|
Activity Number:
|
144
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 7, 2006 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Statistics in Epidemiology
|
| Abstract - #307392 |
|
Title:
|
A Wavelet-Based Method for the Prospective Monitoring of Disease Incidence Counts in Space and Time
|
|
Author(s):
|
J. Brooke Marshall*+ and Dan Spitzner and William H. Woodall
|
|
Companies:
|
Virginia Polytechnic Institute and State University and Virginia Polytechnic Institute and State University and Virginia Polytechnic Institute and State University
|
|
Address:
|
403 Progress Street, NE, Blacksburg, VA, 24060,
|
|
Keywords:
|
public health surveillance ; wavelets ; Poisson regression ; disease clusters ; control charts ; ARL performance
|
|
Abstract:
|
In epidemiology it is useful to monitor disease occurrences prospectively to allow for quicker detection of disease clusters. Here we present a prospective method for monitoring disease occurrences in a geographical region. In this method, a surface of incidence counts is modeled over time in the region of interest. This surface is modeled using Poisson regression where the regressors are functions from the Haar wavelet basis. The surface is estimated each time incidence data is obtained using both past and current observations, weighing current observations more heavily. The flexibility of this method allows for the detection of changes in the incidence surface, increases in the overall mean incidence count, and clusters of disease within smaller areas of the region, by using control charts. The control limits for these charts are determined by average run length performance.
|
