Bayesian Methods in Syndromic Surveillance

• Activity Number: 659
• Type: Topic Contributed
• Date/Time: Thursday, August 5, 2010 : 10:30 AM to 12:20 PM
• Sponsor: Section on Bayesian Statistical Science
• Author(s): Jian Zou*+
• Companies: National Institute of Statistical Sciences
• Keywords: syndromic ; surveillance ; spatial-temporal models ; Markov random field

Abstract

Early and accurate detection of outbreaks is one of the most important objectives of syndromic surveillance systems. In spatial statistics, conditional autoregressive (CAR) models have been widely used for areal data. This article proposes a general Bayesian framework for the syndromic surveillance system. This methodology incorporates Gaussian Markov random field (GMRF) and spatial-temporal CAR modeling. This contrasts with previous approaches based on only spatial or time series models. The model has some nice probabilistic representations, as well as attractive statistical properties. We demonstrated that the model is capable of capturing early outbreaks through extensive simulation studies and synthetic data analysis.