202 Repeatability of estrus detection patch readings and relationship with breeding success

Abstract

Abstract The objective of this study was to assess how well estrus detection patch readings correlated to successful AI breedings and to characterize the within-cow repeatability of estrus detection patch readings across breeding seasons. Data from the Virginia Department of Corrections beef cattle herds were collated for 7 locations over 7 years, with 2 calving seasons per year. Data from Spring of 2011 were missing. The full dataset contained 19,253 individual animal observations, of which 2,389 observations were omitted for failure to report estrus detection patch data. The relationship between estrus patch reading and pregnancy rate to fixed-time AI and subsequent natural service pregnancy rate was assessed by calculating the sensitivity and specificity of patch readings within each location during each breeding season. A true positive was defined as a patch activating and a cow being confirmed pregnant. A true negative was a patch remaining in the non activated state and a cow being confirmed open. A false positive was defined as a patch activating and a cow being confirmed open. A false negative was a patch failing to activate and a cow being confirmed pregnant. Pregnancy rate to AI sensitivity ranged from 0.2 to 1 with a mean of 0.583. Specificity ranged from 0 to 1 with a mean of 0.525. Subsequent natural service pregnancy rate), ranges in sensitivity (0.197 to 1.00, mean 0.563) and specificity (0 to 1, mean 0.545) were similar. Of the 6,249 animals with usable patch data, 934 never presented with an activated patch and 2,064 presented with an activated patch every breeding season. The remaining 3,251 animals averaged presenting with an activated patch 47% of the time and a non-activated patch 46% of the time. Odds of getting pregnant by AI for cows that always flag with the heat patch were 0.415.