Reproductive performance of Thoroughbred mares in the Waikato region of New Zealand: 2. Multivariable analyses and sources of variation at the mare, stallion and stud farm level.

Abstract: The objective of this study was to utilise multivariable statistical methods appropriate for clustered data to identify mare-related explanatory variables that significantly affected the reproductive performance of Thoroughbred mares in the Waikato region of New Zealand. In addition, we aimed to determine the relative contribution of the mare, stallion and stud farm to reproductive performance. Methods: A prospective cohort study was performed involving five stud farms in the Waikato region of New Zealand during three consecutive breeding seasons (2006-2008). A total of 1,482 individual mares contributed 2007 mare years and 3,402 oestrous cycles over the three breeding seasons. Reproductive performance was measured using three parameters; (a) first-cycle pregnancy rate (FCPR), (b) end-of-season pregnancy rate (SPR), and (c) the start-of-mating to conception interval. Results: When controlled for the effects of serving stallion, stud farm and year of study the only significant mare-related variables included in the final models of FCPR, SPR and conception interval were the age of the mare and her reproductive status (classified as dry or foaling). Advancing mare age significantly reduced reproductive performance regardless of reproductive status and foaling mares had significantly poorer reproductive outcomes compared with dry mares when controlled for age. For each additional increase in year of age, the FCPR was reduced by a factor of 0.94 (95% CI=0.92-0.96) and the SPR was reduced by a factor of 0.91 (95% CI=0.88-0.93). Mares older than 14 years of age took longer to conceive after the start-of-mating compared with younger mares. The daily hazard of conception for mares 14 years and older was 0.64 (95% CI=0.47-0.83) times less than mares younger than 9 years of age. Determining the relative contribution of the mare, stallion and stud farm to the FCPR indicated that 95.9% of the variation was at the mare level, 4.1% was at the stallion level and 0% was at the stud farm level. For the SPR the variance components indicated that 92.5% of the variation was at the mare level, 6.7% was at the stallion level and 0.8% was at the stud farm level. Conclusions: The reproductive performance of Thoroughbred mares in the Waikato region of New Zealand is influenced by two main mare-related factors; the age of the mare and her reproductive status (dry or foaling). The majority of variation in reproductive performance was associated with mare-level factors and the contribution of the stallion and stud farm was relatively minor.