Evaluation of the Control of West Nile Virus in Ontario: Did Risk Patterns Change from 2005 to 2012?

The goal of this study was to evaluate and compare the risk distribution of human cases of West Nile virus (WNV) disease in Ontario in 2005 to 2012. The objectives were to: map the risk distribution of WNV in 2005 and 2012, identify clusters of human WNV disease and determine whether the clusters are significantly different between the years 2005 and 2012. West Nile virus surveillance data were used to calculate empirical Bayesian smoothed estimates of disease incidence in southern Ontario for 2005 and 2012. Choropleth maps were generated to visualize the spatial risk distribution, and the spatial scan test was performed to identify clusters of disease. Following identification of clusters for 2005 and 2012, a Poisson model was applied to the 2012 human WNV incidence adjusted for the smoothed human WNV incidence rate from 2005 and the scan test was repeated. Two significant clusters were identified in both the year 2005 and 2012. In 2005, the primary cluster was located in the Windsor-Essex and Chatham-Kent public health units (PHUs). For 2012, the primary cluster was identified in the Golden Horseshoe area. A cluster analysis for 2012 adjusted for those identified in 2005 resulted in one significant cluster in the Windsor-Essex PHU. In 2012, the Windsor-Essex PHU remained as a high-risk area for human WNV disease when compared with the rest of southern Ontario. Although overall risk may change from year to year, public health programming should be employed to decrease the relative risk of WNV in this area.