Comparing Network Analysis Measures to Determine Potential Epidemic Size of Highly Contagious Exotic Diseases in Fragmented Monthly Networks of Dairy Cattle Movements in Ontario, Canada

Abstract

Adult milking cow movements occurring in monthly periods in 2004-2006 were analysed to compare three network analysis measures to determine the lower and upper bounds of potential maximal epidemic size in an unrestrained epidemic: the out-degree, the infection chain or output domain of a farm, and the size of the strong and weak components. The directed networks generated by the movements of adult milking cows were highly fragmented. When all the farms that were not involved in shipments were included in the analysis, the risk of infection transmission through movements of adult cows was very low. To determine the size of an epidemic when an infected farm shipped cows in such a fragmented network, farm out-degree and infection chain provided similar and more reasonable estimates of potential maximal epidemic size than the size of the strong and weak components. Component analysis always provided estimates that were two to three times larger than the out-degree of infection chain approaches. For example, the upper bound was estimated to be 12-13 farms using out-degree and 16-17 farms using the infection chain, the components approach showed a range of 39-51 potentially exposed farms. Strong components provided an inflated measure of the lower bound of potential maximal epidemic size at first diagnosis because the time sequence of shipments was not considered. Weak components provided an inflated measure of the upper bound because both the time sequence and directionality of shipments between farms were ignored. Farm degree and infection chain measures should now be tested to determine their usefulness for estimating maximum epidemic size in large connected networks.

Exploring pig trade patterns to inform the design of risk-based disease surveillance and control strategies

“…Although the GSC size may be an overestimate of a likely epidemic size40, it is useful information for determining the number of potentially exposed farms41. Again, enhancing surveillance and biosecurity at these premises is likely to have significant impact in terms of minimising the extent of a potential epidemic.…”

Simulation of between-farm transmission of porcine reproductive and respiratory syndrome virus in Ontario, Canada using the North American Animal Disease Spread Model

“…Subsequently, the relative frequency of contact from each source production type was calculated for each of recipient production farm types for direct contact based on swine movement (Supplementary material, Table S1) and relative frequency of sharing among production types for indirect contact based on truck sharing (Supplementary material, Table S2) information. Then the contact rates assigned to a production type were adjusted by multiplying the relative frequency by the maximum out-degree for that production type (Supplementary material, Table S3), where maximum out-degree is the maximum number of unique outgoing contacts an individual or group has in the population (Dubé et al, 2008). This approach of using the maximum out-degree between farms is thought to lead to a more realistic epidemic size, as opposed to using either the total number of shipments or average out-degree, which will tend to overestimate or underestimate the epidemic size respectively (Dubé et al, 2008).…”

“…Then the contact rates assigned to a production type were adjusted by multiplying the relative frequency by the maximum out-degree for that production type (Supplementary material, Table S3), where maximum out-degree is the maximum number of unique outgoing contacts an individual or group has in the population (Dubé et al, 2008). This approach of using the maximum out-degree between farms is thought to lead to a more realistic epidemic size, as opposed to using either the total number of shipments or average out-degree, which will tend to overestimate or underestimate the epidemic size respectively (Dubé et al, 2008). Indirect contact rates (Supplementary material, Table S4) from the truck sharing contact matrix were based on the maximum number of farms sharing a single truck within one week.…”

The patterns and risks for disease spreading of cattle movement in China

“…Canada, Britain, Denmark, Sweden and some other countries' research results showed that livestock and poultry movement plays an important role in the spread of animal disease. The researchers also suggested that in the management of animal movements we should pay attention to the following points: 1) Mixed degree, which means the mixed level of directly infected animals and indirectly infected animals; 2) Dispersion degree, that is, one buys infected animals from another seller, then the infected animals involved in trading become scattered in the whole group of animals; 3) Complicated sources degree, namely, the buyer will buy animals from a number of sources [1][2][3][4][5][6][7][8]. However, the relevant research in China is scarce [9].…”

“…In order to reduce the risks for spread of disease in cattle movements, through describing the cattle markets and analyzing the risks in the process of cattle movements, this article wanted to provide the advice about the prevention of animal disease to farmers, practitioners and other related staffs in cattle movements and the reference to establish the disease spread mathematical model and dynamic model [3].…”