|
Activity Number:
|
332
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Wednesday, August 11, 2004 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
| Abstract - #301083 |
|
Title:
|
Exploratory Relative Risk-mapping Applied to Regional Data on Echinococcus Multilocularis Infections in Red Foxes
|
|
Author(s):
|
Olaf Berke*+
|
|
Companies:
|
University of Guelph
|
|
Address:
|
Dept. of Population Medicine, Guelph, ON, N1G 2W1, Canada
|
|
Keywords:
|
veterinary ; epidemiology ; Bayes ; kriging ; cluster ; scan test
|
|
Abstract:
|
Spatial epidemiological investigations are often exploratory with limited knowledge about the putative risk factors. Indeed, often the primary motivation for the spatial epidemiological analysis is to identify unknown geographically varying risk factors. An exploratory approach to mapping the spatial relative risk is to scale the risk map by the background risk of the unexposed population. Exposure to unknown spatial risk factors is defined via specific cluster analysis using the spatial scan test. Identification of spatial disease clusters separates the population spatially into inside and outside high risk areas, i.e., the exposed and unexposed populations. To account for unequal regional populations at risk the data were spatially smoothed by empirical Bayesian estimation. Then the latent risk surface map is generated from regional data by universal kriging. This exploratory approach to relative risk-mapping gives the investigator an impression about the importance and geographical distribution of the unknown spatial risk factors. The approach is applied to data on infections of Red foxes with Echinococcus multilocularis from 43 administrative regions in Lower Saxony, Germany.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2004 program |
