Skip to main content
SLU publication database (SLUpub)

Conference abstract2008

A new reproductive performance indicator for dairy herds adjusted for voluntary waiting period

Löf, Emma; Emanuelson, Ulf; Gustafsson, Hans


Introduction There are a variety of measurements and indices that are used to monitor overall reproductive performance of dairy herds. The most commonly used indicators are time interval measurements such as calving interval, days to first service and days to conception. these is that they can only be calculated for animals that either have a next calving or have inseminations or natural services, thus introducing possible selection bias. This problem can be alleviated by using survival analysis. The 100-day-InCalf-rate (100IC) is an increasingly popular indicator that utilizes this methodology. However, most indicators, including the 100IC, do not take into account the different strategies that are applied at farm level, e.g. the herd’s voluntary waiting period (VWP). The aim of this study was to use survival analysis to construct a new reproductive performance indicator that is adjusted for the herd’s VWP and to compare it with 100IC. Material and Methods We studied all animals that calved from 2004-09-01 to 2005-08-31 in 512 Swedish dairy herds registered in the Swedish Official Milk Recording Scheme (SOMRS). The herds were randomly selected from all herds (n=2728) that had more than 45 milking cows. Registrations were extracted from the SOMRS and from the AI-recording system for each animal. A dataset was built containing calving date and, when known, the date of the consecutive calving, date of pregnancy checks, the date the cow was culled or sold and date of AI or natural service. From this dataset days from calving to conception or to censoring was calculated. A survey asking for the VWP was sent to the herds, with a resulting response rate of 59%. Because herd VWP was not available for all herds, it was also estimated as the days postpartum by which 5% of the cows in the herd-year had received a first insemination. The correlation between the reported and the estimated VWP was calculated. The proportion of cows pregnant 30 days after the herd VWP (PPW+30) was estimated using survival analysis, stratified by herd and accounting for potential censoring, as was also the 100IC. The correlation between the indicators was calculated. Results The reported VWP had a mean of 58 days and an inter-quartile range (IQR) between 50 and 60 days. The estimated VWP for the herds had a mean of 51 days and an IQR between 45 and 57 days. The correlation between the reported and the estimated VWP was 0.51. The new performance indicator PPW+30 had a mean of 0.19 and an IQR between 0.12 and 0.24, which means that on average 19% of the cows at a herd where pregnant 30 days after the herd’s VWP. The 100IC had a mean of 0.30 and an IQR between 0.21 and 0.38, which means that on average 30% of the cows at a herd where pregnant 100 days postpartum. The correlation between PPW+30 and 100IC was 0.73. Conclusion Estimation of the PPW+30 on herd level was feasible and the indicator performed well. The PPW+30 is related to, but slightly different than, 100IC and may be useful to assess other aspects of the herds’ fertility status

Published in


16th International Congress on Animal Reproduction