A two-stage approach to estimate spatial and spatio-temporal disease risks in the presence of local discontinuities and clusters

Abstract

Disease risk maps for areal unit data are often estimated from Poisson mixed models with local spatial smoothing, for example by incorporating random effects with a conditional autoregressive prior distribution. However, one of the limitations is that local discontinuities in the spatial pattern are not usually modelled, leading to over-smoothing of the risk maps and a masking of clusters of hot/coldspot areas. In this paper, we propose a novel two-stage approach to estimate and map disease risk in the presence of such local discontinuities and clusters. We propose approaches in both spatial and spatio-temporal domains, where for the latter the clusters can either be fixed or allowed to vary over time. In the first stage, we apply an agglomerative hierarchical clustering algorithm to training data to provide sets of potential clusters, and in the second stage, a two-level spatial or spatio-temporal model is applied to each potential cluster configuration. The superiority of the proposed approach with regard to a previous proposal is shown by simulation, and the methodology is applied to two important public health problems in Spain, namely stomach cancer mortality across Spain and brain cancer incidence in the Navarre and Basque Country regions of Spain.

Dealing with risk discontinuities to estimate cancer mortality risks when the number of small areas is large

“…9 and Adin et al. 10 applied the proposed methods to problems with a low/moderate number of areal units. For instance, Adin et al.…”

“…9 and Adin et al. 10 requires to fit as many hierarchical models as small areas, which is computationally prohibitive when estimating risks in Spanish municipalities. Then, a new modeling approach is needed to estimate risks in the presence of local discontinuities when dealing with a large number of small areas.…”

“…Anderson et al 9 and Adin et al 10 applied the proposed methods to problems with a low/moderate number of areal units. For instance, Adin et al 10 analyzed cancer mortality in 47 Spanish provinces, and this low resolution lacks the fine spatial detail of disease risk and may mask the locations of possible hot spots. However, a higher geographical resolution such as municipalities would rocket the number of areal units and could make the method computationally unfeasible.…”

Spatiotemporal Data Clustering: A Survey of Methods

“…Napier, et al [69] proposed a novel Bayesian model to identify the cluster of similar temporal disease trends rather than disease estimation and prediction. Adin, et al [70] proposed a two-stage approach to estimate disease risk maps. Compared with traditional methods, their method has the ability to overcome the problem of local discontinuities in the spatial pattern that cannot be modeled.…”

Online relative risks/rates estimation in spatial and spatio-temporal disease mapping

“…In Table A1 the results obtained when fitting all types of space-time interactions are shown, where all the model selection criteria suggest a Type IV (completely structured) is greater than the risk of the whole of Spain with a high probability. Future development of the application will be the integration of sf (simple feature) objects [47] as cartography files to generate maps, to include interactive data visualization maps using the R package "tmap" [48], and to implement other spatio-temporal proposals such as B-spline models accounting for both spatial and temporal correlation [46], models including age-specific patterns [49], or models to estimate disease risks in the presence of local discontinuities and clusters [50].…”