Yes, it's interesting that while the UK govt says it's doing no Lyme research at the moment, and has done none since 1999 (they told this to Parliament a few weeks ago), nevertheless someone in the UK says they ARE doing Lyme research. And guess who? Porton Down, which is the British Fort Detrick...
Here's the relevant extract, followed by the full text of a very interesting document found recently.
Lisa
---------
"14. An example of a predictive modelling study being carried out at HPA Porton Down relates to
the sheep tick, which can transmit a range of vector-borne diseases, including Congo Crimean
Haemorrhagic Fever, Lyme Fever, ehrlichiosis, babesiosis and tick-borne encephalitis.
By undertaking comprehensive ecological and environmental analysis in a sophisticated GIS of the
key parameters that impact on tick ecology (such as land cover type, climate, topography, animal
host distribution and human demographics), the model can identify potential hot spots for tick
distribution and abundance and therefore the potential risk areas for future vector-borne
transmission of infection to humans."
____
MEETING OF THE STATES PARTIES TO THE
CONVENTION ON THE PROHIBITION OF
THE DEVELOPMENT, PRODUCTION AND
STOCKPILING OF BACTERIOLOGICAL
(BIOLOGICAL) AND TOXIN WEAPONS AND
ON THEIR DESTRUCTION
BWC/MSP/2004/MX/WP.51/Rev.1
29 July 2004
ENGLISH ONLY
Second Meeting
Geneva, 6-10 December 2004
Meeting of Experts
Geneva, 19-30 July 2004
Items 5 and 6 of the agenda
A modelling programme on bio-incidents
Submitted by the United Kingdom
1. The UK has traditionally had strong epidemic and mathematical modelling capabilities, and
these are increasingly being applied to enhance response capabilities to future emergency public
health situations. Active links have been established between modelling groups in the Department
of Health and at HPA Colindale and HPA Porton Down, with groups in other government
departments and agencies such as the Ministry of Defence's Defence Science and Technology
Laboratory, as well as in university departments in Oxford, Cambridge, Warwick and at Imperial
College London. International collaborative efforts are increasingly important, including work
related to the Health Security Committee of the EC and the G7 + Mexico Global Health Security
Action Group (GHSAG).
The modelling programme at HPA Porton Down
2. HPA Porton Down has a Microbial Risk Assessment Programme based on work over several
years, to provide evidence based risk assessment and modelling approaches which can contribute
advice to government in the formulation of policy, emergency health planning and disease
countermeasures. This programme concentrates on areas of potential public health concern in which
HPA Porton Down and collaborating institutes have particular expertise, which include a number of
diseases caused by dangerous pathogens in Risk groups 3 and 4 which are not endemic to the UK.
3. The HPA Porton Down programme has a number of key components:
(a) Horizon scanning to ``second-guess'' the types of event for which modelling
applications would be helpful. Events could be natural outbreaks such as pandemic influenza or
SARS, or outbreaks deliberately caused by terrorists. Given the potential for spread of infectious
diseases into new habitats - as seen in the recent introduction of West Nile virus into the
GE.04-62641
BWC/MSP/2004/MX/WP.51/Rev.1
Page 2
United States - attention is also being given to future potentials for spread of vector -borne and
zoonotic diseases in the UK;
(b) Mathematical modelling tools able to address issues thrown up by horizon scanning;
(c) Underlying data requirements. Geographic Information Systems (GIS) are important
because data and the modelling process often rely on specific geographies and spatially referenced
data (e.g. human demographic data, population movements, disease vector distributions, and the
environmental and ecological factors that determine the distribution of all these). The various data
sets currently amount to about 0.6 terabytes;
(d) Dedicated hardware and software to handle, analyse and develop models;
(e) A multidisciplinary team that includes mathematicians, public health microbiologists,
epidemiologists and dedicated GIS and IT/database specialists.
4. The overall strategy is to bring together the data and a predictive, adaptable and scalable
toolbox of models within an integrated IT/database system, so as to assist in forward planning and
for use in real time outbreak/incident management. For forward planning, data that are collected
and warehoused regarding past incidents and outbreaks can be fed through into appropriate
modelling applications to analyse future ``what if'' scenarios. In the event of an actual infectious
disease emergency, real time data feeds would be rapidly mapped in the GIS and analysed to re -
parameterise the models for the particular outbreak in question. These models can then be run in real
time to help with outbreak/incident management by making ongoing short/middle term predictions
of the likely state of affairs in the upcoming time period, of whether the current interventions are
having the required effect in terms of bringing the situation under control, and of the likely time
scale for control to be established.
Example of a `real time' application: the Legionella outbreak in Hereford, 2003
5. The health officials investigating an ongoing outbreak of legionellosis in Hereford in 2003
had identified a cooling tower that they felt could have caused the outbreak, consistent with two sets
of information: the location of those individuals who did not move out of the area during the
outbreak; and the timing and location of the infected individuals from outside Hereford who seemed
to have made only one visit to Hereford. HPA Porton Down was asked to help provide
confirmatory evidence that the tower could have been responsible for the outbreak.
6. For each person who had developed the disease, the outbreak investigation team had
constructed a detailed record of locations visited, with time and date details, during the periods
leading up to the onset of their illness. The modellers took this data for each individual and
compared it with the emissions that would have been predicted from the tower in the days leading
up to the onset of symptoms. Predictions of the likely emissions from the tower for each day of its
operation were based on atmospheric dispersion modelling using software called HPAC (Hazard
Prediction and Analysis Capability). This required inputs of data on the source characteristics -
height and other dimensions of the stack, aerosol flow through the stack, microbial load in aerosol,
BWC/MSP/2004/MX/WP.51/Rev.1
Page 3
survival of pathogen in aerosol - and all relevant meteorological data - such as wind s peed,
direction, height of boundary layer. The meteorological factors varied considerably both during and
between different days. The locations and travel histories of the cases were plotted using GIS, in
relation to the predicted plumes of Legionella release from the tower for each of the relevant days.
