The H5N1 avian influenza virus is now considered endemic in poultry in some parts of the world and the continued exposure in humans suggests that the risk of the virus evolving into a more transmissible agent in humans − a step towards worldwide pandemic – remains high. Universities, with large assembly of students and student movements determined by the class schedules and travel routes between classes, in addition to the faculty and staff located in close proximity, are extremely susceptible environments to the spread of pandemic events. Moreover, large universities in the U.S. often have a good proportion of international students, who commute to/from their home country within their study period. Therefore, a good surveillance system to detect disease outbreaks is essential to support a system that is robust to this high impact low probability disruptive event.
Objective
This paper describes a framework for an aberration detection method − change-point analysis for mean and variance − adapted for Poisson-distributed data, for syndromic surveillance in an academic environment.