There’s a compelling theory that if we do less when horses are younger and give them longer to mature then they’ll have longer, healthier, careers. So why doesn’t the scientific research agree? Social media loves that bizarrely-inaccurate infographic of skeletal maturity and growth plate closure that somehow never gets called out, but underneath the pseudoscience, some concerns are real
What the science actually shows:
There’s a good overview in a 2021 review in Animals, where Team Massey / Chris Rogers – consistently credible researchers into the effects of early exercise – directly examined skeletal maturity. They looked at growth plate closure and somatic growth and saw that horses complete the equivalent of rapid infant growth by weaning (4–6 months) and completes childhood growth by roughly 11 months old. At 2 years old, the horse has achieved most measures of maturity, including the closure of growth plates.
Figure taken from Rogers et al., 2021, summarising findings for relative maturity for horses and humans
All the scientific literature shows that the growth plates of the vertebrae (spine) close around 18 months and that’s pretty much final mature morphology. So why are there so many articles and pictures doing the rounds saying the spine is mature by 5-8 years? Some of the “vertebrae” that form the wings of the pelvis (tuber sacrale and tuber coxae) might fuse late, at around 5-8 years, so it’s likely that’s where part of the confusion originally arose. Similarly there’s a much-cited old dissection paper by Haussler et al. in 1997, showing some evidence for cartilage in the sacral vertebrae in 5-8 yo TBs, which often gets quoted as “the spine”, “open growth plates” and extrapolated to the idea that riders could sprain a horse’s back. More recent research from Elizabeth Collar at UC Davis (2020) using CT and histology found that lumbosacral growth plates are inactive by 2 years of age, as expected, but that cartilage remnants may persist much longer, which is what was found in Kevin Haussler’s dissection. These misunderstandings are always self-propagating and now if you Google it, you’ll get a load of articles reporting misunderstandings as fact. Would it matter if the pelvis had unfused growth plates? If your horse is in a very specific accident then at a stretch perhaps, otherwise no, there’s no reason why it would. It’s not a relevant part and even if it was, actual growth has finished well before closure shows up on x-ray, and growth plates are not a weak point, they just look like a gap on x-rays because it’s a different type of tissue.
The muscles and bone strength are still underdeveloped in a young horse and that does matter. For instance, the back needs to be supported by an abdominal sling and if it’s not then damage can be done, but the right way to develop that is with (controlled, well-judged) work. Bones, cartilage, ligaments and tendons need to develop and the best way to condition these is also controlled, well-judged work, which will decrease the chance of lameness in later years. And yes, the brain is immature so we can’t drill the horse, and it needs to fun because why else should they do it. This is a world away from the idea that working a horse is going to hurt it but still agrees with not doing the wrong sort of loading.
Functional Adaptation 101
1. The “Window of Opportunity” (Tendon & Cartilage)
From a tissue-science perspective, waiting until a horse is “mature” (age 6+) can actually be a disadvantage.
- Tendon Plasticity: Exercising horses at a young age improves tendon quality, and there’s actually a developmental window that closes. Studies (e.g., Dowling et al.) show that certain tendons, like the Superficial Digital Flexor Tendon (SDFT), lose much of their ability to adapt to high-strain loading after about 19 months of age. Indeed my own research showed that tendons and ligaments progressively stiffened until maturity (2-3 years), then stayed much the same until about 8 years old.
- Cartilage Maturation: Articular cartilage is non-vascular and has almost zero regenerative capacity in adults. Scientific consensus (like research from van de Lest) suggests that if cartilage is not “conditioned” through moderate loading while the horse is still a juvenile, it fails to develop the collagen network and glycosaminoglycan content required to survive a high-impact career later. Similarly, there’s quite a nice EVJ paper by Brommer et al (collab with Team Utrecht / Rene van Weeren – also consistently credible in this area) that shows “At age 18 months, functional adaptation [of cartilage], as assessed by the biomechanical characteristics, has progressed to a level comparable to the mature horse and, after this age, no major adaptations seem to occur”
2. Wolff’s Law vs. The “Soft Bone” Myth
The idea that young bones are “soft” and should be bubble-wrapped is also a biomechanical oversimplification.
- Mineralization: While growth plates are indeed cartilage, the cortical bone (the shaft) and the subchondral bone (under the joints) have been repeatedly shown to respond to strain by increasing mineral density, trabecular structure, and cross-sectional area, i.e. increasing strength (Wolff’s Law, kicking around since 1892 and still holding up well to scrutiny).
