Prevalence of obesity in the equine population of Saskatoon and surrounding area

Abstract

A retrospective study determined the prevalence of obesity and over-conditioning in horses in Saskatoon, Saskatchewan. Body condition score (BSC) was assessed for 290 horses from the Field Service practice at the Western College of Veterinary Medicine. The median BSC of horses was 6; however, 59 (20.3%) horses were classified as over-conditioned, and 24 (8.3%) as obese.

Equine obesity is an emerging health and welfare problem which may go unnoticed or under-reported. Horse owners commonly underestimate the condition of their horse, especially if their horse is overweight (1,2). Veterinarians need to be proactive in identifying overweight horses and educating clients about the risks and management of the overweight horse.

The conditions for which obesity is considered a risk factor include laminitis, insulin resistance, hyperlipemia, and equine metabolic syndrome (EMS) (3). Excessive deposition of fat may contribute to exercise intolerance, osteoarthritis, thermoregulatory inefficiency, abnormal reproductive performance, and the development of strangulating lipomas (1,4).

Body condition scoring offers a quick and objective assessment of a horse’s condition. It can easily be used by horse owners and veterinarians to determine if a horse is over- or under-weight, and to monitor weight loss or gain. The 9-point body condition scoring system described by Henneke et al (5), using Kohnke’s modification (6), is a relatively good predictor of body fat in horses (7). A body condition score (BCS) of 6.8 or greater (on the 9-point scale), estimates total body fat to be greater than 20%, thus identifying a horse in need of a weight loss program (7).

Studies in the United Kingdom have shown that 45% of riding horses were classified as fat or very fat, and prevalence of obesity varied from 27% at the end of winter to 35% at the end of summer (1,8). Similarly, in Denmark, 24% of mature Icelandic horses were considered overweight or obese, and in Australia 23% of horses were considered obese; ponies were more likely to be obese than were horse breeds (2,9). Prevalence of obesity and over-conditioning in the United States has been estimated at 51% in Virginia and 48% in North Carolina (10,11). The USDA APHIS NAHMS Equine Study (12) estimated the prevalence of obesity in horses in the United States at 1.7%, but this is likely an underestimate based on other reported work (10,11). Interestingly, a recent study using animal welfare control data in Sweden showed that premises were more likely to have under-conditioned, rather than over-conditioned horses (13). Little data are available in Canada; in Prince Edward Island, the mean BCS for non-racing horses was reported as 5.7 on a 9-point scale (14). To the authors’ knowledge, there are no studies investigating the prevalence of obesity in horses in western Canada. The primary objective of this study was to obtain prevalence estimates for obesity in the equine population of Saskatoon and surrounding area. A secondary objective was to evaluate potential risk factors associated with over-conditioning or obesity in this population. Age, gender, breed, and BCS were obtained from the medical records of 290 horses seen for routine examination from March to July 2014 by Equine Field Service at the Veterinary Medical Centre (VMC) at the Western College of Veterinary Medicine (WCVM) in Saskatoon, Saskatchewan. Breed, age, and gender distributions were representative of the population seen at the hospital.

The BCSs of horses were assessed by 7 trained equine clinicians using the Henneke 9-point scale in which a score of 1 is very poor condition and a score of 9 is extremely fat (5). Horses with a BCS of ≤ 3 were classified as thin, a BCS between 4 and 6 was classified as ideal, a BCS of 7 was classified as over-conditioned, and a BCS of 8 or 9 was classified as obese.

Breed category was defined by considering the breed, body type, and frame size (Table 1). Horses were considered as a light breed [14.2 to 16 hands high (hh), light build], a pony (under 14.2 hh, light build, known pony breed), a warmblood (> 16 hh, moderate to heavy build, known Warmblood breed), a draft (> 16 hh, heavy build, known draft breed), or unknown (for which there was no record or description of the breed/type allowing for categorization, or the horse did not fit into the other categories as per the description) breed type.

Table 1.

