An analysis of data from the Cattle Tracing System has been carried out by the Environment and Sustainability Institute at the University of Exeter. The APHA was involved in the project and it is assumed that the findings will be perused by the government, with particular relevance to future controls for bovine tuberculosis. A paper entitled “Contact chains of cattle farms in Great Britain” (by Helen Fielding and colleagues) has been published by Royal Society Open Science. At the time of writing, further clarification of the issues that are likely to be considered is not available, but it can be anticipated that cattle farmers and their veterinary surgeons will wish to influence future control options. The second sentence that introduces the paper will make easy reading for vets in cattle practice: “Animal movements between farms can be considered to form a directional link from the source to the destination farm, which may therefore indicate potential pathways for direct and indirect transmission of pathogens.”
An introduction to the study
The authors highlight that identifying the importance of a given farm within a trading network could lead to preferential protection, treatment or isolation of a herd, or specific farms might enhance disease control measures that would influence other herds. Infection risks are said to correlate with the number of farms with which a farmer trades, and the number of animals traded.
It is emphasised that the incidence of disease has not been investigated – only the frequency and volume of cattle trading. This is the explanation for using the term “contact chain” rather than “infection chain”, but a chronic disease can be expected to be transferred over a longer period and trading is expected to enhance the spread of disease. Within the study years (from 2001 to 2015), 158 million individual animal movements between premises were recorded. The analysis excluded births (41 million) and deaths (42 million), movements via markets or show-grounds (26 million) and animals going directly for slaughter (34 million).
The originating farm and the destination farm remained in the dataset but the transitory nature (one day) of markets and shows was considered less relevant for transmission and persistence of slow-spreading infections. The authors recognised the potential role of markets in disease spread, but for bTB the decision was taken to record farm-to-farm.
Twelve-month periods (1 January to 31 December from 2001 to 2015) were considered for each farm, and herd size was recorded as the mean daily number of animals on the farm over the same period. Herd type was defined in each year as beef, dairy or dual purpose. Suckler herds were defined as having calves reared by their dams before weaning and had a majority of females. Fattening units were defined as those with a male majority that did not breed cattle but reared for beef production. Dairy farms were producing milk commercially and any farms where one sex did not comprise more than 50 percent were defined as mixed herds.
Findings and implications
The paper refers to ingoing contact chains (ICCs) and outgoing contact chains (OCCs). Farms with high ICCs are considered at high risk for acquiring infections and farms with high OCCs a high risk of spreading infection.
Analysis indicates that as a result of their cattle purchases within 12-month periods, 47 percent of British farms were connected by ICCs to more than 1,000 other farms and 16 percent were connected to more than 10,000 other farms. As a result of their cattle sales within 12-month periods, 66 percent of farms had OCCs that reached more than 1,000 other farms and 15 percent reached more than 10,000 other farms.
Over 19,000 farms had both ICCs and OCCs reaching more than 10,000 farms for two or more years. Herds with high outgoing and ingoing contact chains are referred to as potential “superspreaders”.
Discounting 2001 (because of the foot and mouth effect), the median number of cattle traded from a single farm to another over the 12-month period was two. Suckler herds made up 56.8 percent of all farms and the predominant between-herd flows of cattle were between suckler herds (12 percent), from suckler to fattening herds (9 percent) and from dairy to fattening herds (7 percent).
Fattening farms received the most movements from other farms (21 percent) and dairy farms received the fewest (4 percent). Thirty-five percent of movements were traded through markets, often from breeding herds back to suckler herds or to fattening herds. On average, 34.4 percent of dairy holdings purchased no cattle in any one year, suckler farms 27 percent, mixed holdings 22.5 percent and fattening units 9.64 percent. Individual farm movements tended to stay consistent over time, indicating a likelihood to trade or not trade.
There are in excess of 600,000 individual animals moved between two farms each year. The data suggest that trading partners tend to reciprocate buying and selling cattle to one another, with multiple small groups of well-connected farms. Dairy farms sell animals to many different farms and it is speculated by the authors that infected dairy herds could be disproportionately influential in spreading disease throughout the cattle network and are a likely target for control measures.
A small number of farms act as “hubs”; these farms have many more trading partners than the majority. The trading of cattle and the risk of disease spread had been thought to be predominately due to markets, but the analysis indicates that hub farms might facilitate epidemic spread by creating transmission shortcuts throughout the cattle network.
Hub farms are considered important for control measures as well as markets. However, the authors point out that individual farmers will have control over their direct purchases but may be unaware of the extended cattle chains, and knowledge of the history of herds within contact chains would better equip farmers and veterinary surgeons to make better buying decisions. Farmers who believe that they trade carefully, with a few partners, may have a false sense of security.
Summary
The paper introduces unfamiliar means of analyses with networks, nodes, edges and in and out degrees. On average, the cattle movement pathways have fewer than seven steps. Digesting the results of the study will require considerable awareness of disease and cattle if practical developments are to follow.
It would be helpful if veterinary practices were able to chart contact chains from records held on clients’ herds. Examples of large chains, where 10,000 links are recognised, are difficult to comprehend and case histories would be highly beneficial. It does seem that the influence of one herd on neighbours and trading partners may be an area that should be looked at more carefully. If a few high-risk farms can be identified and disease control measures introduced, the uptake of preventive strategies may benefit farmers, cattle and veterinary practice.