Proper management of dairy cows during the transition period is critically important for both the productivity and profitability of any dairy herd. Nutrition, housing and management during this period all have a direct influence on the health, production and reproduction of cows in the subsequent lactations.
Health disorders occurring in the immediate postpartum period are not only inter-related but also have significant negative consequences for the cow as it moves into lactation. For that reason, best practices should be implemented during that time to minimize the risk of, if not completely prevent, such postpartum health disorders.
Despite best efforts to minimize the risk of postpartum health disorders, dairy producers are not able to prevent all of these from occurring. Even with optimal management, it is not uncommon to see a significant number of cows succumbing to various metabolic and infectious disorders in the first few weeks of lactation.
For that reason, efforts need to be focused not only on prevention but also on the identification of those individual cows at risk for or experiencing postpartum disorders.
The identification of those cows does not come without its challenges. One symptom of a cow experiencing postpartum health issues is being “off feed.” Similarly, cows with low feed intake during the transition period, particularly in the week leading up to calving, are at a greater risk of experiencing postpartum health disorders.
Related to this, while a decline in feed intake during the week prior to calving has been described as common across cows, researchers have shown such a decline in intake does not occur for those cows that remain completely healthy across the transition period.
As such, monitoring intake levels of individual cows across the transition period would tell us much about their health status. For individually housed cows, that may be an option; however, for those group-housed herds, such monitoring is not possible. Further, not every health disorder is always associated with a decline in feed consumption.
As result, we must rely on other metrics to help us identify those cows at risk for or experiencing health disorders. Those metrics may be indicative of feed intake level or may be general indicators of sickness behavior or other changes in physiology. To that end, much recent research has been focused on monitoring behavioral metrics.
While visual behavioral monitoring of individual cows may be difficult, particularly in large herds, there are various commercial precision technologies that can be used on-farm to automatically capture cow behavior (e.g., eating, ruminating and lying behavior).
Such technologies not only allow us to monitor many cows at one time, they may also be more objective and repeatable in their measurement of various behaviors.
One of the opportunities with the use of precision behavior monitoring technologies is to identify those animals prior to calving that may be at risk of succumbing to a health disorder in early lactation. Initial work in this area was done by researchers at the University of British Columbia, who demonstrated cows with reduced feeding time during the week before calving are at higher risk of postpartum health concerns such as metritis and subclinical ketosis.
In similar work, we at the University of Guelph demonstrated cows with reduced rumination time during the week before calving are at higher risk of developing subclinical ketosis post-calving. In our study, we did not see any association in prepartum lying behavior with their health status post-calving.
Collectively, this indicates identifying at-risk cows is most effective using those behaviors indicative of feed intake levels, such as feeding and rumination time.
If and when those cows are identified, producers can investigate means to either manage those individual cows differently or manage the group differently to minimize the risk of that between-cow variability in cow behavior (e.g., feeding management to improve feed access and consistency, minimizing social stressors by reducing stocking density, minimizing pen moves and separating first-lactation heifers from mature cows).
It is noteworthy that, in these studies, cows developing health disorders post-calving continued to demonstrate much lower feeding and rumination activity compared to cows that remained healthy. Further, we also demonstrated that lying behavior actually increased post-calving in those cows with subclinical ketosis.
This suggests these behaviors may not only be indicative of reduced intake but also potential indicators of sickness behavior in these animals. This provides the opportunity, through monitoring, to identify those cows experiencing illness early in the post-calving period.
There is good opportunity to look for early deviations in behavior to identify cows going to succumb to illness before any other clinical symptoms appear. While monitoring cows milked in a robotic system, our University of Guelph research group found deviations in daily rumination time typically occurred earlier than any other symptoms, including changes in milk yield.
For example, we found daily rumination time started to decline already eight days prior to diagnosis and treatment of a displaced abomasum, while milk yield only started to decline four days prior to diagnosis.
Similarly, we found daily rumination time started to decline six days prior to diagnosis of subclinical ketosis, while milk yield only started to decline four days prior to diagnosis. We also found bodyweight started to decline six days prior to subclinical ketosis diagnosis.
This similar timing to the drop in rumination time provides further evidence the rumination drop in these cows is likely linked to a decline in feed intake.
These studies collectively indicate the use of precision technologies for monitoring of cow behavior, both prior to and after calving, may be very valuable for dairy producers.
By monitoring individual cows and how they vary from the rest of the herd or change from their own baseline, those experiencing health concerns or other physiological changes can be identified, and treatment protocols or other interventions may be implemented earlier. 
-
Trevor DeVries
- Associate Professor
- University of Guelph
- Email Trevor DeVries
