FMD dilemma

Modelling the control of foot and mouth

As the UK farming industry reels from the mass culling policy adopted to halt the spread of foot and mouth disease in 2001, a team of researchers is investigating alternative control strategies in the event of a future outbreak.

Although there were only 2030 actual cases of infection during the UK foot and mouth disease (FMD) outbreak of 2001, six million animals were culled in an attempt to stop the disease spreading - at a total economic cost estimated at £5 billion. "The impact on farmers was devastating," says Dr Matthew Keeling at the University of Warwick. "If you've built a cattle herd up over 20 years and you have to cull the entire herd, you're back to square one. You've lost the last 20 years of your life. We never want to go through that again."

Since FMD occurs all over the world, there is no guarantee it won't be introduced into the UK again at some point. So Dr Keeling is teaming up with Professor Mark Woolhouse at the University of Edinburgh and Professor Bryan Grenfell and Dr Stephen Brooks at the University of Cambridge in an attempt to devise control strategies that will minimise the loss for farmers in the event of a fresh outbreak.

The team is building a detailed computer model of the spread of FMD, against which they aim to simulate and test different control strategies. They began building the model during the 2001 outbreak, when they were called in to advise the government on appropriate control strategies. However, as time was short and results needed immediately, the model was by necessity overly simplified. "For the sake of speed we assumed a simple linear relationship between animal numbers and the infectivity of a farm. We also used straight-line distances between farms, instead of looking at road links, which would give a more accurate idea of distance in real terms. In practice, of course, things aren't so straightforward."

The aim is now to develop a more sophisticated and evidence-based model. Thanks to data collected by the Department for the Environment, Food and Rural Affairs (DEFRA) during the 2001 outbreak, the team has access to uniquely detailed documentation about the spread of the disease. "The data collected by DEFRA is vital, we keep looking back at it to check our model is realistic."

Modelling movement

One of the challenges the team faces is the number of variables that need to be incorporated in order to reflect the reality. The density of farms in a given region, and distances between them - measured in terms of road links, rather than straight lines - are obviously crucial in determining the spread of infection. What happens on each farm at the individual level is also important. The size of each farm, the number of animals on the farm and whether they are predominantly sheep, cattle or pigs (different strains of FMD affect different species in different ways) all influence the spread of disease on the farm and between farms. So do seasonal changes in the number of livestock on the farm, and whether they are kept in fields or housed in barns, and regional factors such as climate and the strength of movement restrictions.

Understanding all these variables - and ascertaining which of them are important in determining the spread of the disease - is at the heart of designing optimal control strategies. "You need determine what's happening very early on," says Dr Keeling. "Ascertain what strain is involved. What livestock will it affect? How fast does it spread? Once you've established what the initial situation is, the models can be used to test what would happen if different controls were applied."

Culling versus vaccination

The route they obviously want to avoid is culling all the animals in a given area, as happened during the 2001 outbreak. Vaccination is an alternative to culling but is problematic, as the team noted in a paper published in Nature in January 2003, comparing various vaccination strategies.

"Mass prophylactic vaccination is effective in that it would prevent a major epidemic by getting sufficient immunity in the population so that the disease wouldn't be able to spread," says Dr Keeling. "But then there's the question of what you do with the vaccinated animals. Current EU regulations don't permit the export of vaccinated animals, or meat from vaccinated animals. Consumer confidence would also be affected: although it is safe for people to eat meat or drink milk from vaccinated animals, people would be unwilling to risk it. That's why FMD has such a major economic impact: although infected animals rarely die, you can't export infected animals and you can't sell infected meat or milk at home."

A combination of reactive vaccination - inoculating herds when the first cases appear - and judicious culling might control epidemics without losing too many animals or impacting as heavily on exports. However, the 6-10 day delay before vaccinated animals are protected against being infected and transmitting the disease reduces the ability of the vaccination to get ahead of the epidemic - with possible devastating consequences. The Netherlands used this strategy in 2001. "They vaccinated a 3 km ring around farms and slaughtered all infected animals. However, ring vaccination only works if you have very strong movement controls. Otherwise transport of livestock outside the ring enables the disease to spread."

The modellers have developed a novel approach to reactive vaccination that does appear to be effective. This 'smart' predictive strategy takes advantage of the precision and fluidity of the new model to accurately predict the most at-risk farms which can be most successfully protected by vaccination. Computer simulations have shown this potentially powerful strategy cuts off the tail of the epidemic by many months and could be of major economic significance.

Political considerations

"At the end of the day, though, the question of whether you use reactive or prophylactic vaccinations - or stick to culling - boils down to a political one," says Dr Keeling. "Do you eliminate FMD with mass vaccination and lose exports and consumer confidence? If you use a reactive strategy, you need to look at logistics. How many vaccines are needed? How many doses are available and can be rapidly produced? How quickly can these by transported and administered on farms? And how much does all this cost - is mass culling, devastating as it is, more cost-effective in the long run than vaccination? You also need to consider issues like movement controls and standstill policy."

Other voices will also want to be heard. "During the 2001 outbreak, the Government had a 20-day standstill policy, during which you couldn't move animals off the farm if new animals had moved onto the farm. The farmers wanted this reduced to a six-day period as this was more economically advisable. The Government agreed - but what impact does that have on the spread of disease? We'll be looking at questions like that in the model as well."

The team is assembling scientific evidence from the model and simulations and passing this on to DEFRA and the Office of Science and Technology to help generate optimal policies to tackle future outbreaks in the UK - and possibly internationally. The comprehensiveness and flexibility of the model, and its 'real-time' application to policy means that similar approaches could potentially also be useful for modelling the spread of other livestock diseases, such as African swine fever.

What causes FMD?

FMD is caused by a hardy virus that can persist in contaminated fodder and the environment for up to one month.

What effects does FMD have on animals?
The disease infects cattle, sheep, pigs and other cloven-hoofed animals, causing sore mouths and feet, fever and appetite loss. Although only five per cent of animals die, many remain debilitated. The disease causes pregnant animals to abort, lowers milk yield and reduces weight gain, so there is less meat on the animal. It therefore has an important economic impact.

Can humans catch FMD from animals?
Yes, but this is very rare. There was one human case in 1967 and none in 2001. As with animals, infected humans suffer a mild flu and a sore mouth.

Is meat from animals with FMD infected?
Yes, but the virus is killed if the meat is cooked properly.

How is FMD spread?
The FMD virus is highly contagious. It can be spread when animals are housed in contaminated barns, transported in contaminated vehicles, or exposed to hay, food or water contaminated with the virus. People can also spread the virus: they can carry it in their nasal passages, on their shoes, clothes, even mobile phones.

How did FMD get into the UK in 2001?
The massive increase in free trade makes import restrictions more difficult, so countries around the world are more vulnerable to outbreaks. The UK outbreak occurred as a result of a worldwide epidemic of a virulent strain called the Pan Asian Type O, which reached 60 countries in 1999. It is thought to have spread to the UK from South America through infected meat fed to pigs in Northumberland in swill, although this is not known for sure.

External links