Abstract
The increase in Bayesian models available for disease mapping at a small area level can pose challenges to the researcher: which one to use? Models may assume a smooth spatial surface (termed global smoothing), or allow for discontinuities between areas (termed local spatial smoothing). A range of global and local Bayesian spatial models suitable for disease mapping over small areas are examined, including the foundational and still most popular (global) Besag, York and MolliƩ (BYM) model through to more recent proposals such as the (local) Leroux scale mixture model. Models are applied to simulated data designed to represent the diagnosed cases of (1) a rare and (2) a common cancer using small-area geographical units in Australia. Key comparative criteria considered are convergence, plausibility of estimates, model goodness-of-fit and computational time. These simulations highlighted the dramatic impact of model choice on posterior estimates. The BYM, Leroux and some local smoothing models performed well in the sparse simulated dataset, while centroid-based smoothing models such as geostatistical or P-spline models were less effective, suggesting they are unlikely to succeed unless areas are of similar shape and size. Comparing results from several different models is recommended, especially when analysing very sparse data.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
A. Adin, M.A. Martinez-Beneito, P. Botella-Rocamora, T. Goicoa, M.D. Ugarte, 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), 403ā415 (2017). https://doi.org/10.1007/s00477-016-1269-8
C. Anderson, L. Ryan, A Comparison of spatio-temporal disease mapping approaches including an application to ischaemic heart disease in New South Wales, Australia. Int. J. Environ. Res. Public Health 14(2), 146 (2017). https://doi.org/10.3390/ijerph14020146
Australian Bureau of Statistics [ABS], Census of Population and Housing: Socio-Economic Indexes for Areas (SEIFA), Australia, āStatistical Area Level 2, Indexes, SEIFA 2011ā, data cube: Excel spreadsheet, cat. no. 2033.0.55.001. (2013). www.abs.gov.au/AUSSTATS/abs.nsf/DetailsPage/2033.0.55.0012011
Australian Bureau of Statistics [ABS], Australian Statistical Geography Standard (ASGS): Volume 1 - Main structure and greater capital city statistical areas, July 2011, cat. no. 1270.0.55.001 (2013). www.abs.gov.au/AUSSTATS/abs.nsf/DetailsPage/1270.0.55.001Jul%202011
S. Banerjee, B.P. Carlin, A.E. Gelfand, Hierarchical Modeling and Analysis for Spatial Data. Monographs on Statistics and Applied Probability, 2nd edn., vol. 135 (CRC Press/Chapman & Hall, Boca Raton, 2014)
J. Besag, Spatial interaction and the statistical analysis of lattice systems. J. R. Stat. Soc. Ser. B (Stat. Methodol.) 36(2), 192ā236 (1974)
J. Besag, C. Kooperberg, On conditional and intrinsic autoregressions. Biometrika 82(4), 733ā746 (1995). https://doi.org/10.1093/biomet/82.4.733
J. Besag, J. York, A. MolliĆ©, Bayesian image restoration with application in spatial statistics. Ann. Inst. Stat. Math. 43(1), 1ā20 (1991). https://doi.org/10.1007/BF00116466
N. Best, S. Cockings, J. Bennett, P. Wakefield, Ecological regression analysis of environmental benzene exposure and childhood leukaemia: sensitivity to data inaccuracies, geographical scale and ecological bias. J. R. Stat. Soc. Ser. A (Stat. Soc.) 164 (1), 155ā174 (2001). https://doi.org/10.1111/1467-985x.00194
N. Best, S. Richardson, A. Thomson, A comparison of Bayesian spatial models for disease mapping. Stat. Methods Med. Res. 14(1), 35ā59 (2005). https://doi.org/10.1191/0962280205sm388oa
A.C.A. Clements, N.J.S. Lwambo, L. Blair, U. Nyandindi, G. Kaatano, S. Kinungāhi, J.P. Webster, A. Fenwick, S. Brooker, Bayesian spatial analysis and disease mapping: tools to enhance planning and implementation of a schistosomiasis control programme in Tanzania. Trop. Med. Int. Health 11 (4), 490ā503 (2006). https://doi.org/10.1111/j.1365-3156.2006.01594.x
P. Congdon, Representing spatial dependence and spatial discontinuity in ecological epidemiology: a scale mixture approach. Stoch. Environ. Res. Risk Assess. 31(2), 291ā304 (2017). https://doi.org/10.1007/s00477-016-1292-9
S.M. Cramb, K.L. Mengersen, P.D. Baade, Identification of area-level influences on regions of high cancer incidence in Queensland, Australia: a classification tree approach. BMC Cancer 11, 311 (2011). https://doi.org/10.1186/1471-2407-11-311
