On the use and Evaluation of Prospective Scan Methods for Health-Related Surveillance

Abstract

We review some prospective scan-based methods that are used in health-related applications to detect increased rates of mortality or morbidity and to detect bioterrorism or active clusters of disease. We relate these methods to the use of the moving average chart in industrial applications. Issues that are related to the performance evaluation of spatiotemporal scan-based methods are discussed. In particular we clarify the definition of a recurrence interval and demonstrate that this measure does not reflect some important aspects of the statistical performance of scan-based, and other, surveillance methods. Some research needs in this area are given. Copyright 2008 Royal Statistical Society.

A comparison of CUSUM, EWMA, and temporal scan statistics for detection of increases in poisson rates

“…When used for online monitoring, scan statistic methods should be modified for the detection of a rate change 12 . A comprehensive review of the use of scan statistic methods for online monitoring can be found in Woodall et al 13 . When used in online monitoring, scan statistic methods that use a fixed window size are indeed moving average control charts 12 .…”

Spatio-Temporal Disease Surveillance

“…As such the results from Chen and Glaz can be used to get efficient starting estimates for the threshold used in the surveillance plan. For other literature on the SCAN plan see Tango (1995), Kulldorff and Nagarwalla (1995), Kulldorff (1997), and Kulldorff, M. (2001), Kulldorff et al (2005), Woodall et al (2008) and Han et al (2008). Fig.…”

“…This method is a spatio-temporal moving average plan that systematically scans the target space applying a test to all windows of data up to a given fixed size in time and space. This presents an intuitive approach but has received some criticisms by Woodall et al (2008) and Han et al (2008), including: firstly that it is not as efficient as the cumulative SUM (CUSUM) (see Raubertas, 1989, andYamada, 2004) for outbreak detection and secondly that its ability to detect outbreaks most effectively is dependent on the choice of shape and size of the scanning window. In addition, in some of the literature on the SCAN statistic, the focus is on whether diseases cluster rather than on early detection.…”

A one‐sided MEWMA chart for health surveillance

“…Some methods for prospective spatial surveillance have been presented or summarized by Raubertas 9 , Rogerson 10,11 , Kulldorff 12 , and Woodall et al 13 , among others. A good review of spatial surveillance methods is given by Shmueli and Fienberg 14 .…”