A pilot study on factors influencing the career of Dutch sport horses



Reasons for performing study: Welfare concerns over the reasons for interruption or termination of a sporting career in horses have been expressed.

Objectives: To study the career of competition horses and factors that influence career length.

Methods: In 2004, 46,576 rider-horse combinations were registered with the Dutch National Equestrian Federation (KNHS) in dressage, show jumping, eventing and endurance. From this population, approximately 1% of horses in each discipline were selected at random and all recorded competition data from the KNHS registration system collected and supplemented by detailed rider/owner telephone enquiry, carried out in July 2009.

Results: A total of 520 horses were included aged 7.1 ± 3.2 years; endurance horses were on average slightly older than horses in the other 3 disciplines. A total of 334 horses (64.2%) ended their competition career with their initial rider during the study period; the mean total career time was 3.8 ± 2.8 years. Reasons for this included being sold on (51.5%), veterinary reasons (23.9%), being used for breeding (8.7%), changing to a recreational career (8.1%), rider-related issues (6.3%) and retirement (1.5%). No follow-up was available for horses sold. Orthopaedic problems accounted for the majority of the veterinary career-ending decisions (63.7%). A total of 385 horses (74.0%) had one or more career breaks; main reasons were rider-related issues (39.2%), others included temporary withdrawal from competition (21.6%), veterinary problems (21.8%), breeding (9.1%) and miscellaneous (8.3%).

Conclusions: Veterinary reasons for career breaks or termination of career accounted for 21.8 and 23.9% over the period investigated. They formed the second reason (after selling) why horses stopped competing with their initial rider/owner. Veterinary reasons were the third important cause for a temporary career break.

Potential relevance: Further in depth research of the reasons for interruption or termination of an equine career may be useful for the improvement of equine welfare.


Animal welfare issues have become increasingly important in equestrian sports and other horse-related activities. This is a worldwide trend but, in some countries, these concerns are more urgent than in others and the Netherlands is one of the front runners in this regard. It is the first and to date, the only country where a recognised political party focusing on animal rights (‘Partij voor de Dieren’) has been elected in the national Parliament and is additionally represented in other democratically elected bodies. Public interest in animal welfare has forced the Government to take action. In 2007 the Directive Animal Welfare (Ministerie van LNV 2007) was published by the Ministry of Agriculture, Nature and Food Quality. This Directive contained a specific chapter on horses and expressed concerns about equine welfare issues and use and abuse of sport horses. As a follow-up, the Minister ordered the equine sector to produce within one year a workable plan to safeguard and improve equine welfare in housing, feeding, transport and training methods and required that proposed improvements should be realised within 3 years.

For any action plan to be developed, a good basis of knowledge on the actual situation of equine welfare is necessary; insight into the reasons for wastage in the various equestrian disciplines is imperative if proposals for improvements have to be made. Previous studies on the wastage in Thoroughbred racehorses have been published in United Kingdom (Jeffcott et al. 1982; Rossdale et al. 1985; Williams et al. 2001) and in Australia and South Africa (Bailey et al. 1997; Olivier et al. 1997). Recently, detailed studies on specific aspects of this wastage in Thoroughbred racehorses were published by Boden et al. (2006), Wilsher et al. (2006) and Dyson et al. (2008). To the knowledge of the authors, however, there is little scientific information published on the factors that influence the career of horses used in equestrian disciplines other than racing. Murray et al. (2010) described risk factors for lameness in dressage horses and O'Brien et al. (2005) identified factors associated with the wastage and achievements in competition of eventing horses registered in the United Kingdom. In the latter study it was found that 33.7% of the 2183 horses that were fully registered for the first time in 1999 were not re-registered in the following year. Detailed reasons for the failure of re-registration were obtained from 171 owners: veterinary reasons accounted for 35.1% of failures, 28.1% considered that the horse had underachieved, 24.6% considered the costs to be prohibitive, 4.1% were no longer interested in eventing and 8.2% gave other reasons.

In the Netherlands, as in some other Western countries, dressage and showjumping are by far the most popular equestrian disciplines. Unfortunately, although some concern about the welfare of horses used in competition is expressed, there are no current data on the reasons why horses in these disciplines either have breaks in their career or end their sporting career. Veterinary reasons are often used by critics of equine sport to indicate that the challenges of the sport are inhumane.