- The Paradox of Rest: Research in Thoroughbreds has shown that horses started as 2-year-olds actually have a lower incidence of catastrophic musculoskeletal injury compared to those started at 3 or 4 (e.g. Logan and Nielson, 2022). This is because their skeletal systems undergo “modelling” (shaping) in response to work, whereas older un-modelled skeletons have to go through “remodelling” (repairing micro-damage), which is a much riskier process.
3. Where the Concerns are Actually Biomechanically Sound
The argument for not jumping isn’t necessarily wrong, even if the “growth plate” reasoning is a little crazy.
- Torque and Shear: While straight-line loading (sprinting/hacking) is highly beneficial for bone density, jumping introduces high-magnitude shear forces and torque.
- The Axial Skeleton Gap: Most “early work” studies focus on the appendicular skeleton (legs). The axial skeleton (spine) is a different beast. Putting a rider’s weight on a very young horse while simultaneously asking for the extreme spinal flexion/extension required for grid jumping is where the “adaptation” argument starts to lose its scientific footing and enters the realm of “overloading.”
So don’t wait until the plates close but find the threshold where you trigger adaptation without causing micro-fractures. Jumping grids at 3.5 years old is likely exceeding that threshold for the spine and sacroiliac joint, even if the legs can handle it. (Jean-Marie Denoix is great on the sacroiliac joint, if you’re looking for a literature rabbit hole).
Our concerns about horses being retired early are also valid, but the causes are more complex than “started too young.” We need to look at overtraining, inappropriate training methods, poor surfaces, inadequate recovery time, or riding horses through pain. The growth plate confusion becomes a proxy for a broader welfare argument that deserves a more sophisticated analysis.
The mislabelled pictures and the social media arguments take pieces of real anatomical facts (vertebral physes do close slightly later than limb physes) and draw conclusions the research doesn’t support (therefore no meaningful work before 5-6 years). Partially understood science being misapplied. The horse evolved as a precocious cursorial species. The requirement to avoid predation and cover significant distances while the dam grazes has altered the relative temporal pattern of growth phases. Within their first week, foals cover 7+ km daily. This isn’t an animal whose skeleton evolved to be protected from load until age six. Ironically equine development and equine biomechanics are fields where we have really solid evidence and good scientific support, and it’s always in danger of being drowned out by louder voices.
Selected Refs, badly organised
Brommer, H., Brama, PA, Laasanen, Helminem, van Weeren, PJ, Jurvelin, JA., Functional adaptation of articular cartilage from birth to maturity under the influence of loading: a biomechanical analysis. EVJ 2010, 37,2 146-152
Butler, Janet A., et al. Clinical radiology of the horse. Blackwell Scientific Publications Ltd, 1993. [Vertebrae growth plates closure]
Fretz, P.B.; Cymbaluk, N.F.; Pharr, J.W. Quantitative-analysis of long-bone growth in the horse. Am. J. Vet. Res. 1984, 45, 1602–1609. [growth stops before physical closure]
Logan AA, Nielsen BD. Training Young Horses: The Science behind the Benefits. Animals (Basel). 2021 Feb 9;11(2):463
Łuszczyński, J., M. Pieszka, and K. Kosiniak-Kamysz. “Effect of horse breed and sex on growth rate and radiographic closure time of distal radial metaphyseal growth plate.” Livestock Science 141.2-3 (2011): 252-258.
Moffat, P. A., et al. “The influence of exercise during growth on ultrasonographic parameters of the superficial digital flexor tendon of young Thoroughbred horses.” Equine Veterinary Journal 40.2 (2008): 136-140.
Rogers, C.W.; Gee, E.K.; Dittmer, K.E. Growth and Bone Development in the Horse: When Is a Horse Skeletally Mature? Animals 2021, 11, 3402. https://doi.org/10.3390/ani11123402
Rogers, Christopher W., et al. “Evaluation of a new strategy to modulate skeletal development in racehorses by imposing track‐based exercise during growth: the effects on 2‐and 3‐year‐old racing careers.” Equine veterinary journal 40.2 (2008): 119-127.
Rogers, Chris W., et al. “Early exercise in the horse.” Journal of Veterinary Behavior 7.6 (2012): 375-379.
Strand, Eric, et al. “Radiographic closure time of appendicular growth plates in the Icelandic horse.” Acta Veterinaria Scandinavica 49.1 (2007): 1-7.
Veraa, S.; Scheffer, C.J.; Smeets, D.H.; de Bruin, R.B.; Hoogendoorn, A.C.; Vernooij, J.C.; Nielen, M.; Back, W. Cervical disc width index is a reliable parameter and consistent in young growing Dutch Warmblood horses. Vet. Radiol. Ultrasound 2021, 62, 11–19. [wither height as proxy for vertebral growth]
#equinebiomechanics #biomechanicstheory