Bayesian Disease Mapping Pt II Workshop - Oct 17-19, 2011
Instructor
Professor Andrew Lawson, leading researcher in the field of disease mapping and spatial epidemiology, presented this workshop, which was designed to explore special topics in the area of Bayesian disease mapping applications specific to public health and epidemiology.
Workshop description
The two-and-a-half day workshop formed Part II in the sequence of Bayesian Disease Mapping worskops. Both workshops were based on material from the recent book: Lawson (2009) Bayesian Disease Mapping: Hierarchical Modeling in Spatial Epidemiology. It is designed for those who want to develop further skills in BDM methods, and includes use of WinBUGS and related R functions: R2WinBUGS, BRugs. Part II content focused on special topics in Bayesian disease mapping applications to Public Health as addressed in the later part of this book.
Workshop content focused on spatial and spatio-temporal Bayesian modeling issues, and some knowledge of Bayesian computation and WinBUGS was assumed. Workshop delivery included theoretical input, but also practical elements where participants were involved in the hands-on use of R and WinBUGS in disease mapping applications.
Workshop consisted of sessions dealing with:
Evening Refresher:
- A review of basic concepts applied in Bayesian methods and disease mapping.
DAY 1 Spatial topics
- Spatial models and simple variants: convolution, proper CAR, full MVN
- Special applications: sparse count data: zip and factorial regression
- Special applications: latent structure (L&C and mixtures)
- Spatial survival modelling
- Measurement error, SEMS and Joint modeling.
- WinBUGS, R2winBUGS, BRugs and JAGS
DAY 2 Spatio-temporal modelling topics
- Basic ST models: Bernardinelli, Knorr-Held, Waller; seasonal effects
- Infectious disease models: FMD and influenza outbreaks
Prerequisite
Completion of Introduction to Bayesian Disease Mapping (Part I) or experience using WInBUGS and elementary disease mapping methods.
Additional materials/software requirements
- Laptop with R and WinBUGS 1.4.3 software preloaded. Datasets were provided. R and WinBUGS software can be downloaded from the following websites: http://cran.wustl.edu and/or www.mrc-bsu.cam.ac.uk/bugs
- The text, Bayesian Disease Mapping: Hierarchical Modeling in Spatial Epidemiology was available for purchase at time of registration. Cost: $90
Audience
The workshop was intended for epidemiologists and public health workers who need to analyse geographical disease incidence. In addition, the course may be of interest to statisticians or geographers and planners who deal with spatial and spatio-temporal disease data. Some statistical/epidemiological background would have been beneficial but was not essential. Those who did not attended Introduction to Bayesian Disease Mapping benefited from this course and having the opportunity to review Part I workshop content during the Monday evening refresher session.
Sponsored by Population Data BC, in partnership with the Division of Continuing Studies, University of Victoria
