Flexible age-specific spatio-temporal Bayesian P-splines models

This repository contains the R code to fit in INLA the age-specific spatio-temporal Bayesian P-splines models described in "Flexible Bayesian P-splines for smoothing age-specific spatio-temporal mortality patterns" (Goicoa et al., 2019).

Table of contents

• Data
• R code
• References

Data {#data}

Simulated female breast cancer mortality data (ICD-10 code 50) in Spanish provinces during the period 1985-2010 by age groups.

• BreastCancer_ESP.txt

  This .txt file contains a data set with the following variables:

  • Age.group: numeric vector of age-group identifiers.
  • Age.label: character vector of age-groups labels.
  • Province: numeric vector of geographic identifiers (Spanish provinces).
  • Year: numeric vector of year's identifiers.
  • Obs: observed number of deaths.
  • Pop: Population at risk.

  Note that the observed number of deaths has been modified to preserve data confidentiality.

• Carto_ESP.Rdata: sf object containing the spatial polygons of the Spanish provinces.

• Esp_prov_nb.graph: An inla.graph object with the spatial neighbourhood structure of the 50 provinces of Spain.

R code {#r-code}

R code to fit with INLA (https://www.r-inla.org/) the age-specific spatio-temporal Bayesian P-splines models described in Goicoa et al. (2019). All the R files are written by the authors of the paper.

• Psplines_INLA.R

  Main script including the required functions to fit in INLA the different age-specific spatio-temporal Bayesian P-splines described in the paper. Note that the obtained results will differ from those shown in the original paper, since simulated data are being used here to preserve data confidentiality.

• Figures_and_Tables.R

  This R script contains the necessary functions to reproduce some of the figures and tables shown in Goicoa et al. (2019). The final model fitted with INLA (Model9-SI with posterior patterns) can be downloaded from https://emi-sstcdapp.unavarra.es/Flexible_Psplines_article/Model9_SI.Rdata.

Acknowledgements

This work has been supported by the Spanish Ministry of Economy and Competitiveness (project MTM2014-51992-R), and by the Health Department of the Navarre Government (Project 113, Res.2186/2014).

References {#references}

Goicoa, T., Adin, A., Etxeberria, J., Militino, A.F., and Ugarte, M.D. (2019). Flexible Bayesian P-splines for smoothing age-specific spatio-temporal mortality patterns. Statistical Methods in Medical Research, 28(2), 384-403.