, developing strong lamellar bone.

For example, with Western performance horses, training that focuses on medium and extended trot and canter seems to do the job, says Black. "That training process goes on for a period of several months and accomplishes basic interval training that builds bone density," he says, noting that reining, cutting, and reined cow horses are typically started as early 2-year-olds or sometimes as late yearlings, just like racehorses. "It's my impression over the last 25 years that it doesn't make any difference at all if you start at a young age or wait a year."

The Bottom Line

While there is still much left to learn about the relationship of exercise to age and bone development, the studies to date are enough to confirm that it is humane to train and compete juvenile horses--provided you progress with common sense, wisdom, and good horsemanship. As McIlwraith says, "The bottom line is that exercise is good, and we presume that younger is better. We've pretty well disproven the notion that 2-year-olds are any more vulnerable than older horses." In short, he concludes, "The horse is fragile, but that's not necessarily a function of age."


Biles, Deirdre B. Alliance to Study Training Regimes in Youngstock. ( May 2001.

Early Racing Will Not Harm Horses. Point 2 Point Racing News, Your Guide to Westcountry Horses ( December 2001.

Geor, R. Starting 'em Young. The Horse, January 2001, 57-66. Article #72 online.

Locomotion and Locomotion Disorders of Horses and Cattle. University of Utrecht Research Programmes. June 2003.

Record Crowd Hears Latest on Racehorse Health. Horserace Betting Levy Board Press Releases ( March 2003.

The Fascinating Physics of Training. Grayson-Jockey Club Research Today ( Vol. 18, No. 6, 2, December 2001.

Sellnow, L. Bone Formation With Exercise. The Horse, June 1999, 53-61. Article #338 at

West, C. AAEP Convention: Milne Lecture--Bucked Shins. Article #4066 online.

West, C. AAEP Convention: Milne Lecture 2--Bone and Fracture Treatment. Article #4067 online.


Age and Exercise in Warmblood Foals

Researchers at Utrecht University in the Netherlands have studied the effects of exercise--and lack of exercise--on bone development in Dutch Warmblood foals. Specifically, they looked at subchondral bone, which lies under the articular cartilage and supports the joints. This layer has a high water content in newborn foals, but the water is gradually replaced with calcium and collagen when the foal begins to put weight on his joints. Healthy subchondral bone is considered vital to healthy joints, as studies indicate that when abnormal articular cartilage appears, it's preceded by abnormal subchondral bone.

The Utrecht study placed week-old Dutch Warmblood foals into two groups. One group lived in box stalls while the other group lived at pasture 24 hours a day. After five months, tests showed that the stall-bound foals had significantly decreased calcium deposits, collagen development, and collagen cross-linking (a trait important for collagen strength). The pastured foals had normal development of subchondral bone. This, combined with some additional data, led the researchers to conclude that pasture exercise is sufficient (perhaps necessary) for normal subchondral bone development, and that the critical period for this development is before the age of five months. --Sushil Dulai Wenholz

Source: Brama, P.A.J.; TeKoppele, J.M.; Bank, R.A.; et al. Equine Veterinary Journal, 34(2), 143-149, 2002.


The Link to Rehab

Optimal rehabilitation strategies for horses pulled out of training due to an injury or health problem is a concern. "There is an association between disuse and getting injured later," says Wayne McIlwraith, BVSc, PhD, Dipl. ACVS, director of the Equine Orthopaedic Research Center at Colorado State University. "Last year, we had a very good 2-year-old racehorse that we did arthroscopic surgery on. He was walked for a month, then trotted for a month, then on his first gallop, he shattered his humerus (a bone in the shoulder; he had to be euthanized). Stress fractures were present at the site of fracture, showing that the stress fractures preceded the catastrophic injury. This not an unusual history in such cases, and we still need to find better ways of nursing bone back to full strength."

According to Nunamaker's research, a major reason for this sort of incident is that when a horse is sidelined from training, the bone begins a resorption process (breaking down of bone to initiate remodeling) in response to the decreased stresses on it. If the horse is brought back too quickly, it's not uncommon for an injury to occur. Thus, it appears that the horse must re-start the bone remodeling process. Researchers hope that ongoing studies will uncover more specifics on this subject in time.--Sushil Dulai Wenholz

About the Author

Sushil Dulai Wenholz

Sushil Dulai Wenholz is a free-lance writer based in Lakewood, Colo. Her work appears in a number of leading equine publications, and she has earned awards from the American Horse Publications and the Western Fairs Association.

Stay on top of the most recent Horse Health news with FREE weekly newsletters from Learn More

Free Newsletters

Sign up for the latest in:

From our partners