|
Activity Number:
|
242
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 5, 2008 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #301080 |
|
Title:
|
Spatial Event Cluster Detection Using a Normal Approximation
|
|
Author(s):
|
Rhonda Rosychuk*+ and Mahmoud Torabi
|
|
Companies:
|
University of Alberta and University of Alberta
|
|
Address:
|
4th Floor, Rm 9423 Aberhart Centre , Edmonton, AB, T6G 2J3, Canada
|
|
Keywords:
|
disease cluster detection ; normal approximation ; surveillance
|
|
Abstract:
|
Tests for the detection of geographic clusters seek to identify regions that have higher disease rates than expected. These methods are generally applied to cases of disease, but surveillance of disease-related events may also be of interest. Recently, a compound Poisson approach that detects event clusters by testing individual areas (that may be combined with neighbors) was proposed. However, the required probabilities are obtained from a recursion relation that can be cumbersome if the number of events is large or analyses by strata are performed. We propose a simpler approach that uses a normal approximation. This method is easy to implement and is applicable to situations where the population sizes are large and the population distribution by important strata may differ by area. We illustrate the approach on pediatric self-inflicted injury presentations to emergency departments.
|