Variables explored in the univariate analyses as potential risk factors for horse being classified as over conditioned

Variable explored	Category	Median BCS (range)	Count (%)	P-value
Age in years (N = 290)	≤ 5	5 (4 to 8)	50 (17.2)	0.84
	6 to 10	6 (3 to 8)	89 (30.7)
	11 to 15	6 (4 to 9)	72 (24.8)
	16 to 20	6 (4 to 6)	37 (12.8)
	≥ 21	6 (3 to 8)	25 (8.6)
	Unknown	6 (3 to 8)	17 (5.9)
Gender (N = 290)	Male	5.5 (5 to 6)	2 (0.7)	0.25
	Female	6 (3 to 9)	133 (45.9)
	Gelding	6 (3 to 9)	155 (53.4)
Exercise (N = 107)	No structured	6 (4 to 8)	44 (41.1)	0.5
	Light	6 (4 to 8)	28 (26.2)
	Moderate	5 (4 to 9)	13 (12.1)
	Heavy	5 (4 to 8)	20 (18.7)
	Very heavy	5 (4 to 6)	2 (1.9)
Activity category (N = 107)	None/retired	6 (4 to 8)	26 (9.0)	0.32
	Pleasure	6 (4 to 8)	23 (7.9)
	Hunter/jumper	6 (4 to 8)	11 (3.8)
	Dressage	7 (6 to 9)	9 (3.1)
	Lesson	6 (5 to 7)	7 (2.4)
	Trial	6 (5 to 8)	6 (2.1)
	Other	4 (4 to 4)	2 (0.7)
Breed type (N = 290)a	Light	6 (3 to 9)	170 (58.6)	0.91
	Pony	6 (4 to 8)	37 (12.7)
	Warmblood	6 (4 to 9)	78 (26.8)
	Draft	7.5 (7 to 8)	2 (0.7)
	Unknown	6 (6 to 8)	3 (1.0)
Housing (N = 84)	Pasture (> 2 acres only)	6 (4 to 7)	43 (51.2)	0.05
	Paddock (< 2 acres only)	6 (4 to 8)	23 (27.4)	0.06
	Daily turnout	6 (4 to 9)	17 (20.2)	a
	Stall only	4 (4 to 4)	1 (1.2)
Grazing access (N = 84)	No access	6 (4 to 8)	24 (28.6)	0.56
	Part of day	5 (4 to 8)	18 (21.4)
	All day	6 (4 to 9)	42 (50.0)

^aReference 1.

Owners of enrolled horses were contacted via e-mail and telephone to participate in a survey regarding the type, frequency, and intensity of their horses’ activity as well as their horses’ turnout and access to grass. Exercise was classified as none, light, moderate, or heavy/very heavy as defined per the US National Research Council’s Nutrient Requirements of Horses (15). The types of activity in which the horses participated were classified as none/retired, pleasure riding only, dressage/dressage plus another activity, hunter/jumper, lesson horse, trail horse, or other. Housing was classified as follows: stall with daily turnout, paddock, pasture, or stall only. Daily access to grass included no access (0 h/day), all day access (24 h/day), or partial day access.

Analyses were performed using commercial statistical software (IBM SPSS Statistics 23.0). For the risk factor analysis, body condition scores were dichotomized into normal/thin (BCS ≤ 6) and over-conditioned/obese (BCS ≥ 7) and logistic regression univariant analyses were used to examine each of the following variables as potential risk factors for horses being classified as over conditioned/obese (BCS ≥ 7): age, gender, exercise intensity, activity type, breed type, housing and access to grass. Age was analyzed as a continuous variable whereas gender was divided into females, males, and geldings. The other factors were analyzed in the categorical variables described.

The median BCS of the 290 horses in the Saskatoon, Saskatchewan area was 6 (range: 3 to 9) on a 9-point scale. Most of the horses (204, 70.3%) had an ideal BCS, but a large proportion of the horses (83, 28.6%) had a BCS ≥ 7 and were over-conditioned or obese. This is consistent with findings in other regions of North America and Europe that indicate obesity is a significant issue in equine populations.

