|
Activity Number:
|
305
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Tuesday, August 5, 2008 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statisticians in Defense and National Security
|
| Abstract - #301212 |
|
Title:
|
EWMA Smoothing and Bayesian Spatial Modeling for Health Surveillance
|
|
Author(s):
|
Huafeng Zhou*+ and Andrew B. Lawson
|
|
Companies:
|
Merck & Co., Inc. and University of South Carolina
|
|
Address:
|
UG1D-44, North Wales, PA, 19454,
|
|
Keywords:
|
health surveillance ; spatial model ; exponentially weighted moving average (EWMA) ; Bayesian method ; Markov chain Monte Carlo (MCMC)
|
|
Abstract:
|
The development of appropriate methods of analysis for surveillance of spatial health data is of growing importance for public health applications. The need for relevant and flexible surveillance methods has been stressed many times in the context of bioterrorism. In this study a novel method for the monitoring of disease maps over time in a surveillance setting is developed. The approach relies upon the use of a spatial model which is fitted to current spatial data and is smoothed with historical spatial estimates. The method of smoothing is a vector EWMA procedure. A simulation study with a range of scenarios is presented and finally a case study of monitoring infectious disease spread in the state of South Carolina is presented. Our new approach for health surveillance is easy to implement, and can substantially reduce the computational time.
|