The present study was carried out in collaboration with the Dutch National Equestrian Federation, as a pilot study to assess the competition career of sport horses in 4 different equestrian disciplines (dressage, showjumping, eventing and endurance) and identify the main reasons for the interruption or termination of a competition career of sport horses.

The aim of the present study was to identify the reasons for interruption or termination of a competition career in a sport horse population that can be seen as representative for many Western countries where racing is not the main equestrian activity, using the database of the Dutch National Equestrian Federation and to investigate the veterinary reasons in more detail. Additional aims were to investigate the mean age in the first competition year (as competing at young age is alleged to be a welfare issue) and percentage of insurance cover, as (lack of) insurance is supposed to affect longevity. It was hypothesised that wastage for veterinary reasons would differ per discipline but for dressage would be below the 34% found earlier in eventing (O'Brien et al. 2005).

Materials and methods


The horses registered by the Dutch National Equestrian Federation (KNHS) in 2004 were used as a reliable source of information on competition career activity. The year 2004 was selected arbitrarily as the pilot year to allow length of career of horses to be examined for a long enough period (1995–2004) as well as after that year (until the start of the study in July 2009).

For practical reasons, data from around 500 horses was considered to be the maximum number that could be analysed in the available time and approximately 1% of the horses in each discipline were selected at random from the records. This resulted in a study population of 520 horses: 357 dressage, 134 showjumping, 12 eventing and 17 endurance horses.

Set-up of investigation

All competition data from these 520 horses, retrospectively as far back as 1995 and up to July 2009, were collected from the KNHS registration system and supplemented with information derived from an extensive telephone enquiry with the rider registered in combination with the horse in 2004.

Telephonic enquiry

As the KNHS registration system only contained individual competition data of horse-rider combinations, a telephone enquiry was used to obtain additional information from the rider (often also the owner). The questions were asked as ‘open questions’, but classified in closed categories by the interviewer. This approach was chosen as detailed personal discussion with the rider/owner of the horse was considered to be the most reliable way to obtain comparable information from each individual horse in the study group. In order to standardise the method of questioning and facilitate interpretation of replies, all telephone enquiries were performed by the second author over a 10 week period immediately after July 1st 2009.

Data collection

The following data were collected:

  • 1) Horse: name, registration number, date of birth, gender, breed, breeding association number/transponder number (from database)
  • 2) Rider: name and contact details (from database)
  • 3) Competition record: year of first KNHS-registration and year of the last KNHS-registration (= career time), discipline(s) started and level achieved, occurrence of one or more career breaks (defined as a period without any competition events of more than 3 months in dressage and showjumping and of more than 6 months in eventing and endurance) and eventual career end (from datebase)
  • 4) Additional information: was the horse insured in 2004? (from questionnaire)
  • 5) Reason(s) for eventual career break(s)? (from questionnaire)
  • 6) Career end: for what reason was the horse no longer registered for competition? (from questionnaire)
    • • Is the horse known to have another career?
    • • Was there a veterinary reason to end the career and, if so, which?
    • • Was there a rider problem (e.g. time, financial or personal constraints)?
    • • Was the horse sold (domestically or abroad)?
    • • If the horse had died or been subjected to euthanasia - why?

Statistical evaluation

All data are given as mean ± s.d. Data for mean age in 2004 and mean age at first competition were evaluated statistically using a one way ANOVA with a Bonferroni post hoc test where appropriate (P<0.05). Data for total career time in relation to veterinary and nonveterinary causes of career break and career end were evaluated statistically using a Cox-proportional-hazard model (P<0.05). Data for cause of career break and career end were evaluated using a Pearsons Chi-square test (P<0.05).


