Development of a statistical modelling framework to inform real-time immunization strategies for elimination of poliomyelitis and other diseases

The campaign to eradicate polio, a debilitating and potential-fatal disease, officially began in 1988 at the World Health Assembly. Control of polio has been largely achieved through the use of vaccines, and in most countries the oral polio vaccine (OPV) is used out of choice due to its low cost and ease of use. Since 1988 widespread use of the OPV reduced the number of children paralysed by polio from an estimated 350,000 in 1988 to just 1,349 in 2010. Circulation of the disease causing virus has not yet been stopped in Nigeria, India, Pakistan and Afghanistan. In these countries hundreds, sometimes thousands, of children have been paralysed due to infection each year. Outside of these countries, and particularly in Africa, outbreaks of disease are frequent, and can spread across communities rapidly, requiring additional vaccination in regions already stretched by other disease burdens.

The aim of my research is to improve the current understanding of what affects the spread of polio. By understanding what influences the emergence and spread observed in previous outbreaks, I will be able to advise on likely regions at-risk of experiencing cases, and who in the communities are most at risk. The advantages of this kind of approach are two-fold; at-risk regions and communities can be vaccinated more in advance of an epidemic, and during an epidemic there will be an increased understanding of where cases may be detected next, and vaccination can begin ahead of detecting cases. The research is likely to reduce the number of people affected by poliomyelitis, and contribute to global eradication.

In collaboration with the Global Polio Eradication Initiative (GPEI) I will use mathematical and statistical techniques to analyse surveillance data on the timing and location of polio cases in Africa since 2003. The surveillance data are collected by the GPEI to monitor where cases are, and vaccinate affected communities (no clinical trials or additional human samples will be collected for this research). Previous research has shown that the spread of polio at a country scale is exacerbated by low population immunity and exposure to polio cases elsewhere in Africa through international migration. However, questions still remain; what affects spread between smaller groups of communities such as villages and towns? How fast does polio spread across the continent and what affects this? How many people should be vaccinated in response to an outbreak, and who is most at risk? Although there are policy guidelines developed by WHO in relation to who and how many to vaccinate (2-5 million children within each affected country less than 28 days after the first case of an outbreak), more could be done in terms of the scientific insight that guides these policies.

I am an experienced scientist that works on the control of infectious disease, and my current focus is on the eradication of poliomyelitis. I am currently employed at Imperial College, in the Medical Research Council Centre for Outbreak Analysis and Modelling, and it is here that I plan to be based during the fellowship. The MRC Centre specialises in the development of quantitative tools for the control of infectious diseases, and includes experts in this field such as Professor Nicholas Grassly. I will also be working in close collaboration with the WHO. I will benefit greatly from training provided by Dr Metcalf and Dr Hay at Oxford University.

Objectives and methodology:
> Determine spatial and temporal trends in population immunity of children in Africa
A cohort model of the immunization status of children aged 0-5 will be developed, and parameterised using surveillance data. Estimates of vaccine associated population immunity will be provided along with measures of uncertainty
> Develop a model framework to identify demographic and mobility networks that influence poliomyelitis spread
A metapopulation approach (using a susceptible-infected-recovered model and exploring other methods such as transmission trees) will be used to test hypotheses on the factors affecting the spatial and temporal spread of disease at a fine spatial scale (districts within countries within Africa). Further developments of this methodology will be included to account underreporting and asymptomatic infection of poliovirus.
> Develop risk maps for poliomyelitis and validate findings using prospective sampling
Inferences made from the proposed model will be validated using prospective sampling from historical outbreaks of poliomyelitis in Africa. Methods to assess predictive ability will be reviewed and updated. During the four years of the fellowship, I will also test the findings on emergent outbreaks.
> Test different vaccination strategies to limit and prevent poliomyelitis outbreaks
From the findings identified above, different vaccination strategies will be trialled using computer simulation. I will identify optimal control strategies that focus on the scale of vaccination that will limit outbreak size and duration.

Scientific and medical opportunities:
New methodologies will be developed specific to capturing dynamics of infectious diseases that are partially observed. Insights will also be applicable to other infectious diseases in humans and animals. The research will provide valuable insight for policy makers and stakeholders within the GPEI, and will contribute to the eradication of poliomyelitis.

The Global Polio eradication Initiative (GPEI), and especially the World Health Organisation will benefit from the scientific advances made during the Fellowship. Hypotheses will be tested that relate directly to policy, and consequently more informed decisions can be made on the control and eradication of poliomyelitis. This will enable a more economic approach, in terms of finances and resources, by focussing control on areas that will have the largest effect, as shown through my analysis. The findings will influence the time to eradication, impacting upon the quality of life of millions of people.

An improved understanding of the transmission of infectious diseases in Africa will be made, which will have wider implications for other researchers and health-care professionals working on other vaccine-preventable diseases. Many of the findings are likely to be transferable to other disease systems, both in humans (measles, meningitis) and animals (foot-and-mouth disease).

Development of novel analytical techniques for understanding poliovirus epidemiology, and carrying out this research in the MRC Centre for Outbreak Analysis and Modelling, will have large benefits for the research group. The Centre already has an excellent reputation in epidemiology, and remaining at the Centre will only increase this reputation. By working with Professor Nicholas Grassly and other post-doctoral researchers who specialise in polio epidemiology, the Centre will strengthen its reputation as a world class centre for excellence, and will attract more researchers and students to Imperial College.