Bayesian modelling for the integration of spatially misaligned health and environmental data

The analyses of spatially misaligned data sets are on the rise, primarily due to advancements in data collection and merging of databases. This paper presents a flexible and fast Bayesian modelling framework for the combination of data available at different spatial resolutions and from various sources. Inference is performed using INLA and SPDE, which provides a fast approach to fit latent Gaussian models proving particularly advantageous when dealing with spatial and large datasets. The Bayesian modelling approach is demonstrated in a range of health and environmental settings. Specifically, a spatial model is developed to combine point and areal malaria prevalence data, and to integrate air pollution data from different sources. These examples illustrate how to manage data at disparate spatial scales to yield more precise predictions and improved estimation of associations. A spatial model is also specified to estimate the relative risk of lung cancer and assess its relationship with covariates that are misaligned with the response variable. This showcases the model's ability to effectively synthesize misaligned health outcomes and environmental exposure data. These case studies highlight the adaptability of Bayesian spatial methods in overcoming the challenges posed by spatial data misalignment, thus providing valuable insights for decision-making in health and environmental fields.