Title: Early Detection of Disease Outbreaks
Speaker: Martin Kulldorff, Harvard University
Date: April 10, 2006 12:00 - 1:30 pm
Location: DIMACS Center, CoRE Bldg, Room 431, Rutgers University, Busch Campus, Piscataway, NJ
The ability to detect disease outbreaks early is important in order to minimize morbidity and mortality through timely implementation of disease prevention and control measures. Many state and local health departments are setting up systematic syndromic surveillance systems with daily analyzes of hospital emergency department visits, ambulance dispatch calls, pharmacy sales or other health data for which traditional denominators such as census population numbers are unavailable or irrelevant.
We present a space-time permutation scan statistic for the early detection of disease outbreaks that only uses case events, with no need for denominator information. Critically, it makes minimal assumptions about the time, geographical location or geographical size of the outbreak. In a non-parametric fashion, it adjusts for purely spatial and purely temporal variation due to for example consistent non-time varying geographical differences in health care utilization patterns or naturally occurring day-of-week variation. A multivariate version is also presented
The new method is evaluated using daily analyses of hospital emergency department visits in New York City and ambulatory care visits in Boston.
Co-Authors: Renato Assuncao, Luiz Duczmal, Jessica Hartman, Richard Heffernan, Farzad Mostashari, Ken Kleinman, Richard Platt, Katherine Yih.
Bio: Martin Kulldorff is an Associate Professor and biostatistician in the Department of Ambulatory Care and Prevention, Harvard Medical School and Harvard Pilgrim Health Care. His main research interest is the development and application of statistical methods for disease surveillance.
see: DIMACS Computational and Mathematical Epidemiology Seminar Series 2005 - 2006