General data

In 2004, a total of 46,576 horses were registered with KNHS of which 32,896 (70.6%) were in dressage, 11,832 (25.4%) in showjumping, 1058 (2.3%) in eventing and 790 (1.7%) in endurance, indicating the relative importance of these disciplines in the Netherlands. The randomly selected 520 horses comprised 12 stallions, 257 geldings and 251 mares with an average age in 2004 of 7.1 ± 3.2 years (dressage 7.1 ± 3.3 years, showjumping 6.7 ± 2.8 years, eventing 7.3 ± 2.0 years and endurance 9.8 ± 3.4 years). The average age at the first year of registration for competition was 5.3 ± 2.1 years (dressage 5.4 ± 2.1 years, showjumping 4.7 ± 1.3 years, eventing 5.6 ± 1.1 years and endurance 8.3 ± 3.3 years). The mean age of horses in endurance was significantly higher (P = 0.03) than that of horses in the other 3 disciplines and horses in dressage were significantly older in their first year of competition than horses in showjumping (P = 0.002).

The mean age in the study population was not significantly different between genders: stallions 5.9 ± 2.5, geldings 7.0 ± 3.2 and mares 7.2 ± 3.2 years old. Also, mean age at first year of registration was not significantly different between gender: stallions were 5.0 ± 2.2, geldings 5.1 ± 1.9 and mares 5.5 ± 2.2 years old.

Table 1 shows the details of the study population according to the maximally achieved competition level. In the study period, a total of 184 horses (35.4%) reached a basic level (B- or L-class, or class I or II, respectively) and 336 horses (64.6%) an advanced level (M-, Z- or ZZ-class, or class III or IV, respectively). On average, geldings reached a significantly higher competition class than mares: 68.5% of all geldings achieved an advanced level, compared to only 59.8% of the mares (P = 0.04). The number of stallions (n = 12) was too low for statistical power, but 83.3% of these started in an advanced class.

Table 1. Classification according to competition level of the horses in the study population per discipline
  1. N.A. = nonapplicable. Dressage, showjumping and eventing classes consist of starters level (B), light (L), intermediate (M), advanced (Z) and very advanced (ZZ). For endurance there are 4 classes related to the mileage/day (class I: 25–39 km/day; class II: 40–79 km/day; class III 80–119 km/day; class IV: ≥120 km/day).

B (I)24813
L (II)1003945
Basic level 124 47 5 8
M (III)884038
Z (IV)1163141
Advanced level 233 87 7 9

Career ending

A total of 334 horses (64.2%) ended their career between the reference year 2004 and end of the study period on July 1st 2009 (Table 2). Mean total career time was 3.8 ± 2.8 years (dressage 4.0 ± 2.8 years, showjumping 3.6 ± 2.8 years, eventing 5.8 ± 1.3 and endurance 3.2 ± 2.4 years). There were no significant differences in mean total career time between disciplines.

Table 2. Career ends and career breaks in sport horses competing at basic and advanced levels
Total horses184336520
Career end   
 No career end45141186
 Career end139195334
  Veterinary causes324880
  Nonveterinary causes107147254
Career break   
 No career break4887135
 Career break136249385
  Veterinary causes325284
  Nonveterinary causes104197301

Significantly more horses competing at basic than at advanced level ended their career in the study period (75.5 vs. 58.0%, respectively; Table 2, P<0.001). However, in both groups the causes for career ending were comparable: veterinary causes accounted for 23.0% in basic level horses and 24.6% in advanced level horses and nonveterinary causes accounted for 77.0 and 75.4%, respectively. Career ends for veterinary reasons calculated on basis of the total number of horses in the initial population were 17.4 and 14.3% for basic and advanced level, respectively.

The reasons for ending a sport horse's competition career were either rider- (6.3%) or horse-related (93.7%). The horse-related reasons for ending a career included being sold (domestic or abroad) with no follow-up (55.0%), veterinary reasons (25.5%), use for breeding (9.3%), changing to a recreational career (8.6%) and retirement (1.6%). The specific results for each of the 4 disciplines are shown in Table 3. Gender had no significant influence on the reasons for ending a horse's competition career.

