|
Centre for Genetic
Improvement of Livestock Survival Data, At Its Best
|
|||
|
|
Canadian systems
that measure and record dairy cattle longevity are being improved at the
University of Guelph.
Research associate Dr. Paul Boettcher, Centre for Genetic Improvement of Livestock, is working to find the best approach for the statistical analysis of dairy longevity data. Longevity is an important trait for dairy cows' genetic evaluation, by organizations such as the Canadian Dairy Network, that compile a range of records from Canadian dairy farms and statistically predict the genetic value of animals in a population. A Longer Life A longer life span is associated with lower production costs and decreased culling. Raising a cow is expensive; it can cost a producer up to $2,000 to care for a calf until it's old enough to be milked. Improved longevity makes this upbringing more economical and leads to increased milk production, because older cows produce more milk than younger cows. Longevity is based on how long each animal lives -- expressed as the number of lactations the cow survives -- from which valuable herd life information can be determined. Longevity is measured in a variety of ways. "In Canada, longevity is assessed as a binary trait," says Boettcher. "It's based on yes or no answers as to whether the cow survived through the lactation. This system has its advantages in that it's simple to implement...but it has some theoretical weaknesses." However, Boettcher says Canada's current method for assessing longevity doesn't account for the differences in days survived within a lactation. Also with this system, records of survival are not added to the database until a reported re-calving or culling. This can delay the results of longevity evaluation for up to a year or more. Finding An Alternative Boettcher says that an alternative longevity measuring system, called "survival analysis," is better than Canada's current system. It estimates a producer's need to cull an animal at any period throughout its lifetime. Boettcher believes survival analysis may be a more appropriate way to analyze survival because it's a more precise statistical model. It can account for differences in the number of days lived between cows that survive the same number of lactations. Also, the survival model includes data on the number of days a cow has survived even while it's lactating, instead of having to wait for the start of lactation or culling. This makes greater amounts of information available earlier, because it's immediately recorded. Boettcher doesn't believe the survival analysis approach on its own is ideal, either. So he's devising a more appropriate method of data analysis that combines existing methods with survival analysis. "If survival data is analyzed using the two different approaches, there will be a clear difference in longevity results," says Boettcher. "This makes it worthwhile to change the system." This research was funded by the DairyGen (formerly the Cattle Breeders' Research Council). |
