Spatial Pattern Analysis of Multivariate Disease Data

In geographical epidemiology, outcomes measured at the same spatial location may be correlated, so that the spatial structures of such outcomes across the region under consideration are very similar, perhaps because they reflect the same set of spatially distributed unobserved or unmeasured risk factors. Alternately, one outcome might lead to the presence of another over a region. Most studies fail to account for correlation among multiple outcomes. We demonstrate how a joint outcome modeling approach can improve the predictive accuracy of disease incidence over space in an infectious disease application in the forestry context.