Evaluating Statistical Methods for Syndromic Surveillance

Published in: Statistical methods in counterterrorism: game theory, modeling, syndromic surveillance, and biometric authentication / by Alyson G. Wilson, Gregory D. Wilson, and David H. Olwell (New York : Springer, 2006), Chapter 1, p. 1-32

Posted on RAND.org on January 01, 2006

by Michael A. Stoto, Ronald D. Fricker, Arvind Jain, Alexis Diamond, John O. Davies-Cole, Chevelle Glymph, Gebreyesus Kidane, Garrett Lum, LaVerne Jones, Kerda Dehan, et al.

The goals of this chapter are (1) to introduce the statistical issues in syndromic surveillance, (2) to describe and illustrate approaches to evaluating syndromic surveillance systems and characterizing their performance, and (3) to evaluate the performance of a couple of specific algorithms through both abstract simulations and simulations based on actual data. Section 1 of this chapter introduces and discusses the statistical concepts and issues in syndromic surveillance, illustrating them with data from an ER surveillance system from the District of Columbia. Section 2 presents methods from the statistical process control (SPC) literature, including variants on existing multivariate detection algorithms tailored to the syndromic surveillance problem, and compares and contrasts the performance of univariate and multivariate techniques via some abstract simulations. Section 3 then compares Evaluating Statistical Methods for Syndromic Surveillance 143 the new multivariate detection algorithms with commonly used approaches and illustrates the simulation approach to evaluation using simulations based on actual data from seven Washington, DC, hospital ERs. We conclude with a discussion about the implications for public health practice.

This report is part of the RAND Corporation External publication series. Many RAND studies are published in peer-reviewed scholarly journals, as chapters in commercial books, or as documents published by other organizations.

Our mission to help improve policy and decisionmaking through research and analysis is enabled through our core values of quality and objectivity and our unwavering commitment to the highest level of integrity and ethical behavior. To help ensure our research and analysis are rigorous, objective, and nonpartisan, we subject our research publications to a robust and exacting quality-assurance process; avoid both the appearance and reality of financial and other conflicts of interest through staff training, project screening, and a policy of mandatory disclosure; and pursue transparency in our research engagements through our commitment to the open publication of our research findings and recommendations, disclosure of the source of funding of published research, and policies to ensure intellectual independence. For more information, visit www.rand.org/about/principles.

The RAND Corporation is a nonprofit institution that helps improve policy and decisionmaking through research and analysis. RAND's publications do not necessarily reflect the opinions of its research clients and sponsors.