How do features of dressage arenas influence training surface properties which are potentially associated with lameness?

Results from a previous study indicated that there are specific arena surface characteristics that are associated with an increased likelihood of lameness in dressage horses. It is important to understand what modifiable arena factors lead to these detrimental surface characteristics. The aim of this study was to describe the use of training surfaces and arenas for United Kingdom dressage horses and to investigate any relationships between arena/surface variables and detrimental surface characteristics. Data from a questionnaire returned by 22.5% of all 11,363 registered members of British Dressage were used for the study. Univariate and multivariable logistic regression models were developed with each of the previously identified surface characteristics as dependent variables. Respondents reported that the majority of arenas were privately owned, sized 20 × 40 m and had a sand and rubber surface. The results indicated that wax-coated and sand and rubber surfaces were associated with less detrimental surface properties than sand, sand and PVC, woodchips or grass. Woodchips were most strongly associated with the detrimental characteristic of slipping, and sand with tripping. The findings indicated that any arena surface should have a base, with limestone the recommended surface, and that crushed concrete was best avoided. This information supported previous studies in racehorses that indicated that surface maintenance is essential, especially when many horses are using an arena daily. Problems were less likely if an arena was privately owned.

Identification of risk factors for lameness in dressage horses.

The aim of this study was to describe the prevalence of illness and lameness at different anatomical sites in registered United Kingdom dressage horses and to identify risk factors for lameness. A questionnaire was sent to all 11,363 registered members of British Dressage in 2005, with one questionnaire assigned per horse. Four multivariable logistic regression models were developed for each section of the questionnaire. A final mixed effects logistic regression model was developed which combined the results from all prior models. Owners reported that 33% of horses had been lame at some time during their career, with 24% of these within the previous 2 years. A number of factors were associated with the occurrence of lameness in the last 2 years, including age, height, indoor arenas, horse-walkers, lunging (as protective), back problems, arenas that become deeper in wet conditions and sand-based arenas. These factors were included as variables in a final model to provide information for selection of horses, development of safer arenas and more effective training regimens to minimise the onset of lameness.