Defra - Department for Environment, Food and Rural Affairs.

Science Search

Science and Research Projects

Return to Science Search homepage   Return to Project List

Development and testing of a model epidemiological framework to optimise the detection and intervention strategies for plant pathogens of statutory concern - PH0406

Description
The aim of this project is to develop a model framework to guide policy decisions aimed at: (i) minimising the risk of further invasion of novel pathogens and pathogen strains, (ii) optimising strategies for control of invading pathogens and pathogen strains. The framework will comprise a suite of temporal and spatio-temporal models that can be used to guide policy decisions in screening and preventing further invasion. It will inform sampling strategies for the detection and estimation of the prevalence of invading plant pathogens in natural, agricultural, horticultural, forest and suburban environments. The framework will also be designed to inform policies on the epidemiological and economic cost-effectiveness of strategies for control and management of invading pathogens.

The project builds on the strengths of the researchers from Cambridge University, Rothamsted Research and the Central Sciences Laboratory. Cambridge University and Rothamsted Research already have a strong track record for many of the mathematical and statistical techniques. The challenge now lies in assembling this into a strategic framework. Central Science Laboratory has a strong record in handling disease management problems and in the identification of threats to invasion.

The models and methods will be developed and made operational in three steps.
(i) We will use expert knowledge to identify a representative and tractable set of scenarios for potential invader plant pathogens. A suite of models will be developed that is sufficiently flexible to describe the epidemic spread of invader plant pathogens.
(ii) The models and methods will be tested using a range of test data sets of plant pathogens that have invaded in the UK and/or in the USA. A unique aspect of the project is that the research team has access to datasets on invasions that are detailed enough to allow model testing.
(iii) The models and methods will then be used to study a series of trial scenarios of plant pathogens that have not yet invaded into the UK but are considered potential threats to plant health. This step makes the models operational to such an extent that, by the end of the project, there is a group of trained researchers that can apply the models and methods when a new invader plant pathogen appears in the UK.

When a new invasive pathogen is found in the UK for the first time it is of key importance to be able to assess the prevalence of the pathogen in the areas where its hosts occur, because the effectiveness of intervention methods to contain or eradicate the pathogen depends on the prevalence of this invader at the moment of implementation of the intervention. Therefore we will integrate the epidemic models with modules describing sampling/monitoring methods. The key questions are:
• what sampling strategy is most time- and cost-effective to map the prevalence of the pathogen, upon first detection?
• given this sampling strategy, what is the probability of detecting an invading plant pathogen for the first time, before it has reached a given level of prevalence?
• given first detection in a new area, can we predict the prevalence it might have reached in this area?

When a map is produced of the prevalence of the pathogen in the UK intervention strategies might be implemented. Presently, intervention strategies aimed at containment or eradication of an invading plant pathogen are developed on a case by case ad hoc basis. These case by case intervention programmes are usually not informed by the existing knowledge about the epidemiology of the disease, nor by predictive modelling of how interventions interact with this epidemiology. The reason for this lack of guidance from epidemiological and modelling studies in informing intervention strategies reflects the absence of a ready-to-use theoretical framework. We will therefore integrate the models with modules describing intervention strategies aimed at containment or eradication of the pathogen. Key questions for implementation of intervention methods are:
• how can we match the spatial scale of intervention strategies with the spatial and temporal scales of the epidemic so as to optimise disease control and management?
• how do intervention strategies affect the prevalence and further spread of the disease?
• how does uncertainty about pathogen biology and host distribution affect the probability of success and the risks of failure of control strategies?
• what is the most cost effective method of intervention?

In addition to developing and testing a theoretical epidemiological framework to optimise the detection and intervention strategies for plant pathogens of statutory concern, the project will also provide a set of protocols describing how the models should be used in the case of an epidemic outbreak of a plant pathogen together with a a group of researchers trained to use the models within Plant Health Service.
Objective
Aim: To develop a model framework to guide policy decisions aimed at: (i) minimising the risks of further invasion of novel diseases and pathogen strains, (ii) optimizing strategies to control invasions. The framework will comprise a suite of temporal and spatio-temporal models that can be used to guide policy decisions in screening and preventing invasions: it will inform sampling strategies for the detection and estimation of the prevalence of non-native and other invading plant pathogens in natural, agricultural, horticultural, forest and suburban environments. It will also be designed to inform policies on the epidemiological and economic cost-effectiveness of strategies for control and management of invading pathogens. Subject to adequate resources (detailed below), it will include testing of statistical methods to optimise data collection for parameter estimation.

For convenience we distinguish three broad objectives. These concern epidemic models, parameter estimation and sampling, intervention strategies

OBJECTIVES
I. Epidemic Models.
I.1: To use expert knowledge of: (i) pathogen biology; (ii) host spatial distribution and (iii) host temporal dynamics in order to identify a representative and tractable set of scenarios for potential invader plant pathogens.

I.2: To map the scenarios comprising classes of potential invaders and host distributions onto a generic, mutually-related, suite of models for temporal and spatial dynamics for pathogen spread.

I.3: To select and organise test datasets from existing and archival data on invading plant pathogens for model testing.

I.4: To use (i) the test datasets for invading pathogens together with a set of hypothetical trial scenarios to tailor the generic models for specific cases, and (ii) to parameterise the models under uncertainty, in order to simulate and test the robustness of the models for practical use.

II. Sampling and estimation: mapping pathogen prevalence after first
detection and estimating transmission and dispersal parameters.

II.1: To integrate the epidemic models with modules describing sampling/monitoring methods in order to develop time and cost effective sampling strategies to map pathogen prevalence following first detection. This will include: (i) quantification of the probability of detecting the presence of a plant pathogen in a new area for the first time, before it has reached a pre-set level of prevalence and (ii) initial prediction of its prevalence.

II.2: To test methods for estimation of transmission and dispersal parameters of emerging epidemics of plant pathogens.

II.3: To test the mapping and estimation methods using the selected test datasets.

II.4: To apply the mapping and estimation methods to the trial pathogen species.

III. Intervention strategies.

III.1: To integrate the models with modules describing intervention strategies in order to quantify the epidemiological and economic effectiveness of: (i) intervention strategies on disease prevalence; (ii) the probability of eradication or containment for different intervention strategies.

III.2: To test the methods on the selected test datasets.

III.3: To integrate the intervention models with a simple economic impact module to calculate economic costs of intervention.

III.4: To apply the methods to the trial scenarios to test the implementation of the methods.

IV. Knowledge transfer and training.

IV.1 To transfer knowledge to CSL Plant Health Service modellers and provide training in the application of the sampling methods and models to the trial examples. This objective is combined with objective II.4.

IV.2 To transfer knowledge to CSL Plant Health Service modellers and provide training in the application of the intervention strategy methods and models to the trial examples. This objective is combined with objective III.3.

IV.3 To transfer the methods, models and programmes to CSL Plant Health Service modellers.

Project Documents
• OTH - Other : PH0406 Publication list   (24k)
Time-Scale and Cost
From: 2007

To: 2013

Cost: £715,417
Contractor / Funded Organisations
Rothamsted Research (BBSRC), University - Cambridge, Central Science Laboratory
Keywords
Plant diseases              
Plant health              
Plants and Animals