Wednesday, February 13, 2008

Re-examining the situation..

In a new report, scientists from the University of Stirling have re-examined data on cattle movements - which have been the subject of some wild accusations regarding the spread of bovine Tb.

Although not backed up by spoligotype data, much was made of a couple of lightweight studies which examined cattle movement data from 2002 and 2003, immediately after Foot and Mouth restocks - data which carried a health warning for the rest of us 'Do Not Use' - and concluded cattle were responsible for spreading Tb.

If a cow has Tb when she steps onto a cattle lorry, she is unlikely to have a miraculous recovery during the journey. The question of course should be, has she been the cause of onwards transmission of bTb? And if cattle are not the cause of an outbreak of this infectious disease, what is? New work by the team at Stirling, based on 2004 data, concludes that in around 75 % - yup that's right, SEVENTY FIVE PERCENT of cases - the cause of bTb outbreaks is not cattle movements, neither local nor national.

In an abstract now published, a piece entitled:
"Estimates for local and movement-based transmission of bovine tuberculosis in British cattle" concludes that 75% of bTb outbreaks were caused by what is quaintly described as 'local effects'. A new word for badgers perhaps?

The authors point out that "Both badgers and livestock movements have been implicated in contributing to the ongoing epidemic of bovine tuberculosis (BTB) in British cattle. However, the relative contributions of these and other causes are not well quantified".

We used cattle movement data to construct an individual (premises)-based model of BTB spread within Great Britain, accounting for spread due to recorded cattle movements and other causes. Outbreak data for 2004 were best explained by a model attributing 16% of herd infections directly to cattle movements, and a further 9% unexplained, potentially including spread from unrecorded movements.

So, using 2004 data on actual bTb outbreaks and matching that to cattle movements on to the farm, both local and national (should there have been any) the team found that just 16% could be attributed to such 'On' movements, with a further 9% unexplained. They explain that the model which best matched the actual circumstances on farm assumed low levels of cattle-to-cattle transmission and continue:
The remaining 75% of infection was attributed to local effects within specific high-risk areas.

Pointing out that "Annual and biennial testing is mandatory for herds deemed at high risk of infection, as is pre-movement testing from such herds", the Stirling team confirm that
"herds identified as high risk in 2004 by our model are in broad agreement with those officially designated as such at that time".

In other words, what was actually happening on the ground, was predicted by the modeling exercise which they used.

And of these outbreaks, three quarters were attributed neither to local cattle movements nor national. But to 'local effects'. And a name they dare not speak.

4 comments:

Anonymous said...

You say that this report "concludes that in around 75 % - yup that's right, SEVENTY FIVE PERCENT of cases - the cause of bTb outbreaks is not cattle movements, neither local nor national".

Unfortunately LOCAL movements are not recorded, and have therefore not been analysed.

As I suspect you bloggers are well aware, in many TB hotspots cattle 'holdings' are not single parcels of land, but include grazing land in separate locations some as many as twenty miles apart.

This does often provide plenty of opportunity for herd to herd contact 'over the hedge' that may well be a significant factor in disease spread.

research cannot analyse data that does not exist!

Matthew said...

Anon; 1.33

Not all farmers have what we think you are referring to as 'linked holdings' which may be blocks of land in various locations. The authors of this site do not - but their cattle, particularly the dairy cattle regularly crossed the road for milking. Four times a day.

It matters not one jot, as when a Tb breakdown occurs, the SVS (now Animal Health) back track all cattle movements going back to several weeks prior to the last clear Tb test. And they include such movements which have to be recorded in the farms' statutory Movement Record.

A case study is then created to ascertain risk and probable cause.

It is interesting to note that the Stirling study's figures correlate very closely with those published in the Dunnet report which analysed data 1972 - 84, and found that 83 percent of outbreaks could be attributed to badgers, 14percent to purchased cattle and 3 percent contiguous - which may or may not have had a common denominator.

A more recent study in the SW found that from 2001 - 2003, the Veterinary Officer determined cause of 85 new breakdowns were 88 percent badger related. 10 per cent occurred in purchased cattle and 2 percent on contiguous farms. Unlike Dunnet, that study had the benefit of spoligotypes to match outbreaks, and it was noted by the author that of the purchased cattle, not all exhibited the spoligotype of their original herd. Some had been infected after purchase and were showing the spoligotype indigenous to the area.

Your point is also assuming, quite wrongly, that any bovine which does move, is likely to be highly infectious. That it has lesions and that those lesions are infectious enough to spread disease on casual contact. Irish work (Costello et al) in the late 1980's and repeated ten years later, housing such cattle with in-contact cattle for several months, sharing water and feed, proved that in fact cattle are extremely low transmitters of bTb.

Jim said...

It's worth recalling that the ISG's mathematical model ascribed "roughly equal importance" to the following sources of infection: "local infection for example across farm boundaries, infection from animals bought, in particular but not only, from high incidence areas, and infection from wildlife, especially badgers" (ISG report, para. 7.24). Perhaps the good professor would now do farmers and taxpayers the courtesy of re-running his figures with the (rather different) percentages which the scientists at Stirling have produced.

It's perhaps also worth noting that one of the studies which I think you are referring to (Gilbert et al, 2005), and which is often cited by those who wish to blame cattle movements for the spread of bTB, actually had next to no data on badgers to put into their own mathematical model. The Supplementary Information which accompanies the published paper says: "Detailed information about badger distribution in Great Britain has not been possible to obtain. Information published in summary form for 1988 and 1997 was unavailable because the original data 'were collected by volunteers on the strict understanding that they would remain confidential.' Alternative, freely available, badger data are, at best, patchy." Mention is made of certain other records but the paper concludes: "Even when aggregated to 10km, these records appear unlikely to provide a very realistic representation of actual distribution." In other words, there was a rather large hole at the core of their model.

Another thought which occurs to me is that the 2004 data used by the Stirling team predate pre-movement testing of cattle. Presumably, therefore, one could expect their 16% figure for infection attributable to cattle movements to have reduced since 2004. Given that bTB infection levels have nevertheless continued to rise, there must be another factor increasingly at work - perhaps their "local effects".

Matthew said...

Jim, your final point is good. Stirling ran their model over 2005 data as well and achieved similar results, but although they mentioned preMT, that didn't begin until 2006, (with cattle over 15 months) and 2007 for cattle over 42days.
They do mention in the full paper, previous work which attributed 89 percent of Tb outbreaks 'Not to cattle movements'.
That was MAFF work in 1991 - but we still got Krebs.