Rear top herds on diet of data

Databases could help raise better livestock, says John Woolliams.

One of the issues that the traumas of BSE and foot-and-mouth disease have highlighted is the need for more and better information about livestock farming.

Information technology can transform the management of livestock farming by combining the miniature "databases" held on farms. Farmers record a range of information on livestock for legal and professional reasons. For cattle, such information includes the birth of calves, movements of cattle, use of veterinary medicines, milk yields and milk quality. Some information is kept on paper, and some is captured automatically and stored electronically.

Much debate has centred on how information is made available for managing crises day-to-day, but there is a longer-term strategic issue of how we can be more effective by building and using databases. To benefit farmers and the public, the Department for Environment, Food and Rural Affairs should promote the effective use of existing data and work to improve data quality. The argument applies to all livestock, but the illustrations that follow are specific to cattle.

The benefits will come from being able to inform and guide solutions to problems, such as the use of antibiotics in livestock. There will always be a need for veterinary antibiotics to treat disease, but the European Union is developing a policy of reducing our reliance on them so as to minimise risks to human health.

One of the most effective ways is to improve animal welfare by reducing the incidence and severity of common bacterial diseases. How can this be done? If information on veterinary treatments were centrally and readily available, statistical techniques could be used to identify those animals that are genetically most resistant to disease. By breeding from the more resistant animals, the population could be made more resistant to the disease and less reliant on antibiotics.

Such concepts are not new. In the UK, this approach is used to improve the yield and quality of milk in the dairy herd. But the ability to address wider objectives such as disease resistance is limited by the quality of the data collected on the farms, its availability and the infrastructure needed to obtain and collate it. Solutions to these data problems exist in Scandinavian countries - Danish scientists recently used such data to identify a gene that is responsible for abortions in cattle. The data problems are becoming easier to solve because of continuing developments in farm management and communications technology.

Remote-sensing technology could also play a role, although sensors are unlikely ever to replace a good stockman because quality livestock care requires the interpretation of often subtle visual cues. Monitors are being developed for milking parlours that can "sniff" for the presence of metabolites or minute quantities of reproductive hormones in milk. These can reveal the wellbeing of the cow or help diagnose reproductive problems. Initially these sensors would be used as daily management aids, but their long-term potential for revealing genetic solutions to reproductive and welfare problems is huge.

The benefits of better information on livestock are not limited to genetics. Looking at large amounts of data from different angles stimulates new ideas and helps sort plausible from implausible hypotheses, so avoiding years of wasted research. In determining the extent to which BSE might be spread from cow to cow, big advances were made only after data became available to different scientific groups with different perspectives and skills.

The quality of the information obtained will depend on farmers' participation. Though the technology helps with daily management, the wider benefits will come only from pooling information across farms - it is this that allows researchers to separate genetic effects from other factors. Not every farm would need to take part if a significant number of representative farms is included; for the dairy sector this might be one farm in five.

Livestock farming has a very low level of profitability and farmers view recording data as a job that is desirable but very unrewarding in the short term. If we want livestock farmers to address wider issues, we must provide some of the tools and infrastructure to help them to do so.

Foot-and-mouth disease and BSE have put management of rural affairs and the direction of agricultural subsidies higher on the political agenda than at any time in recent history. E. M. Forster in Howards End recommends us to "only connect". His view of the world may be too narrow for modern livestock farming, but there is a strong argument for following his advice as one component of a new arrangement of rural affairs.

John Woolliams is a quantitative geneticist at the Roslin Institute, Edinburgh. He will speak in the "Risk management and its communication" session on Tuesday afternoon, September 4 at the BA Festival of Science in Glasgow.