The Scottish Government

« Previous | Contents | Next »

Listen

ANNEX 2 (a) Transmission of bluetongue virus within and between farms

This appendix provides a summary of the models used to describe the transmission of bluetongue virus ( BTV) within and between farms in Great Britain ( GB), including the assumptions made and the data used.

Within-farm model

• The structure of the model has been described elsewhere (Gubbins et al. 2008). However, in the GB spread model a stochastic, rather than a deterministic formulation is used.
• The model includes two host species (cattle and sheep) and one vector. For simplicity, the population sizes (i.e. number of cattle, sheep and vectors on a holding) are assumed to be constant.
• Parameter estimates were obtained from the published literature, using those applicable to the GB situation wherever possible (Gubbins et al. 2008) _¹².
• Explicit temperature dependence was included for the reciprocal of the time interval between blood meals (related to the biting rate), the vector mortality rate and the extrinsic incubation period.
• Other critical parameters (probability of transmission from vector to host, vector to host ratios) were set by sampling from appropriate ranges for the parameters.
• For the remaining parameters point estimates were used where robust; otherwise, values were sampled from appropriate ranges.

Between-farm model

• A stochastic, spatially explicit farm-level model with a daily time step was developed to describe spread between farms. A farm is considered infectious once the first newly infectious vectors appear on the farm.
• Transmission between farms is modelled by a generic dispersal kernel, which includes both animal and vector movements. The probability of transmission depends on the distance between farms (i.e. the kernel) and the species composition of the farms.
• If a farm acquires infection, the within-farm model (see above) was simulated based on the number of cattle and sheep kept on the farm. An affected farm was assumed to be detected if an animal died due to BTV infection or if overt clinical signs appeared. For each affected farm the following are recorded:
• • time of challenge;
  • time at which farm becomes infectious;
  • time of appearance of clinical signs in cattle and sheep;
  • time at end of outbreak;
  • number of cattle ever infected and number of cattle dead; and
  • number of sheep ever infected and number of sheep dead.
• Parameters for the transmission probability were estimated using data on clinically affected holdings in northern Europe in 2006 (Albers et al. 2007; EFSA 2007). Estimation was done principally using data for the initial Maastricht focus, but similar estimates were obtained for the Ghent and Cologne foci. The estimates implicitly take into account the impact on transmission of movement restrictions in place at this time.
• Species-specific probabilities for an animal showing overt clinical signs were estimated from OIE reports for 2007.

Overwintering

• Overwintering of BTV occurs in the model via two mechanisms:
• • Hosts infected at the end of one season and which have a very long duration of viraemia could act as a source of infection for newly emerged vectors in the following season.
  • Vectors infected at the end of one season and which survive to the beginning of the following season will act as a source of infection (Wilson et al. 2007). The likelihood of this happening will depend on the vector mortality rate at low temperatures, which for the temperature-dependent function used in the within-farm model is very low. However, this is based on extrapolation from data which considered vector mortality over a restricted temperature range (10-30°C; Gerry & Mullens 2000).
• Three further overwintering mechanisms have been posited: vertical transmission in the vertebrate host (Anonymous 2008); covert persistence of BTV in immune cells in the vertebrate host (Takamatsu et al. 2003); and vertical transmission in the vector (White et al. 2005). However, none of these mechanisms has been quantified in the field, and so they cannot be easily incorporated in the model.
• In summary, it is likely that model overestimates the importance and frequency of overwintering in the vector and underestimates that in the ruminant host. Consequently, the overall frequency of overwintering is probably captured by the model, if not the precise mechanisms.

Vaccination

• It is assumed that vaccination acts by reducing the probability of vector to host transmission (i.e. acquisition of infection) and host to vector transmission (to reflect lower virus titres in infected, vaccinated animals).
• The effect of vaccination is assumed to start at zero at the time of vaccination and increases linearly until full protection is reached. Data on the time to full protection in sheep and cattle were obtained from information supplied by the vaccine manufacturers. In sheep this is reached at 14 days post vaccination (dpv); for cattle it is reached at 60 dpv.
• Vaccination is assumed to be 100% effective in all animals.
• Farms belonging to a protection zone ( PZ) are vaccinated so that a specified percentage of holdings are covered.