By integrating the outputs from the dispersion modelling with the relevant data on the legionellosis
cases within the GIS it was possible to demonstrate that the locations and trajectories of the cases
were consistent with the postulated exposures from the tower.
7. The analysis was extended to determine whether the timing of the onset of the individual
cases was consistent with the timing of the operation of the tower. Luckily, good independent data
that c ould be used to derive an incubation period distribution representative of Legionella infection
were available from an earlier outbreak in The Netherlands in 1999 following the discrete exposure
of individuals at a flower show. Individuals who had visited the show once subsequently developed
disease from the single exposure, so providing a clear picture of the individual variation in the
length of the incubation period of the disease. In the HPA investigation, an individual-based
stochastic model was deve loped based on the Dutch data. By carrying out Monte Carlo simulations
that were fitted to the observed data by maximum likelihood analysis, a predicted Probable
Distribution of Exposure times with confidence intervals was derived. (That is, it was ``backcalculated''
from Observed Distribution of Onset Times). In this way, it was possible to show that
the dynamics of the Hereford epidemic were consistent with the predicted operation of (and releases
from) the cooling tower, and in particular with the date that the cooling tower was ultimately
cleansed, to within 95% confidence limits.
Examples of `ahead-of-time' applications
8. Three examples are given of how HPA Porton Down is using predictive mathematical
models to help in the development of policy and contingency planning ahead of time for potential
infectious disease outbreak emergencies in the UK.
(a) An outbreak of smallpox
9. Modelling of a supposed UK smallpox outbreak has been used to investigate the potential
impacts of different public health intervention strategies over both time and space - thus allowing
comparison of the outcomes of different intervention strategies that are geographically targeted. A
typical analysis used administrative districts as the geographic base, and data on the demogr aphics
of each `patch' from the 2001 (population) Census, along with data on the extent to which people
move between patches. The model was parameterised based on the published information on past
smallpox outbreaks with appropriate adjustments to take account of uncertainties and changes in
social structure and demographics. The model can predict the development and transmission of the
disease both within and between patches. The model can be run in a stochastic mode - run 1000s of
times - with the results being subjected to statistical and sensitivity analysis. It can be used to
investigate the potential effect of public health interventions, such as different efficiencies of :
BWC/MSP/2004/MX/WP.51/Rev.1
Page 4
(i) Case finding and isolation;
(ii) Contact tracing, vaccination,
monitoring and isolation;
(iii) Vaccination in districts where cases have occurred or may be expected;
(iv) Vaccination on a national scale.
10. The particular problem for smallpox is to optimise for an intervention strategy that
minimises the number of cases and deaths from the disease, while at the same time minimising the
number of adverse events, including deaths, that would be associated with widespread vaccine use.
(b) An airborne release of anthrax
11. GIS, dispersion modelling (using HPAC) and epidemic modelling have been combined to
predict the potential impacts of anthrax aerosol releases on a civilian population in the UK, and the
possibilities that may be available for public health interventions. Initial airborne dispersion
modelling data for anthrax has been integrated by GIS with representative mapping and
demographic layers. This allowed visualisation of the dosage contours for the anthrax exposure of
the different parts of the population in relation to a range of useful features, such as the geogr aphic
boundaries of health care areas. By taking account of the different population densities in various
Census output areas, it is possible to calculate the expected casualty counts and also the relative
dosage of anthrax that each exposed individual has received.
12. By applying further bespoke stochastic individual-based models that incorporate appropriate
public health responses, including some logistical delays, the anthrax modelling studies have led to a
number of conclusions:
(i) With increasing distance from the release site the dosages of anthrax received on the
ground get progressively less;
(ii) The probability of infection correspondingly decreases, but the numbers exposed
increase as the plume widens;
(iii) As the received dosage decreases, the incubation period of subsequent infections
increases considerably, from a time scale measured in hours for the highest dosages
to a time scale measured in days and possibly weeks for the lowest dosages;
(iv) People receiving the highest dosages will probably not be identified and treated
quickly enough, but these cases will alert the appropriate authorities that there has
been a release;
(v) People receiving the lowest dosages, which is the largest proportion of those
exposed, could be found and treated with antibiotics.
BWC/MSP/2004/MX/WP.51/Rev.1
Page 5
(c) Future risks from vector-borne diseases in the UK
13. Vector-borne diseases are generally not thought to pose an immediate public health problem
in the UK. However, the situation could change with the increasing international movements of
people, animals and insects, and increasing trade in exotic items, tourism, etc. These trends could
lead to the introduction of new vector-borne zoonotic agents and /or new vectors into the UK.
14. An example of a predictive modelling study being carried out at HPA Porton Down relates to
the sheep tick, which can transmit a range of vector-borne diseases, including Congo Crimean
Haemorrhagic Fever, Lyme Fever, ehrlichiosis, babesiosis and tick-borne encephalitis.
By undertaking comprehensive ecological and environmental analysis in a sophisticated GIS of the
key parameters that impact on tick ecology (such as land cover type, climate, topography, animal
host distribution and human demographics), the model can identify potential hot spots for tick
distribution and abundance and therefore the potential risk areas for future vector-borne
transmission of infection to humans.
____
UBBFriend: Email this page to someone!
I think it's more like 7 days if any at all.
Printer-friendly view of this topic