We are pleased to invite applications for 3-year PhD studentships in the Medical Research Council Centre for Environment and Health (MRC-CEH).

The studentships will be based in  the Department of Epidemiology and Biostatistics (EBS) at Imperial College

We develop and apply statistical methods driven by problems in environmental exposure and environmental epidemiology. Our applications span from climate change and its impact on health, to spatio-temporal surveillance and nowcasting of disease incidence and prevalence, to the evaluation of the impact of policies on population health. To answer these public health questions we develop and employ a wide range of methods, mainly in the Bayesian statistical framework, exploiting the data structure in space and time.

We closely collaborate with other department across Imperial, as well as with public health institutions within the UK and internationally.

Research ideas

We are now seeking outstanding candidates for the following two PhD projects:

1. “Modelling grass pollen concentration in England and its health-related impacts” (Supervisors: Marta Blangiardo and Garyfallos Konstantinoudis):

Climate change and in particular temperature increase is expected to increase the duration, intensity and geographical spread of the grass pollen season. At the same time, pollen is an established respiratory irritant causing higher morbidity in sensitized individuals with asthma and other lung diseases. Nevertheless, there is limited available data about the temporal and geographical variance in pollen counts in England due to the reliance on data from only a small number of environmental pollen monitors.

The student will use different sources of data to predict daily pollen concentrations, including various counts of allergens collected by the UK Met Office and predictors such as temperature, wind speed, rainfall, green space etc. They will develop advanced Bayesian spatio-temporal models to integrate the different sources of data and predict grass pollen counts at higher spatiotemporal resolution. They will consider the SPDE (stochastic partial differential equation) class of models and inference will be conducted with INLA (Integrated Nested Laplace Approximation). Cross-validation will be employed to examine accuracy and robustness of the predictions. Finally, the student will assess the role of the predicted daily grass pollen count (together with their uncertainty) on respiratory hospitalisations for asthma in England.

2. “Spatio-temporal modelling of pathogens in wastewater concentration to inform public health policies” (Supervisors: Marta Blangiardo, Matt Wade – UKHSA):

Wastewater-based epidemiology (WBE) is defined as a collection of tools and methods for surveillance and monitoring disease outbreaks using biochemical analysis of wastewater samples as the primary outcome measure. The first use of WBE was to track illicit drug use. Over the years, WBE has been successfully used in polio eradication and for retrospective prediction of several disease outbreaks, such as noroviruses and hepatitis A, especially in resource-limited settings. During the COVID-19 pandemic, WBE has been recognised as an economically efficient approach for disease surveillance and methods to detect the presence of SARS-CoV-2 RNA in wastewater have been developed in a number of countries.

The student will develop a spatio-temporal statistical model to estimate the relationship between pathogen concentration in wastewater at selected point locations (wastewater programme network) and a set of covariates which represent socio-demographic and environmental characteristics of the network catchment areas. Based on this they will provide probabilistic prediction of spatio-temporal pathogen concentration in wastewater at fine resolution across the study region and evaluate the extent to which wastewater can be used to track “traditional” metrics of the disease measured using clinical data and/or survey data and to investigate the form of the relationship between the wastewater metric and the traditional metrics (e.g. in terms of lag and population characteristics). Finally, they will build a dynamic tool which can provide early warning to inform public health interventions.

The closing date for applications is 1 April 2023.

Successful candidates are expected to start their studentships no later than 1st October 2023.