Foundational Review of U.S. Female Fertility Trait Evaluations

Abstract

The Council on Dairy Cattle Breeding provides four female fertility evaluations for U.S. dairy producers: daughter pregnancy rate (DPR), cow conception rate (CCR), heifer conception rate (HCR), and early first calving (EFC). These evaluations were first introduced in 2004 for DPR, 2009 for CCR and HCR, and 2019 for EFC. Currently, these traits are expressed on six different breed bases: Ayrshire, Brown Swiss, Guernsey, Holstein, Jersey, and Milking Shorthorn. Over time, the data and methods used to calculate these traits have evolved in response to changes in availability, recording practices, and management systems. In recent tri-annual evaluations, unexpected season fluctuations have been observed in the Sire Estimated Breeding Values (EBV) of recently born bulls. The objective of this project was to identify the cause of these fluctuations and implement changes to improve stability across evaluations. In collaboration with the USDA Animal Genomics and Improvement Laboratory, this project also involves the research and development of a potential new trait to be added to the fertility evaluation, First Service to Conception (FSC), and re-estimation of genetic parameters for all five traits. Comprehensive tests were conducted to refine models, pre-adjustments, and data edits, including the use of both truncated and full datasets. Although data truncation showed promise in mitigating historical biases, it introduced higher variability in smaller breeds (Guernsey and Ayrshire). Additional tested changes included stricter calving year restrictions, improved data extraction procedures, updated CCR and HCR pre-adjustments, the inclusion of a days-in-milk covariate at first insemination for CCR and FSC, and the addition of a random herd-by-sire effect. Tests also examined whether modeling days open to pregnancy rate as a linear or non-linear trait, modeling traits as uncorrelated, performing unweighted analyses, or stricter convergence criteria of the traditional evaluation mixed model equations solver were appropriate. While the findings suggest that current methodologies provide a robust foundation, ongoing work is required to address the persistent slight negative trends reported in young bulls, where the underlying causes remain unclear. The project team is well-positioned to further enhance the stability of female fertility trait evaluations for U.S. dairy producers.