Table 3. Reasons for career end in the 4 equestrian disciplines
Total horses3571341217520
Total career end202115413334
Rider reasons 1062321
 Health problem20002
Horse reasons 192109210313
 Sold domestic49130062
 Sold abroad446501110
 Veterinary reasons56151880
Euthanasia/death 23 5 0 0 28
 Tendon injury31004
 Respiratory problem10001
 Skin problem10001
 Back problem20002
 Navicular disease11002
Veterinary problems 33 10 1 8 52
 Tendon injury1230520
 Neck/back problem52007
 Navicular disease31004
 Kidney problem10001
 Dental problem01001
 Patellar ligament problem10001
 Lame (unspecified)41005

When calculated in relation to the initial study population, veterinary reasons caused an end to the career in 15.7% of dressage horses during the study period, in showjumpers this was 11.2%. If calculated in relation to the initial population minus the horses sold (and lost to follow-up), these figures are 21.2 and 26.8%, respectively. For endurance and eventing the numbers are too low to provide a reliable outcome, although there seems to be an indication of a high level of veterinary items in endurance (with 47.1% of the initial population and 50% of the population with complete follow-up ending their career for veterinary reasons).

The age at which a horse starts its athletic career matters in show jumping and endurance. In both of these disciplines for every year of age the horse is older at the beginning of its career there is an increase in the chance of the career ending for a veterinary reason with a factor of 1.1 (P<0.05).

Orthopaedic problems accounted for 63.7% of the veterinary career ending decisions, internal medicine related problems 16.2%, neck and back problems 11.3%, trauma 7.5% and, in one horse, the veterinary reason was unknown (1.3%). Thirty-five percent of the horses that ended their career for a veterinary reason were subjected to euthanasia or died. Overall, in 313 horses with a horse-related reason for career end, euthanasia or death was the reason in 8.9% and in horses of the study population with a career end (334 horses) this was 8.4%.

With regard to the nonveterinary reasons for the end of a sporting career, horses in show jumping had a 2.5 times higher chance of ending their career for a nonveterinary reason than horses in any of the other 3 disciplines (P<0.001), but this was largely due to the large proportion of showjumpers being sold.

Career breaks

One or more career-breaks occurred in 385 horses (74.0%). The reason for the first career-break was either rider- (39.2%) or horse-related (60.8%) (Table 4). Veterinary reasons accounted for 21.8 % of the first career breaks, with tendon problems (27.4%) being the most prevalent. Excluding the rider-related career breaks the veterinary reasons accounted for 35.9%.

Table 4. Reasons for career break in the 4 equestrian disciplines
Total horses3571341217520
Total career break27687814385
Rider reasons 1212352151
 Health problem1720120
Horse reasons 15564312234
 For sale3161020
 Veterinary reasons61171584
Veterinary reasons 61171584
 Tendon injury1650223
 Neck/back problem31004
 Respiratory problem40004
 Digestion problem60006
 Patellar problem20002
 Lame (unspecified)61007
 Muscle problem21003

In the group of 136 horses performing at basic level, veterinary problems accounted for 23.5% of career breaks comparable with the 20.9% of advanced horses with a career break.

Additional information

A total of 485 riders provided information on the insurance status of their horses: 282 (58.1%) horses were not insured, 141 horses (29.1%)were insured for ‘nonsuitability’ and 62 horses (12.8%) for ‘nonsuitability + health care’. Therefore, only 12.8% of the horses were insured for the costs of veterinary services.


This is the first study that has specifically investigated the competition career of sport horses with respect to the age at first competition, career time and reasons for career breaks and career endings.

In a study among British dressage horses, lameness was the most frequently reported health problem with 33% of the horses having been lame at some time during their career, while 8.5% suffered from an internal medicine-related problem (Murray et al. 2010). In dressage horses in the present study, lameness-related problems accounted for 20.0% of horse-related career breaks and 17.2% of horse-related career ending decisions while internal medicine-related problems accounted for 7.7 and 5.7%, respectively, which represents a similar ratio of 3 times as many orthopaedic as medicine-related problems. Although the 2 studies are not really comparable since the number of horses and method of data collection differ significantly, the outcome confirms the statement that musculoskeletal disorders are the most important causes of wastage due to veterinary reasons in the nonracing disciplines, as they are in racing (Riggs 2010).

In racing, the importance of musculoskeletal disorders is even greater. A recent study on days lost from training in 2- and 3-year-old Thoroughbred racehorses showed that if total days lost for training was set at 100%, 80–81% was due to lameness, 5–6% to medical problems, 7–10% due to trauma and 3–5% for unknown reasons (Dyson et al. 2008). Similarly, in a study on factors associated with failure of Thoroughbreds to train and race, Wilsher et al. (2006) showed that of the 6 most common ailments in training in both 2- and 3-year-olds, 5 were injuries of the musculoskeletal system. The findings of the present study are also in accordance with a study by Loomans et al. (2007) showing that Dutch equine practitioners spend most of their clinical time investigating the locomotor system.

