Abstract:
|
Motivated by population-based geocoded stillbirth and live birth surveillance data for the years 2005-2011, we sought to identify spatio-temporal variation of stillbirth risk. The high-quality surveillance data consisting of point locations of events makes a Bayesian Poisson point process approach natural to consider to evaluate the spatial pattern of events. Due to the large epidemiologic dataset, we implemented the integrated nested Laplace approximation (INLA) to fit the conditional formulation of the point process via a Bayesian hierarchical model and empirically showed that INLA, as opposed to Markov chain Monte Carlo (MCMC) sampling, is an attractive approach. Furthermore, we modeled the temporal variability in stillbirth to better understand how stillbirths are geographically linked over the 7-year study period and demonstrate the similarity between the conditional formulation of the spatio-temporal model and a log-Gaussian Cox process governed by discrete space-time random fields. After controlling for important features of the data, the Bayesian temporal relative risk maps identified areas of increasing and decreasing risk.
|