The detection of spatially localised outbreaks in campylobacteriosis notification data

This paper applies a Bayesian hierarchical model designed to identify potential outbreaks of campylobacteriosis from a background of sporadic cases. We assume that such outbreaks are characterized by spatially-localised periods of increased incidence. As well as calculating an outbreak probability for each potential disease cluster, the model simultaneously estimates the underlying spatial and temporal distribution of sporadic cases. The model is applied to notification data from a region of New Zealand for the period 2001–2007 and correctly identifies known outbreaks, whilst highlighting an appropriate number of potential outbreaks for further investigation. Using simulated data, we show that if additional epidemiological information is included in the construction of the model then it can outperform an established method.

► A Bayesian model identifies campylobacteriosis outbreaks from notification data.
► Outbreaks identified via spatio-temporally localized increase in incidence.
► Known and potential outbreaks are identified in data from New Zealand.
► A simulation study shows the approach can outperform the spatial scan statistic.
► Incorporating spatial and epidemiological knowledge enhances detection.