Statistics in Medicine · 1998 · 290 citations · 3 references
Public HealthBiostatisticsPreventive MedicineEpidemiologic MethodsEpidemiologySpatial ModelingBayesian Spatial AnalysisBayesian StatisticsGeographic Variation In CancerSpace-time EpidemiologySpatial ConfoundingSpatial Statistical ModelingBayesian Disease ModelingHierarchical Bayesian EstimationSpatially Varying CovariatesTime-varying Covariate AdjustmentTime-varying Covariate EffectsUnmeasured Spatial ConfoundingUnobserved ConfoundingUnobserved Heterogeneity
The Ohio lung cancer data set has attracted attention due to a hypothesis that a southwest nuclear facility may have increased regional cancer rates. The study aims to integrate longitudinal and spatial modeling within a hierarchical Bayesian framework to assess time‑ and space‑varying covariate effects. The authors implement Markov chain Monte Carlo estimation, applying two spatial adjustment strategies—random effects for unobserved heterogeneity and an urbanization proxy for smoking—to reanalyze Ohio lung cancer data from 1968–1988. The authors conclude that the Ohio data are insufficient to properly investigate the alleged nuclear facility–lung cancer link. © 1998 John Wiley & Sons, Ltd.
This paper combines existing models for longitudinal and spatial data in a hierarchical Bayesian framework, with particular emphasis on the role of time- and space-varying covariate effects. Data analysis is implemented via Markov chain Monte Carlo methods. The methodology is illustrated by a tentative re-analysis of Ohio lung cancer data 1968–1988. Two approaches that adjust for unmeasured spatial covariates, particularly tobacco consumption, are described. The first includes random effects in the model to account for unobserved heterogeneity; the second adds a simple urbanization measure as a surrogate for smoking behaviour. The Ohio data set has been of particular interest because of the suggestion that a nuclear facility in the southwest of the state may have caused increased levels of lung cancer there. However, we contend here that the data are inadequate for a proper investigation of this issue. © 1998 John Wiley & Sons, Ltd.
3