Initial conditions

• The initial conditions were set depending on how the model was being used.
• For the incursion scenarios (i.e. spread in GB) the model was initialised with a single infected farm (Baylham Farm, near Ipswich) on 04 August 2007. Six long range transmission events in Cambridgeshire, Kent and Sussex (i.e. affected farms identified at a long distance from the main East Anglia cluster) were seeded in the model to allow for spread which cannot be easily replicated in the simulations.
• For the epidemiological scenarios where BTV was introduced to Scotland via northwards spread the initial status of farms in Scotland (and the four northernmost counties in England) were extracted from the results of the GB model run until the end of December 2008. For each replicate of the Scotland model the initial status was selected at random from one (out of ten) replicate of the GB model.
• For the epidemiological scenarios where BTV was introduced to Scotland via imported infected animals a single farm was selected as the initial affected location. The initial farm was selected in a two-step process. First, a county was selected at random with probability given by the proportion of movements from England to that county. Second, a farm within the county was selected at random with each farm having an equal probability of being selected. The number of animals imported was drawn from a negative binomial distribution parameterised for each species according to the distribution of batch sizes from the movement data.

Input data

• Farm locations and number of sheep and cattle on the holding were obtained from June agricultural survey data for 2006.
• Hourly temperature records were obtained for 19 meteorological stations throughout GB. Data from 2006 were used for most (14) stations, while data from used for 2005 for the remaining 5 stations; these represent the most complete data-sets. Each farm used temperature records for the nearest station.

Movement data

• Movement data for 2006 for cattle and sheep were obtained from CTS, AMLS and SAMS.
• These data have been aggregated by county and by month to provide:
• number of cattle movements to live and to slaughter from one county to another; and
• number of sheep movements to live and to slaughter from one county to another.
• By linking these with the time at which each county becomes part of a PZ, it is possible to estimate the number of movements of each type lost as a result of movement restrictions.

References

Albers, A.R.W., Mintiens, K., Staubach, C., Gerbier, G., Meiswinkel, R., Hendrickx, G., Backx, A., Conraths, F.J., Meroc, E., Ducheyne, E., Gethmann, J., Heesterbeek, J.A.P., de Clercq, K., Unger, F. & Stegeman, J.A. 2007 Bluetongue virus serotype 8 epidemic in north-western Europe in 2006: preliminary findings. In Proceedings of 25th Meeting of the Society for Veterinary Epidemiology and Preventive Medicine, March 2007 (ed. D.J. Mellor & J.R. Newton), pp. 231-245. Edinburgh: Society for Veterinary Epidemiology and Preventive Medicine.

Anonymous 2008 Bluetongue virus might overwinter in fetuses. Vet. Record162, 328.

EFSA 2007 Epidemiological analysis of the 2006 bluetongue virus serotype 8 epidemic in north-western Europe (available online at: http://www.efsa.europa.eu/EFSA/1178620925100/efsa_locale-1178620753812_Bluetongue.htm ).

Gerry, A.C. & Mullens, B.A. 2000 Seasonal abundance and survivorship of Culicoides sonorensis (Diptera: Ceratopogonidae) at a southern Californian dairy, with reference to potential bluetongue virus transmission and persistence. J. Med. Entomol.37, 675-688.

Gubbins, S., Carpenter, S., Baylis, M., Wood, J.L.N. & Mellor, P.S. (2008) Assessing the risk of bluetongue to UK livestock: uncertainty and sensitivity analysis of a temperature-dependent model for the basic reproduction number. Journal of the Royal Society Interface5, 363-371.

Szmaragd, C., Wilson, A., Carpenter, S., Mertens, P.P.C., Mellor, P.S. & Gubbins, S. 2007 Mortality and case fatality during the recurrence of BTV-8 in northern Europe 2007. Vet. Record161, 571-572.

Takamatsu, H., Mellor, P.S., Mertens, P.P.C., Kirkham, P.A., Burroughs, P.A. & Parkhouse, R.M.E. 2003 A possible overwintering mechanism for bluetongue virus in the absence of the insect vector. J. Gen. Virol.84, 227-235.

White, D.M., Wilson, W.C., Blair, C.D. & Deaty, B.J. 2005 Studies on overwintering of bluetongue viruses in insects. J. Gen. Virol.86, 453-462.

Wilson, A. J., Carpenter, S., Gloster, J. & Mellor, P.S. 2007 Re-emergence of BTV-8 in northern Europe in 2007. Vet. Record161, 487-489.

« Previous | Contents | Next »