This presentation is for public health practitioners and methodology developers interested in using statistical methods to combine evidence from multiple data sources for increased sensitivity to disease outbreaks. Methods described will account for practical issues such as delays in outbreak effects between evidence types. Presented examples will include outbreaks from multiple years of authentic data as will as simulations. The ensuing discussions with attendees will explore the role and scope of multivariate surveillance for the situational awareness of public health monitors.
Presenter
Linus Schiöler, Statistical Research Unit, Department of Economics, University of Gothenburg, Gothenburg, Sweden Article: Schiöler L. and Frisén, M. (2012): Multivariate outbreak detection,
Journal of Applied Statistics, 39:2, 223-242, http://dx.doi.org/10.1080/02664763.2011.584522
Howard Burkom, Johns Hopkins Applied Physics Laboratory, Laurel, Maryland, USA Article: Burkom H. S., Ramac-Thomas L., Babin S., Holtry R., Mnatsakanyan Z. and Yund C. (2011), An integrated approach for fusion of environmental and human health data for disease surveillance. Statistics in Medicine, 30: 470–479. doi: 10.1002/sim.3976
Learning objectives
- Overview of methods using multiple data sources for increased sensitivity to disease outbreaks
- Presentation of examples of outbreaks with authentic data
- Discussion of roles and scope of multivariate analysis