Journal de biométrie et biostatistique

Models for spatial variation in relative disease risk often consider posterior probabilities of elevated disease risk in each area, but for health prioritisation, the interest may also be in the broader clustering pattern across neighbouring areas. The classification of a particular area as high risk may or may not be consistent with risk levels in the surrounding areas. Local join-count statistics are used here in conjunction with Bayesian models of area disease risk to detect different forms of disease clustering over groups of neighbouring areas. A particular interest is in spatial clustering of high risk, which can be assessed by high probabilities of elevated risk across both a focus area and its surrounding locality. An application considers univariate spatial clustering in suicide deaths in 922 small areas in the North West of England, extending to an analysis of bivariate spatial clustering in suicide deaths and hospital admissions for intentional

self-harm in these areas.