Cow-level prevalence and risk factors for estrus detection inaccuracy in seasonal calving pasture-based dairy cows

• Published in: Theriogenology Vol. 161; pp. 41 - 48
• Main Authors: Kelly, Emmet T., McAloon, Conor G., O’Grady, Luke, Furlong, J., Crowe, Mark A., Beltman, Marijke E.
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.02.2021
• Subjects: artificial insemination; conception; cross-sectional studies; Dairy cow; Estrus; estrus detection; Inaccuracy; insemination; Milk; odds ratio; pregnancy; Progesterone; radioimmunoassays; regression analysis; Risk factor; risk factors; spring; Dairy cow; Estrus; Progesterone; Inaccuracy; Risk factor; Milk
• ISSN: 0093-691X; 1879-3231; 1879-3231
• DOI: 10.1016/j.theriogenology.2020.11.011

High submission rates and pregnancies per AI are essential to ensure compact calving is achieved in seasonal calving pasture-based systems. Estrus detection inaccuracy (EDI) is one area that negatively impacts pregnancies per AI as it increases the inseminations per pregnancy with little probability of conception, while also having the potential to disrupt established pregnancies. The aims of this cross-sectional study were to provide cow-level estimates of EDI prevalence and determine cow-level risk factors for EDI in seasonal calving pasture-based systems. A total of 1071 milk samples were obtained from 984 cows on 19 farms in spring 2018 and analyzed by radioimmunoassay to determine the progesterone concentration at the time of artificial insemination. Based on a validation study on a subset of cows, an inaccurate estrus detection was described as a concentration of progesterone in foremilk of ≥3 ng/ml which corresponded to a composite milk progesterone value of 5 ng/ml. To investigate selected risk factors for EDI, we conducted statistical analyses using two multivariate logistic regression models, stratifying by insemination number (first versus repeat). The overall prevalence of EDI was 4.7% with a prevalence of 3.3% of EDI at first insemination and 14.1% at repeat insemination. Absence of a mounting abrasion (Adjusted odds ratio (AOR) = 3.0) was a significant risk factor for EDI on first insemination while abnormal preceding repeat interval (AOR = 9.5), the absence of an observed standing estrus (AOR = 12.5) and the absence of a mounting abrasion (AOR = 4.1) were significant risk factors for EDI on repeat insemination. The results indicate that cow-level estimated prevalence of EDI in a selection of pasture-based herds was low at first insemination but higher for repeat insemination. It confirms that certain cow-level risk factors existed for EDI, thus providing preliminary evidence for potential future investigation into the targeted use of on-farm progesterone assays in pasture-based herds. •The overall cow-level prevalence of estrous detection inaccuracy was low at 4.7%.•Estrous detection inaccuracy at first AI was lower than at repeat AI.•Lack of mounting abrasions were a risk factor for detection inaccuracy at first AI.•No observed standing estrus and an abnormal preceding repeat interval were risk factors for inaccuracy at repeat AI.