US-UK Collab: Adaptive surveillance and control for the elimination of endemic disease

Quantitative observations of infectious disease incidence define the scope of human and animal health problems, and, over time, reveal the underlying dynamics that allow us to develop, and then evaluate, effective control strategies. As such, surveillance systems are crucial to developing evidence based policy. Developing an efficient surveillance network requires knowledge of the system under surveillance. This proposal develops a framework for the evolution of surveillance systems and their adaptation in tandem with the development of models to support decision-making, changes in control and elimination policies, and the evolving dynamics of disease transmission in response to control. In many settings where surveillance networks remain rudimentary (e.g. in under-resourced settings or diseases that are not yet considered priorities), development of evidence-based policy is hampered by lack of locally-specific demographic, surveillance, serotype, and movement/connectivity data. While efforts to collect these data will always advance scientific understanding, we propose that model-based prioritization of data collection and surveillance system design will lead to more efficient targeting of data collection to specifically support disease control and eradication goals. This work will develop general methods for model-based prioritization of data collection and surveillance system design using highly resolved, long-term, strain-specific surveillance data on foot-and-mouth disease (FMD) virus incidence in Turkey from 2001-2012.
This proposed work will develop the first country-scale, strain-specific models of endemic FMD transmission. FMD is a significant livestock disease in many low and middle income countries (LMICs) and the models developed here will provide insight into the mechanisms of FMD persistence and serve as a guide for the development of locally specific models to support policy development in endemic countries. This proposal will develop novel methods for developing hybrid surveillance systems to efficiently mix 1) passive surveillance with model-based allocation of active surveillance effort, and 2) field diagnosis with diagnostic confirmation.
The data available for Turkey reflect a "gold standard" for what could be achieved in counties that are currently developing FMD control and elimination plans. The methods that will be developed in this proposal will address the iterative development of empirical models and surveillance systems for FMD and will provide a roadmap for countries that currently do not have highly resolved surveillance data. The co-investigators will engage with FMD managers in Turkey, through annual meetings, to transfer models and insights on surveillance and control strategies, and to develop additional control strategies to be evaluated with the models developed herein. The co-investigators will further engage with FMD managers in Kenya and Uganda through existing EuFMD training workshops to develop applications of lessons from the Turkey case-study in those countries. Many control programs for endemic disease suffer from limited, or imperfect, data collection. The examples and methods developed in this proposal will be broadly applicable to human and animal disease settings where surveillance, models, and policy need to be developed in tandem. The co-investigators for this proposal have a long history of engagement with both human and animal health policy organizations and will work to translate lessons from the FMD case study to additional settings. Junior project staff will be trained on the methods developed in this proposal in application to FMD and encouraged to develop applications in other endemic animal and human disease systems (e.g. avian influenza, measles, rubella, meningitis, rotavirus) studied by the assembled co-PIs. These individuals will serve a valuable role in the onward translation of quantitative science to regional and national agencies throughout their careers.

Mathematical models have historically been used in epidemic settings to provide predictions and policy advice during foot-and-mouth disease (FMD) outbreaks, most notably during and in the aftermath of the 2001 FMD outbreak in the UK. However, accurate parameterisation of such models relies upon a regular flow of epidemiological and demographic data, that are not available in many endemic settings. In FMD-endemic countries, it is estimated that the global economic costs of the dise Fiase are as high as $20 billion in lost production and costs of disease control. FMD has been identified as an important contributor to poverty in endemic regions and this has led to the initiative by EuFMD, the Food and Agriculture Organisation of the United Nations (FAO) and the World Organisation for Animal Health (OIE) of the "Progressive Control Pathway for FMD Control" (PCP), a process that is implemented in endemic countries to follow a stepwise route to FMD freedom.

In this project we will utilise a suite of modelling tools to establish efficient surveillance and vaccination protocols that could be deployed to facilitate progression for endemic countries through the PCP framework towards disease freedom. In particular, we will develop the first country-scale, strain specific models of endemic FMD transmission that will be used to investigate the predictive capacity of models in endemic settings where often precise demographic and epidemiological information are not available. Our work will also reveal the dynamics of FMD serotype replacement at the landscape scale and develop models to evaluate serotype-specific vaccination strategies. Throughout our project we will interact with EuFMD real time training courses in Kenya and Uganda, facilitating targeted collection of epidemiological and demographic data that will enable us to assess targeting strategies for surveillance and vaccination policies that can reduce the impact of FMD in the future.

Given the applied nature of the research in this project, obtaining impact is a key priority area and numerous individuals and groups within the UK, the USA, Turkey, Kenya, Uganda and worldwide will directly benefit from our findings.
A range of policy makers and associated organisations will be given access to our findings and will receive policy-ready documentation that outlines our main results in clear non-scientific language. These include government agencies such as the Department for the Environment, Food and Rural Affairs in the UK, the US Department of Agriculture, the Ministry of Food, Agriculture and Livestock in Turkey, the Ministry for Agriculture in Kenya and the Ministry of Agriculture, Animal Industry and Fisheries in Uganda, and non-government organisations such as the Food and Agriculture Organisation of the United Nations, FAO and the European Commission for the Control of Food and Mouth Disease, EuFMD. The investigators have an excellent track record of working with and supplying information and advice to such organisations that will assist in disseminating this information. In addition, EuFMD, as collaborators on this proposal, will work closely with the Ministries in Turkey, Kenya and Uganda to ensure that outputs are reported directly to stakeholders. Policy makers and stakeholders will therefore benefit from access to a range of epidemiological forecasts based on detailed and ground-truthed data together with a quantitative assessment of the likely impact of policy recommendations. This will ensure that any policy decisions or practical recommendations benefit from the latest epidemiological advice. This advice will extend beyond the region of immediate study and will provide much needed guidance across all regions affected by foot-and-mouth disease (FMD).
In order to communicate the outputs from our models and to liaise closely with policy makers and stakeholders, we will visit Turkey, Kenya and Uganda once per year and engage with veterinary training workshops led by EuFMD. These workshops will provide us with the opportunity to report directly to regional stakeholders to ensure that our findings have a benefit to the livestock farming community.
In addition, those commercially involved with livestock farming in the regions will benefit from a greater understanding of the risks that different activities pose in terms of bringing infection into the farm environment and intervention strategies that may reduce risk of disease spread. Outbreaks of FMD can be financially damaging to the individual farmer and the industry as a whole and therefore it is vital that our recommendations to farmers are resonant with the practicalities of this industry, while industry support for our work is extremely beneficial. Poor and rural farmers will potentially have the greatest and widest gain from our findings from this component of the project. We hope that the livestock farming industry benefits from focused and practical advice that can reduce future disease risk, thereby safeguarding an important industry whilst ensuring the population at large has continued access to cheap and safe food sources.
Our overarching aim is to develop an understanding of the ability of adaptive surveillance and control strategies to reduce the impact of FMD and to improve the livelihoods of farmers in lower and middle income countries. This aim, together with our dedication to interact with a range of local organisations, should provide substantial and sustained benefits to Turkey, Kenya, Uganda and other countries around the world.