Article Text

Download PDFPDF

Short Communication
Prevalence of obesity in a population of horses in the UK
  1. H. M. Stephenson, BVMedSci,
  2. M. J. Green, BVSc, PhD, DCHP, DipECBHM, MRCVS and
  3. S. L. Freeman, BVetMed, PhD, CertVA, CertVR, CertES, DipECVS, MRCVS,
  1. School of Veterinary Medicine and Science, University of Nottingham, College Road, Sutton Bonington, Leicestershire LE12 5RD
  1. E-mail for correspondence sarah.freeman{at}

Statistics from

OBESITY predisposes to important conditions in horses, such as laminitis and equine metabolic syndrome (Treiber and others 2006). Wyse and others (2008) reported a prevalence of obesity of 45 per cent in pleasure riding horses in Scotland, but the prevalence of obesity in the wider UK horse population is currently unknown. This study assessed the prevalence of obesity in a selected population of leisure horses. The relationships between the estimated body condition score of the horses, owners' perceptions of body condition and owners' attitudes towards equine obesity were investigated.

Clients were selected from the client database of an equine veterinary practice. The inclusion criteria were that the client was situated in Leicestershire or Nottinghamshire, had joined the practice in the past five years, and was not a professional horse owner. Professional owners were defined as owners of breeding, livery or riding stables, or those competing at a professional level. A questionnaire, body condition scoring sheet and explanatory letter were sent to all eligible clients during August and September 2008. All the owners were invited to an evening of seminars at the conclusion of the study, discussing obesity and related health problems, and the outcomes of the study.

The questionnaire consisted of closed, multiple choice questions concerning the type, management and feeding of the horse. The owners were asked to score the body condition of their horse, using a scoring sheet (Anon 2010). Descriptive data analysis was conducted by plotting the condition scores against feeding and exercise management to evaluate trends.

A total of 506 questionnaires were distributed, of which 160 (31.6 per cent) were returned. Table 1 shows the distribution of body condition scores reported by the owners; 20.6 per cent of the horses were assigned scores that were overweight (score >3). Grass was the main source of forage (49 per cent of horses), and coarse mix was the main source of concentrates (50.6 per cent). Only 10.6 per cent of the horses received no concentrate feed. Forty-nine per cent of the horses were exercised for up to five hours per week, 42 per cent for six to 10 hours, and the remainder were exercised for more than 10 hours per week. No significant trends were identified for age, breed of horse, type of work, concentrate or forage feeding or hours of exercise per week with respect to body condition score. Overall, 28.1 per cent of the horses had had lameness (excluding laminitis), and 11.3 per cent had had laminitis in the past three years. No significant associations between lameness and body condition score were identified.

Table 1

Body condition score of 158 horses, as assessed by their owners (0 Very poor, 5 Very fat)

On the basis of the results from this initial questionnaire, conventional sample size estimates identified that a sample of 14 horses would be needed to make a comparison between owner-scored and researcher-scored horses (assuming a detectable difference in mean body condition score of ≥0.5, power = 0.80 and significance P=0.05). Having given informed consent, 15 clients were randomly selected ( for a visit and interview, scheduled during September and October, during which the researcher body condition scored the horse, and asked the owner five structured questions regarding the horse's weight control and diet. The researcher was blinded to the owner's body condition scoring of the horse. The body condition scores assigned by the owner and the researcher were compared using the kappa statistic (Minitab v15).

None of the 15 clients that were visited had body condition scored their horse previously, although 12 of them reported using other methods of weight assessment. For these 15 horses, the mean (median) condition score assigned by the researcher was 3.8 (4), significantly higher than the score assigned by the owners 3.1 (3) (Kruskal-Wallis test, P=0.03). Eight of the owners who were visited had scored their horse at least one grade lower than the researcher. There was poor agreement between the body condition scores assigned by the owner and the researcher (κ=0.04).

In this study, the questionnaire completed by the owners reported the prevalence of overweight horses as 20.6 per cent, which is markedly lower than the prevalence identified by Wyse and others (2008) in pleasure horses in Scotland. However, comparison of the body condition scores assigned to a subgroup of the horses by their owners and by a researcher showed that many owners underestimated their horse's condition score. This may have been due to lack of experience using the scoring system or owner bias. If this underestimation error were to be extrapolated to the questionnaire data, then the prevalence of overweight horses would be 54.1 per cent, a prevalence similar to the prevalence of 45 per cent reported by Wyse and others (2008).

This pilot study identified a relatively high prevalence of obesity in the selected population of horses, and that underestimation of body condition score by the owners was likely. A larger study would be useful to establish the prevalence and risk factors for obesity in different populations of horses across the UK.



  • Provenance not commissioned; externally peer reviewed

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.