Who mooves where?
Writing about web page http://www.biomedcentral.com/1746-6148/7/31
As a consequence of The BSE outbreak in the late 1980s in the UK, the UK government set up a computerised system for recording the births, deaths, and movements of cattle. Spurred on by the realisation that the 2001 foot and mouth disease outbreak was initially spread around the country by people buying infected animals, these data have been made available to researchers. This means that we now have a comprehensive picture of the lives, locations, and deaths of cattle in the UK for over a decade – some 43 million animals, and 157 million animal movements!
This is a great resource for epidemiologists, as every trade in live animals is risky – if I sell you cattle from my farm, then if there are infectious diseases circulating on my farm, there’s a chance that you’ll buy infected cattle from me, and thus I have transmitted infection from my farm to yours. But, trading animals is essential to the farming industry in the UK, so there will always be a balance to be struck between economics and infectious disease control.
So what do all these data tells us? Well, for the full details, have a look at my recent paper in the journal BMC Veterinary Research, but since it’s open access, I can show you a couple of the figures here. In common with other European countries, British cattle tend to not move that often during their lives, and to not move great distances, although a few cattle really rack up the miles:
Of course, the other thing you can do with a decade’s worth of data is to look at trends across time. If we look at the number of movements per month, you can can clearly see the troughs caused by foot and mouth disease in 2001 and 2007:
(The peak in 2000 is an artifact of “Cattle Count 2000”, not a real surge in movements)
A seasonal trend is pretty clear, too – more animals are moved in the spring and autumn, co-inciding with the main calving times. Beyond that, there isn’t much change over time; if we look at the numbers of cattle each farm moves on and off in a year, there was some increase between 2002 and 2005, but things have rather levelled off since:
(the boxes are median and quartiles, whiskers are the central ninety-five percentiles, the solid lines are the means)
This, actually, shows the value of taking a long-term view: some scientists looked at the changes between 2002 and 2005 and concluded that there was a real increasing trend in movement numbers (and some associated disease-risk measures); we can now see that this was only a short-term issue. These data, as well as providing an insight into the structure of cattle farming in the UK, might well be the basis for a predictive model of livestock movements. That would be really useful in trying to understand the impact of proposed regulatory changes – rather than introducing things like pre-movement testing for tuberculosis and hoping they’re beneficial, we could attempt to model what changes would occur, and so design better regulations to keep the UK cattle industry safe.
You might be wondering where the infectious disease modelling is in this work – how has the UK’s vulnerability to an invading disease of cattle changed in the last decade? I’ve wondered that too, and have been doing some work on it, but that paper’s still with referees, so you’ll have to wait a little longer before I can write about it here…