Design of Repeated Sampling for Disease Detection Problem

• Activity Number: 359
• Type: Contributed
• Date/Time: Tuesday, August 6, 2013 : 10:30 AM to 12:20 PM
• Sponsor: Section on Statistics in Epidemiology
• Author(s): Yinan Fang*+ and Chong Wang
• Companies: Iowa State University and Iowa State University

Abstract

Almost every year, animal husbandry all over the world has to deal with the problems coming from outbreaks of some epidemics. In this paper, our goal is to develop a sequential sampling scheme for disease surveillance in a target population. First, we develop geometric distribution models for disease detection by repeated sampling against a stable target prevalence. Binomial model were used to model test outcome in large populations. Furthermore, the S(susceptible)I(infectious) model is introduced to describe the dynamic process of this epidemic. Then under this model, the sequential disease diagnostic testing is conducted and the geometric model is applied to describe detection time of a spreading disease. We theoretically show how sampling frequency as well as sample size influences the test power and the detection time.