Adsa Logo White Adsa Title White

Population approach for determining behavioral thresholds associated with health status during transition.

P. Pinedo

Events

06-22-2020

Join P. Pinedo on this page for a live text chat!
6:00 PM - 8:00 PM GMT

Abstract:

M30
Population approach for determining behavioral thresholds associated with health status during transition.
D. Manriquez, P. Pinedo*. Colorado State University Fort Collins, CO.

Monitoring of behavioral variables has been used to assess individual health and estrus presentation in dairy cows. The objective was to evaluate population dynamics of behavioral variables during postpartum to determine cut-off values associated with the subsequent occurrence of health events up to 21 DIM. Cows (n = 1,350) in an organic certified herd in CO, were affixed with CowManager tags (AABV, the Netherlands) for monitoring of rumination, eating, and active times (min/d). Health events were retrieved from on-farm software (PCDart, Raleigh, NC). Cows were categorized as healthy (HT) or diagnosed with at least one health event (DZ) within 21 DIM. Explanatory variables included behavioral variables summarized in 20 min intervals at 0, 1, and 2 DIM. Logistic regression and ROC curves analysis were used to calculate cut-off values maximizing sensitivity (Se, %) and specificity (Sp, %). Calving season, parity, and dystocia were included in the models as controlling variables when P < 0.1. Overall, 17.2% of the cows were in the DZ category. Increments of the behavioral variables were associated with lower odds of disease presentation. For each 20 min/d increment in rumination at 0, 1, 2 DIM the odds of DZ decreased by 0.95 (95% CI: 0.92—0.98), 0.92 (0.9—0.95), and 0.90 (0.87—0.92), respectively. For each 20 min/d increment in eating time at 0, 1, 2 DIM the odds of DZ decreased by 0.89 (0.86—0.92), 0.85 (0.82—0.89), and 0.84 (0.8—0.88), respectively. For active time, the odds of DZ decreased by 0.96 (0.93—0.99), 0.94 (0.91—0.96), 0.95 (0.92—0.98), respectively. Eating time at 1 DIM had the greatest area under curve (AUC = 0.69), followed by eating (AUC = 0.68) and rumination time (AU = 0.64) at 2 DIM. The cut-off values that optimized Se and Sp at these points were 233.5 (Se = 65; Sp = 61), 285.9 (Se = 63.2; Sp = 61.3), and 438 (Se = 61.3; Sp = 52) min/d, respectively. AUC for active time was < 0.6 and Se and Sp < 50%. This study suggests that rumination time and eating time have discriminative value at early lactation for cows at risk of health disorders and these assessment points could be used for developing targeted prophylaxis after calving.

Keywords: behavior, disease, performance.