Breeding for Health
Recording herd health data could lead to strategies that would let you choose economically important genetic traits.
In the last few years, dairy producers and the breeding industry have recognized health, well-being and long-term fitness of cattle as economically important. At the recent semi-annual open industry session of the Canadian Dairy Network's Genetic Evaluation Board, many of the papers presented and discussed focussed on reproduction and health traits. This work clearly indicates that we need to do a better job of recording breeding and health events. New and innovative ways of analysing these data would provide useful genetic information for dairy producers.
Diseases cost you in many ways. There are costs of treatments, disposal of milk from treated cattle, lost milk production and death loss, for instance. As well, you will likely be forced to rely less on treatments and drugs in the future. There is also concern that ignoring the genetic side of health traits would cause the fitness and health level of our cattle to generally decline. These are all compelling reasons to take a look at how we can collect health information and use it to the industry's advantage.
Providing a start is DairyComp, a computerized herd management system available through dairy herd improvement [DHI] organizations. An increasing number of dairy producers have been using it since 1998 to record health and disease data. As producer participation increases, we have greater potential for using this information. Canadian DHI is modifying its database to collect and organize health data nationally.
A group of University of Guelph researchers, led by Jalal Fatehi, Larry Schaeffer and Janusz Jamrozik, have done some analysis of existing health trait data in the DHI system. They examined data quality and did preliminary analysis prior to genetic evaluations for resistance to health problems.
The researchers looked at data from 2,251 herds from Ontario and Western provinces. After eliminating duplicate records and data with no animal identification, they had just over 64,000 records of disease occurrences in 33,981 animals.
They looked at the frequencies of 14 health problems and distribution by breed, lactation number and year of recording. The table shows relative frequencies of disease conditions. The most common problem reported was mastitis, followed by cystic ovary, retained placenta and lameness.
When the researchers looked at the frequencies of disease conditions according to lactations one through five-plus, the most commonly reported-mastitis, cystic ovary and retained placenta-tended to be consistent from one lactation to another. The relative frequency of some other disease conditions, however, appeared to vary from one lactation to another. Milk fever incidence increased over lactations, while first-lactation cows had a higher frequency recorded of lameness, metritis and dysentery.
Provided it includes proper animal identification, this kind of information in the DHI database means health-related data can be linked to pedigree information, production, calving ease and fertility data. If there are relationships among the recorded traits, a proper analysis will make it possible to develop selection programs to improve dairy cattle profitability. There are other databases of health information but at present they lack the linkage with DHI and the consistent identification that goes with it.
Research reported a few years ago by the University of Wisconsin's Dr. Nate Zwald showed that while heritabilities are low, some of these health and disease conditions have enough genetic variation to make it possible to select for improvement.
The condition of cystic ovaries, for example, has a low heritability. Years of selection against cystic ovaries in some Scandinavian countries, however, has reduced its incidence in their cow population. Progress can be made but to improve something you have to measure it first.
In Canada, we have problems with the quality of health data. Some producers keep complete data on all cows in their herds, recording all health and disease events. At least initially, these herds will provide most of the data for genetic analysis and developing benchmarks for these traits.
On the other hand, a large number of herds have no recorded data for these traits. Since no herd has perfect health status, it is clear that these producers don't record health and disease information.
The most difficult data to deal with and interpret come from herds that sporadically record only some information. It is difficult to tell if their data are complete or should be used in the analysis. If data are incomplete and used in the analysis, they can bias results.
This sort of detail has to be recorded daily and not left to recollection at some later date. Partial recording of these traits is worse than no records at all.
Integration of health events into total herd management is essential. Benchmarks and trends have to be analysed in a herd situation to show what is improving or what is getting worse and in need of attention. This sort of information needs to be integrated with production and reproduction data for the herd to make it economically useful to you.
Because this information will benefit all herds, the industry may decide to pay co-operating producers to accurately record this information. These producers will also find it useful for herd management, ultimately making them more profitable.
To be consistent, however, guidelines are needed on what to record. We need to resolve how these guidelines are going to be determined. Finally, a validation process is needed to ensure that the data we use in genetic evaluation truly represent the traits we want to improve.
For example, we need a uniform understanding of what constitutes a "case" of metritis or a "case" of milk fever. Some producers record only conditions requiring treatment. Since they have to record treatments, that is when they record a disease condition. A number of conditions, such as a retained placenta, lameness or ketosis, may at times go without treatment. This information also has to be recorded and passed on to the DHI program.
Summarizing health and disease data in the milk recording system is a start towards understanding and using this information to advantage. Recording health data of value to dairy improvement will require a joint effort of producers, milk recording organizations, practising veterinarians and animal breeders. The result would be development of breeding strategies that would let you make informed decisions on economically important traits for your herd.
Health Problem Frequencies in CanWest DHI Data Set
|Author:||Blair Murray - Dairy Genetic Improvement Specialist/OMAFRA|
|Creation Date:||17 May 2007|
|Last Reviewed:||3 June 2010|