The industry has encouraged dairy producers to accurately record disease events to monitor incidence and react in a timely manner to any significant increases.
Many goals or benchmarks are also available – both within herds and between multiple herds – to help producers set “alert” levels to signify a health problem or outbreak.
As with all data, we must be confident in the recording method, the identity of animals represented and, most importantly, the diagnosis itself. It is important to ask the question: “Does everyone on the dairy really understand the ‘case definition’?”
For a few diseases, the diagnosis can be straightforward. A retained placenta (RP) is generally easy to spot. Similarly, most people will not question a uterine prolapse diagnosis unless they really do not know cow anatomy.
For other diseases, the case definition may not be as clear. Diseases such as pneumonia or cancer/lymphoma often become “catch-all” diagnoses, with incidence recorded based on some clinical signs. But not all cows with a fever are pneumonia cases. Does she have other signs consistent with the disease? Perhaps the cow does have pneumonia, but have other diseases been ruled out?
If the definition is undefined or different for various individuals on the dairy, how can we really track the true incidence and code it correctly?
In cases where the diagnosis is unknown, second opinion examinations by the veterinarian can be useful to help fine-tune the process and minimize incorrect diagnoses. It is important to take time with your veterinarian to not only develop proper treatment protocols, as most dairies have, but also invest in developing case definition criteria, and follow through with animal health staff training.
Benchmarking between herds
Recognizing case definition challenges within a dairy, it’s even more difficult to benchmark data between dairies. Is a mastitis event coded on one farm the same as on another? If it’s not based on the same diagnostic criteria, then certainly not.
One farm may have a mastitis case definition that only includes clinical signs in the milk and/or udder. If clots, flakes or watery milk are not present, then it is not mastitis. Other farms may use results of routine fresh cow cultures, somatic cell counts or California Mastitis Test (CMT) to make the diagnosis. These are very different case definitions.
As another example, when it comes to ketosis, most consultants will agree we want to keep incidence below 5%. But incidence will depend on our case definition. Do we only code clinical cows, or any cows with an elevated BHBA on a ketosis meter?
So what do we do with the data? If the case definition is clearly defined and we are confident in it, then the data can be very useful to track incidence changes within a dairy over time. However, depending on herd variations in case definitions, using this approach to compare incidence between different operations can lead to some very false conclusions, and establishing benchmark goals is difficult.
One final note: Not every cow will survive. Use these as opportunities to perform necropsies, determining why the cow died and verifying diagnoses. If the cow was diagnosed and treated for pneumonia, but found to have had toxic metritis on post-mortem examination, a review with the animal health crew may be in order.