Predicting outbreak detection in public health surveillance: quantitative analysis to enable evidence-based method selection

Public health surveillance is critical for accurate and timely outbreak detection and effective epidemic control. A wide range of statistical algorithms is used for surveillance, and important differences have been noted in the ability of these algorithms to detect outbreaks. The evidence about the...

Full description

Saved in:
Bibliographic Details
Published in:AMIA ... Annual Symposium proceedings 2008-11, Vol.2008, p.76-80
Main Authors: Buckeridge, David L, Okhmatovskaia, Anna, Tu, Samson, O'Connor, Martin, Nyulas, Csongor, Musen, Mark A
Format: Article
Language:English
Subjects:
Algorithms
Communicable Diseases - epidemiology
Disease Notification - methods
Disease Outbreaks - statistics & numerical data
Evidence-Based Medicine
Humans
Population Surveillance - methods
Proportional Hazards Models
Risk Assessment - methods
Risk Factors
United States
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Public health surveillance is critical for accurate and timely outbreak detection and effective epidemic control. A wide range of statistical algorithms is used for surveillance, and important differences have been noted in the ability of these algorithms to detect outbreaks. The evidence about the relative performance of these algorithms, however, remains limited and mainly qualitative. Using simulated outbreak data, we developed and validated quantitative models for predicting the ability of commonly used surveillance algorithms to detect different types of outbreaks. The developed models accurately predict the ability of different algorithms to detect different types of outbreaks. These models enable evidence-based algorithm selection and can guide research into algorithm development.
ISSN:1559-4076