Inference methodologies and algorithms for fitting spatial models have advanced to the point where a large class of methodological problems in spatial epidemiology have been essentially solved, and increasingly sophisticated statistical methods are appearing in high-impact health science journals. Two important aspects of spatial epidemiology which have yet to be adequately addressed are spatio-temporal models and models for spatially aggregated data.

This talk will cover:

- Geostatistical models in spatial epidemiology: mortality in India;

- Area-level data and changing boundaries: estimating malaria risk; and

- Spatio-temporal models: spatial models on steroids.

Patrick E. Brown (PhD Lancaster, UK) is a Scientist in the Centre for Global Health Research at St. Michael's Hospital, and Associate Professor in the Department of Statistical Sciences at the University of Toronto. His research focuses on models and inference methodologies for spatio-temporal data, motivated by problems in spatial epidemiology and the environmental sciences. Current statistical methods research involves Bayesian inference for non-Gaussian spatial data, and non-parameteric methods for spatially aggregated and censored locations. Applications include estimating disease risk over time when boundaries of administrative regions are changing, modelling and calibrating environmental data in satellite images, and spatial modelling of 'open data' on health outcomes. He has developed and maintains several R packages for modelling and visualizing spatial data.