Spatio-Temporal Disease Surveillance: Forward Selection Scan Statistic

Abstract

The scan statistic sets the benchmark for spatio-temporal surveillance methods with its popularity. In its simplest form it scans the target area and time to find regions with disease count higher than expected. If the shape and size of the disease outbreaks are known, then to detect it sufficiently early the scan statistic can design its search area to be efficient for this shape and size. A plan that is efficient at detecting a range of disease outbreak shapes and sizes is important because these vary from one outbreak to the next and are generally never known in advance. This paper offers a forward selection scan statistic that reduces the computational effort on the usual single window scan plan, while still offering greater flexibility in signalling outbreaks of varying shapes. The approach starts by dividing the target geographical regions into a lattice. Secondly it smooths the time series of lattice cell counts using multivariate exponential weighted moving averages. Thirdly, these EWMA cell counts are spatially smoothed to reduce spatial noise and leave the spatial signal. The fourth step uses forward selection approach to scanning mutually exclusive and exhaustive rectangular regions of dynamic dimensions. In the fifth step, it prunes away all insignificant scanned regions where counts are not significantly higher than expected. An outbreak is signaled if at least one region remains after pruning. If all regions are pruned away - including the scan of the target region, then no outbreak is signaled.