Neither age nor gender influenced the likelihood that a horse would be overweight (Table 1). The median BCS was lower in horses 5 y of age and younger, compared with horses more than 5 y of age. Horses 5 y or younger had a median BCS of 5, compared to a median BCS of 6 for all other horses (Table 1). This is consistent with the physiological differences in young animals. Body condition scoring systems are less accurate when used in young and geriatric horses because of changes in muscling due to growth and sarcopenia, respectively (N. Cymbaluk, unpublished). Median BCS was assessed for all age categories; and even the geriatric group (≥ 21 y old) had a median BCS of 6 (Table 1). It was expected that this group of older horses would also have a lower median BCS, due to physiologic changes in shape and muscle mass associated with aging, but this did not prove to be true in this population.

Breed differences in BCS were not compared as there were too many breeds in this population of horses. Instead, breeds were categorized into pony, light, warmblood, draft, and unknown breed type (Table 1), based on the breed of horse, the height of the horse, and the build of the horse. Horses were classified as unknown when no record or breed or type had been made and that information was not available. Based on breed categorization, there was no association between breed category and likelihood of over-conditioning/obesity (Table 1).

The duration and intensity of exercise also did not appear to be related to the horse’s BCS in this population. However, exercise is an important part of weight management and weight loss programs for horses, and was expected to be associated with the condition of the horse (Table 1). The reason for not finding an association between BCS and exercise intensity might be due to factors such as misclassification of the intensity of the exercise, a relatively small sample size and the potential variability among clinicians scores leading to misclassification of the BCS.

Housing and management was the only statistically significant variable related to BCS in this study. Management survey results were obtained for 84 of the 290 horses (Table 1). Horses that were housed exclusively on pasture were 8.8 times (P = 0.05) more likely to be over-conditioned or obese than horses that were housed in stalls with daily turnout. Thirty-five percent (8/23) of horses housed exclusively on pasture were classified as over-conditioned or obese, whereas only 6% (1/17) of horses housed in stalls with daily turnout were classified as over-conditioned or obese. This finding is likely due to owners being able to regulate the quantity of intake when horses are not exclusively housed on pasture. Horses that were housed in paddocks or horses that were housed in stalls exclusively were not significantly more likely to be over-conditioned or obese than horses housed in stalls with daily turnout (Table 1).

Seasonal variation in weight has been demonstrated in horses living outdoor, with the largest variation seen in horses on pasture (11). In the earlier months of the current study pasture may not have been adequate to meet the horses’ nutritional needs, whereas in the later months of the study pasture may have provided excess energy. Seasonality is a limitation of this study.

Body condition is determined by a balance between feed intake and exercise. Comparing only one factor of the equation does not represent the whole picture. For example, a horse receiving heavy exercise but fed in excess of their energy requirements could still be overweight. On the other hand, a horse receiving no structured exercise and an adequate amount of feed could be an ideal BCS. This concept is illustrated well by Thatcher et al (10) as they identified the feeding of 0 to 1.4 kg of grain as a risk factor for obesity. This volume of grain most likely represented a “treat” in addition to the horse’s daily diet, thus increasing their energy intake above their expenditure. Feeding larger volumes of grain was not identified as a risk factor by this group; most likely, this practice is used in horses performing intense exercise as a way to meet energy demands. When assessing a horse’s risk for obesity it is important to consider diet and exercise together, rather than separately.

Our findings align with previous work and demonstrate that 28.6% of the Saskatoon area horse population was overweight or obese. Obesity is a disease that can easily be treated by owner education and changes in horse management practices. While susceptibility to obesity and EMS may vary between breeds, the American College of Veterinary Internal Medicine consensus statement on EMS emphasizes that EMS and obesity can be prevented through good management practices (15). The key to controlling obesity is recognition and client education on management changes, such as diet and exercise, to promote weight loss. Owners commonly underestimate their horses’ body condition score, particularly if their horse is fat (1,2). It is important for veterinarians to educate clients on proper body condition and weight management to reduce obesity and associated health risks.