Center for
Biosikring og
-Beredskab
Modelling in Biological Preparedness
Modelling as a tool for decision making
Mathematical modelling is a helpful tool for response managers in complex situations where decisions have to be taken within a short time in order to respond adequately to a biological threat with a high impact potential. Wrong decisions may disrupt civil society. In such situations there may not be time to investigate details or to analyse consequences in depth. One must often rely on insufficient data and best estimates.
The response to a biological incident may follow successive stages:
Attack ® Identification ® Modelling ® Resource Management
In each stage different problems require different modelling capabilities.
An
incident can be caused by a deliberate open air release of biological
agent by means of a dispersion device: spray, explosive, or dry agent
delivered by
an envelope. Historically, several biological incidents have taken place
with casualties ranging from single assassinations to large scale
attacks on civilian populations.
A release may be recognised immediately (e.g. visually or a claim by a terrorist organisation) or indirectly by detecting a disease outbreak days later.
After detecting a release, wind dispersion modelling tools can be used to calculate the extent of the area affected. This estimate in turn can generate an estimate on how many people have been exposed and with knowledge (e.g. by field and laboratory investigation) of the agent involved casualty estimates can be made.
Some
biological agents cause epidemics. The magnitude and duration of an
epidemic can be modelled and help to inform public health officials and
the health care system about resource needs to be met.
In the health care system modelling tools are used to suggest the optimal distribution of resources: the most effective combination of workforce, utensils, and equipment to provide the maximum treatment capacity.
Alert
|
Covert attack - disease surveillance A covert release may only
be identified by a sudden increase in symptomatic cases.
Surveillance systems that monitor disease frequencies have been
implemented in order to identify a disease outbreak as early as
possible. Computer automated systems can report abnormal high
frequencies automatically and be used to initiate further
investigations.
|
Overt attack - dispersal assessment Aerial dispersion of an
agent can be modelled if some initial information is available:
where, when, what. The output from the dispersion calculations is
foremost an estimate of size and position of the affected area and
thus persons at risk plus derived calculations such as expected
casualties.
|
|
|
Sudden increase in disease frequency |
Dispersion calculation displayed on a map |
Epidemic modelling
Modelling the spread of disease in a population is an old discipline that is still evolving into more and more sophisticated models. Modelling can help health professionals to organise control measures that can reduce the number of people falling ill and modelling can also help to effectively administrate treatment resources. Additionally some tools have been developed to model consequences of implementation of countermeasures such as the effect of reducing contacts, medication or personal protective equipment.
Two models, Epi-X and Epi-Y are available for estimation of epidemic
parameters: ill, susceptible and immune/dead
Resource management
As
part of the total response capability, hospital treatment capacity is a
key parameter in the management of large and sudden disease outbreaks.
The surge treatment capacity must be dimensioned in relation to
realistic scenarios and plans to expand capacity within a short time
should be developed.
Survey of operational modelling capabilities and needs in Europe
Tlf: +45 3268 8127
cbb@ssi.dk
Opdateret:
10. March 2009
Center for Biosikring og -Beredskab,
Statens Serum Institut,
Artillerivej 5,
DK-2300 København S