In the present study the sale of horses proved to be an important reason for ending competition with their initial rider: 51.5% of horses were sold on and this rose to 67.8% in showjumping horses. This is probably due to the fact that the horse trade is an important factor in the Dutch equine industry. These horses were lost to follow-up, but it can be assumed that most of them will have pursued their athletic career for some time at least. Hence, this category cannot be classified as genuinely ‘career ended’.

None of the horses in the present study died or were subjected to euthanasia during competition itself. All reported trauma incidents had occurred at home, at pasture, in traffic or in the stable. This is contrary to the situation in racehorses where trauma during races is a highly significant factor in ending a career. The risk of fatality in an Australian study was 0.44 per 1000 flat starts and 8.3 per 1000 jump starts (Boden et al. 2006). It would have been interesting to study the number of competitive starts for individual horses in relation to career breaks and career endings. However, these data were not available from the KNHS computer system within the study period.

This study is a first and preliminary investigation into a so far hardly explored area. The authors appreciate that the study suffers from several weaknesses which include the retrospective character and (partial) data collection by telephone enquiry. Further, numbers in the disciplines eventing and endurance are too small to draw any conclusions. Last, but not least, the large number of sold horses lost to follow-up reduces the effective study population as the eventual fate of these animals is unknown. It can be presumed that most will have continued their athletic career for a certain period, but some will have faced a career end due to veterinary reasons within the study period in their new surroundings. Therefore, the percentage of animals (of the entire population) classified as having a career end for veterinary reasons in this study (15.7% in dressage and 11.2% in show jumping) is an underestimation. If it was calculated only on the basis of the fraction of horses with known follow-up during the study period then the figures for dressage and jumping rise to 21.2 and 26.8%, respectively. However, these figures are overestimations because they do not account for horses that competed soundly for a varying period before being sold. Nevertheless, although imperfect, the data generated by this study undoubtedly give a valuable first impression and it can be concluded that the wastage range of 10–25% over the 2004–2009 study period is considerably less than the 34% reported by O'Brien et al. (2005) for eventers in their first year, confirming one of the hypotheses of this study. However, it should be realised that our study is a cross-sectional study with only a limited number of horses competing for their first year and, in the older categories, a selection may have taken place already. The difference between the major disciplines of dressage and showjumping in terms of career ends for veterinary reasons was minor. Numbers for the other disciplines were too low to draw any valid conclusions.

The present study shows that, although not to the extent suspected by those who criticise equestrian sports, veterinary problems indeed play a significant role in career breaks and termination of competition activity (career end) in dressage, showjumping, eventing and endurance with most emphasis being placed on the locomotor system. These results have been reported to the Dutch National Equestrian Federation so that further investigations can be conducted and possible interventions considered. More extensive and detailed studies into the nature, incidence and causes of common injuries are necessary to facilitate the improvement of veterinary care of horses involved in each discipline. In relation to lameness-related problems prevention deserves special attention because of the inadequate healing tendency of many disorders of the locomotor system, especially those concerning tendons and joints. In breeding, a strong and injury-resistant locomotor system has already been the subject of primary interest amongst the KWPN (Royal Dutch Warmblood Studbook) for more than 25 years and has resulted, amongst other things, in a highly significant reduction in the prevalence of navicular disease and bone spavin over the past 2 decades. The roles of rearing conditions (early) training and competition and training surfaces may warrant further studies in the future.

It is the responsibility of the National Federations and the Fédération Equestre Internationale (FEI) to support and coordinate such studies to contribute to the welfare of the equine athlete.


The authors would like to thank the owners, riders and employees of the Dutch National Equestrian Federation (KNHS) for their collaboration. We are also grateful for the support of Prof Derek Knottenbelt, Dr Frits Sluyter and Prof Ab Barneveld in writing the manuscript and to Dr Jan van den Broek for assistance with the statistical evaluations.

Conflict of interest

The authors have declared no potential conflicts.