Thursday, September 11, 2014

Dry cow therapy


Starting a blog is one thing, maintaining it is yet another thing. The waiting time on airports when travelling is then an excellent reason to write another item for a blog. Many times the travel is done around interesting meetings in which some inspiration for a blog entry is fond. Friday I was at the DACH tagung which was held at the University of Zurich. The DACH tagung is a symposium for German language veterinary epidemiologists and the theme of this year was Tiergesundheit und Ökonomie. I had the honour to be the closing keynote speaker and was asked to give a presentation on economic analysis for different levelsof decision making. As always you can find the presentation at slideshare. Besides a classification of different types of animal diseases, levels of decision making and the decision makers involved I presented a few examples. One example was on dry cow therapy. This is an intervention aimed at a typical production disease and the, already 20 year old discussion on selective vs blanket dry cow therapy. We did publish a paper in 2007 and that was used as example. In another paper, published a little earlier we concluded the same. Note that the methods used in both papers differed (Monte Carlo simulation modelling and decision tree analysis), showing once more that there are more methods possible to tackle the same problem.

The papers showed that, on average, selective dry cow therapy is the cheapest option at the herd level, but a farmer does accept more clinical mastitis when applying selective dry cow therapy.  However, currently there is regained interest in dry cow therapy because of the discussion on the (preventive) use of antibiotics. Dutch dairy farmers have to reduce the use of antibiotics and drying-off antibiotics make up quite a large proportion of the antibiotic use. The question is then how to optimally use the limited amount of antibiotics a farmer may use. That is a typical optimization model and a student (Luuk Maas) dedicated his MSc thesis to this problem. He linked his linear programming model to data from a large trial of the AnimalHealth Service in Deventer.

The basic background of this work was that cows with an increased SCC before drying off (primiparous cows > 150.000 cells/ml and multiparous cows > 250.000 cells/ml) will be dried off with antibiotics. For the other cows you want to think about the most optimal level of use of antibiotics. It did show that it was most efficient to not treat all cows with antibiotics, even if there was no constraint (something we knew already), but Luuk also showed that when there is a reduction in the use of antibiotics of more than 35 %, there will be costs involved for the farmer. This limit is dependent on the farm.

This type of knowledge can be used when further setting thresholds for responsible use of antibiotics in dairy cattle. Yes: economics are important to support decisions. Moreover, decisions regarding production diseases are not alone a matter of the farmer. Since society and dairy processors are more and more interested  in production diseases, they also should take notice of the potential that animal health economics can offer them.