Aberration Detection in R Illustrated by Danish Mortality Monitoring

The objective of biosurveillance in this chapter is the detection of emerging incidence clusters in time of a health related event. Reviews on temporal surveillance can be found e.g. in Sonesson and Bock (2003), Bravata et al. (2004), Buckeridge et al. (2005) and Tennant et al. (2007). In recent years a pleasant development has been a synthesis of surveillance methods with methods from statistical process control, see e.g. Woodall (2006) for a survey. One important aspect to ensure a transfer of methodological developments into practice is the availability of appropriate software implementations and their documentation. With the present chapter we want to introduce one such open-source software implementation into a public health context: the R package surveillance. In order to demonstrate functionality, we use Danish mortality data from the ongoing European monitoring of excess mortality for public health action (EuroMOMO) project (Anonymous, 2009). The R system is a free software environment for statistical computing and graphics distributed under a GNU-style copyleft license and running under Unix, Windows and Mac (R Development Core Team, 2009). Several documents and books provide an introduction (Dalgaard, 2008; Venables et al., 2009; Muenchen, 2009). The add-on package surveillance offers functionality for the visualization, monitoring and simulation of count data time series in R for public health surveillance and biosurveillance. It provides an implementation of different aberration detection algorithms for epidemiologists and an infrastructure for developers of new algorithms. The package is freely available under the GNU GPL license and obtainable from the Comprehensive R Archive Network (CRAN). To install the package from CRAN, the following call in R has to be performed once: