The Myth about "Local Spread"

Until disease transmission is analysed for the truly infected premises, it is difficult to separate myth from reality.

It is interesting to note that there appears to be a discrepancy between DEFRA's , Elliot Morley's, Prof Mark Woolhouse's and the Imperial College team's views on local spread.
That is: the distance the disease was spreading and also how it was spreading.

a) Distance

i) ' 78% of infected premises were within 3 km of another infected premises'
NAO report 18/06/02 - data extracted from DEFRA's Epidemiology report of 21 Oct 2001

ii) Some 45 per cent. of all outbreaks in the disease were within 1 to 2 km of the initial outbreak and 60 per cent. of all outbreaks were within 1 to 3 km.
Elliot Morley ( Hansard 12/11/01)

iii)' 50% of new cases turned up in the immediate neighbourhood of a previous cases'
Professor Woolhouse EFRA select Committee 7/11/01 (189) (neighbourhood is about 3km)

iv) The median distance of the original kernel was about 2 km (March model).
'The median distance of the newly estimated kernel is about 4km.' (October Model) Imperial College team- Nature 4/10/2001 vol 413

Comment on the Imperial College team's view.

My understanding of the Imperial College models, is that in the March model (ie. the one used to determine culling policy) the midpoint distance between index cases and secondary infections was about 2 km. When constructing the October model the Imperial team found that, because the cases had been more accurately traced, the spatial relationship between index cases and secondary infections was not as they had previously believed in March:-

'The newly estimated spatial kernel (October Model) differed significantly from that previously derived from the infectious contacts identified by DEFRA (March model) with considerably more long distance transmission events being predicted. This implies significant biases in the DEFRA contact-tracing process, with closer contacts being more easily identified.'

They discovered that the midpoint distance between index premises and secondary premises was now about 4 km not 2 km. That is: contiguous farms were less at risk then previously thought.

b) Method of spread

i) The Department considered that most local spread was attributable either to aerosol spread between animals or from contamination, for example of roads or common facilities, by poor biosecurity on leaving and returning to farms.
NAO report 18/06/02 - data extracted from DEFRA's Epidemiology report of 21 Oct 2001

ii) Spread was due to animal contact and short distance aerosol spread - March Model

Spread was due to movement of animals, personnel or vehicles rather than through animal contact or aerosol spread - October model.
Imperial College team- Nature 4/10/2001 vol 413

Reality

The reality is until the infected premises are analysed taking into account their laboratory test results ( that is 1324 definitely had FMD, 401 probably didn't and 301 may or may not have had FMD) we can not say with any certainty how the disease was spreading and over what distance.

However, we can say , all the published evidence suggests, that there were many more market related infections than initially thought and we now know that many of these infections were not traced for weeks even months after they had occurred. In addition, there are several counties where there was very little disease transmission once the movement ban was implemented; though the scale of the slaughter in these counties suggests that DEFRA, incorrectly, believed at the time that the disease was spreading.