A Flexible Space-Time Scan Statistic for Disease Outbreak Detection and Monitoring

A time periodic geographic disease surveillance system based on a cylindrical space-time scan statistic proposed by Kulldorff [1] has been used extensively for disease surveillance along with the SaTScan software. This statistic is based on a circular spatial scan statistic. On the other hand, many different tests have been proposed to detect purely spatial disease clusters. In particular, some spatial scan statistics such as those developed by Duczmal and Assuncao(2004), Patil and Taillie (2004), and Tango and Takahashi(2005) are aimed at detecting irregularly shaped clusters which may not be detected by the circular spatial scan statistic. However, due to the unlimited geometric freedom of cluster shapes, these statistics have a risk to detect quite large and unlikely peculiarly shaped clusters. A flexible spatial scan statistic proposed by Tango and Takahashi[2], which has been used along with the FleXScan software[3], has a parameter K as the pre-set maximum length of neighbors to be scanned, to be avoid detecting a cluster of unlikely peculiar shape. The flexible spatial scan statistic can be easily extended to space-time alerting methods in syndromic surveillance. Objective: This paper proposes a flexible space-time scan statistic for early detection of disease outbreaks.