In spatio-temporal disease mapping models, identifiability constraints affect PQL and INLA results
- 144 Downloads
Disease mapping studies the distribution of relative risks or rates in space and time, and typically relies on generalized linear mixed models (GLMMs) including fixed effects and spatial, temporal, and spatio-temporal random effects. These GLMMs are typically not identifiable and constraints are required to achieve sensible results. However, automatic specification of constraints can sometimes lead to misleading results. In particular, the penalized quasi-likelihood fitting technique automatically centers the random effects even when this is not necessary. In the Bayesian approach, the recently-introduced integrated nested Laplace approximations computing technique can also produce wrong results if constraints are not well-specified. In this paper the spatial, temporal, and spatio-temporal interaction random effects are reparameterized using the spectral decompositions of their precision matrices to establish the appropriate identifiability constraints. Breast cancer mortality data from Spain is used to illustrate the ideas.
KeywordsBreast cancer INLA Leroux CAR prior PQL Space-time interactions
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). We would like to thank the National Epidemiology Center (area of Environmental Epidemiology and Cancer) for providing the data, originally created by the Spanish Statistical Office. Thanks are also given to two anonymous reviewers for their comments that have contributed to improve the paper.
- Adin A, Martínez-Beneito MA, Botella-Rocamora P, Goicoa T, Ugarte MD (2016) Smoothing and high risk areas detection in space-time disease mapping: a comparison of P-splines, autoregressive, and moving average models. Stoch Environ Res Risk Assess. 31(2):403–415. doi: 10.1007/s00477-016-1269-8 CrossRefGoogle Scholar
- Breslow NE, Clayton DG (1993) Approximate inference in generalized linear mixed models. J Am Stat Assoc 88(421):9–25Google Scholar
- Harville DA (2008) Matrix algebra from a statistician’s perspective, 2nd edn. Springer, New YorkGoogle Scholar
- Leroux BG, Lei X, Breslow N (1999) Estimation of disease rates in small areas: a new mixed model for spatial dependence. In: Halloran M, Berry D (eds) Statistical models in epidemiology, the environment, and clinical trials. Springer, New York, pp 179–192Google Scholar
- Martino S, Rue H (2009) Implementing approximate Bayesian inference using integrated nested laplace approximation: a manual for the inla program. Department of Mathematical Sciences, NTNU, NorwayGoogle Scholar
- R Core Team (2016) R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria, https://www.R-project.org/
- Rue H, Held L (2005) Gaussian Markov random fields: theory and applications, vol 104. CRC press, Boca RatonGoogle Scholar
- Ugarte MD, Goicoa T, Militino AF (2010) Spatio-temporal modeling of mortality risks using penalized splines. Environmetrics 21(3–4):270–289Google Scholar