Course

Course description: Spatio-temporal methods in environmental epidemiology

Description: This course will emphasize the spatio-temporal methods used in environmental epidemiology to characterize the distribution of morbidity and mortality over space and time. The course relies mainly on the first edition that was published after the course was given, as it lays the groundwork for the second edition.

Some of the methods that are described in the course, involve a single area and look at variations in the health responses over time in relation to those of predictors, including environmental hazards. Some involve a single time period in which case, it’s the variations over space that can help identify relationships between these responses and possible explanatory variables. However the scale of modern investigation involves means those relationships over time and space must be both be handled leading to the need for sophisticated methods whose development is an active area of risk in statistical science.

The course does assume a solid background in statistics although the required material will be briefly reviewed. These include but are not limited to: likelihood methods; linear and logistic regression; Bayesian methods; the generalized linear model. Other topics will be developed from first principles including: Poisson regression; quasi likelihood.

The course will not assume a background in epidemiology however. Thus the relevant building blocks from that subject will be described including all necessary definitions and terminology. In particular, we will give an overview of: types of study; rates and standardization; measures of risk; the role of confounders and the problem of causation.

Statistics and epidemiology come together when we address important special topics including, disease mapping, ecological analysis, disease clustering, infectious disease modelling, depending on the interests of those enrolled in the course and time available.

Labs and projects for the course are being included in the GitHub site for Reference 2 https://spacetime-environ.github.io/stepi2

References:

  1. Le, N.D. and Zidek, J.V. Statistics Analysis of Environmental Space-Time Processes. New York: Springer. (2006)

  2. Shaddick, G. and Zidek, J.V. Spatio-temporal methods for environmental epidemiology. London: Chapman and Hall/CRC (2015)

  3. Shaddick, G. and Zidek, J.V and Schmidt, A., Spatio–Temporal Methods in Environmental Epidemiology with R: Second Edition (2023)

  4. Sudipto Banerjee, Alan E. Gelfand and Bradley P. Carlin (2005). Hierarchical Modeling and Analysis for Spatial Data. Chapman & Hall/CRC Monographs on Statistics & Applied Probability.

  5. Noel Cressie and Christopher K Wikle (2011). Statistics for spatio-temporal data. Wiley Series in Probability and Statistics.

  6. Peter J. Diggle (2002). Statistical analysis of spatial point patterns, Second Edition. Hodder Education

  7. Peter J. Diggle and Paulo J. Ribeiro Jr. (2007). Model-based geostatistics. Springer series in Statistics.

  8. Andrew B. Lawson (2006). Statistical Methods in Spatial Epidemiology, Second Edition. Wiley Series in Probability and Statistics.

  9. Nhu D. Le and James V. Zidek (2006). Statistical Analysis of Environmental Space-Time Processes. Springer Series in Statistics

  10. Duncan Thomas (2009). Statistical Methods in Environmental Epidemiology. Oxford University Press