D.G.T. Denison, C.C. Holmes, Bayesian partitioning for estimating disease risk. Biometrics, 57(1), 143ā149 (2001). https://doi.org/10.1111/j.0006-341x.2001.00143.x
L.E. Eberly, B.P. Carlin, Identifiability and convergence issues for Markov chain Monte Carlo fitting of spatial models. Stat. Med. 19(17), 2279ā2294 (2000). https://doi.org/10.1002/1097-0258(20000915/30)19:17/18<2279::aid-sim569>3.0.co;2-r
J. Geweke, Evaluating the accuracy of sampling-based approaches to the calculation of posterior moments, in Bayesian Statistics, vol. 4, ed. by J.M. Bernardo, J. Berger, A.P. Dawid, A.F.M. Smith (Oxford University Press, Oxford, 1992), pp. 169ā193
T. Goicoa, M.D. Ugarte, J. Etxeberria, A.F. Militino, Comparing CAR and P-spline models in spatial disease mapping. Environ. Ecol. Stat. 19(4), 573ā599 (2012). https://doi.org/10.1007/s10651-012-0201-8
P.J. Green, S. Richardson, Hidden Markov models and disease mapping. J. Am. Stat. Assoc. 97(460), 1055ā1070 (2002). https://doi.org/10.1198/016214502388618870
C. Kandhasamy, K. Ghosh, Relative risk for HIV in Indiaāan estimate using conditional auto-regressive models with Bayesian approach. Spatial Spatio-temporal Epidemiol. 20, 13ā21 (2017). https://doi.org/10.1016/j.sste.2017.01.001
S.Y. Kang, S.M. Cramb, N.M. White, S.J. Ball, K.L. Mengersen, Making the most of spatial information in health: a tutorial in Bayesian disease mapping for areal data. Geospat. Health 11(2), 428 (2016). https://doi.org/10.4081/gh.2016.428
L. Knorr-Held, G. RaĆer, Bayesian detection of clusters and discontinuities in disease maps. Biometrics 56(1), 13ā21 (2000). https://doi.org/10.1111/j.0006-341x.2000.00013.x
S. Lang, A. Brezger, Bayesian P-Splines. J. Comput. Graph. Stat. 13(1), 183ā212 (2004). https://doi.org/10.1198/1061860043010
A.B. Lawson, A. Clark, Spatial mixture relative risk models applied to disease mapping. Stat. Med. 21(3), 359ā370 (2002). https://doi.org/10.1002/sim.1022
D. Lee, A comparison of conditional autoregressive models used in Bayesian disease mapping. Spatial Spatio-temporal Epidemiol. 2(2), 79ā89 (2011). https://doi.org/10.1016/j.sste.2011.03.001
D. Lee, CARBayes Version 4.7: An R package for Spatial Areal Unit Modelling with Conditional Autoregressive Priors (University of Glasgow, Glasgow, 2017). https://CRAN.R-project.org/package=CARBayes
D. Lee, R. Mitchell, Boundary detection in disease mapping studies. Biostatistics 13 (3), 415ā426 (2012). https://doi.org/10.1093/biostatistics/kxr036
D. Lee, R. Mitchell, Locally adaptive spatial smoothing using conditional auto-regressive models. J. R. Stat. Soc. Ser. C (Appl. Stat.) 62(4), 593ā608 (2013). https://doi.org/10.1111/rssc.12009
D. Lee, C. Sarran, Controlling for unmeasured confounding and spatial misalignment in long-term air pollution and health studies. Environmetrics 26(7), 447ā487 (2015). https://doi.org/10.1002/env.2348
B.G. Leroux, X. Lei, N. Breslow, Estimation of disease rates in small areas: a new mixed model for spatial dependence. Stat. Models Epidemiol. Environ. Clin. Trials 116, 179ā191 (2000). https://doi.org/10.1007/978-1-4612-1284-3_4
H. Lu, C. Reilly, S. Banerjee, B. Carlin, Bayesian areal wombling via adjacency modelling. Environ. Ecol. Stat. 14(4), 433ā452 (2007). https://doi.org/10.1007/s10651-007-0029-9
D.J. Lunn, A. Thomas, N. Best, D. Spiegelhalter, WinBUGSāa Bayesian modelling framework: concepts, structure, and extensibility. Stat. Comput. 10(4), 325ā337 (2000). https://doi.org/10.1023/a:1008929526011
Y.C. MacNab, Spline smoothing in Bayesian disease mapping. Environmetrics 18(7), 727ā744 (2007). https://doi.org/10.1002/env.876
T.G. Martins, D. Simpson, F. Lindgren, H. Rue, Bayesian computing with INLA: new features. Comput. Stat. Data Anal. 67, 68ā83 (2013). https://doi.org/10.1016/j.csda.2013.04.014
O. Mersmann, microbenchmark: accurate timing functions. R package version 1.4-6 (2018). http://CRAN.R-project.org/package=microbenchmark
P.A.P. Moran, Notes on continuous stochastic phenomena. Biometrika 37(1/2), 17ā23 (1950). https://doi.org/10.2307/2332142
F.S. Nathoo, P. Ghosh, Skew-elliptical spatial random effect modeling for areal data with application to mapping health utilization rates. Stat. Med. 32(2), 290ā306 (2013). https://doi.org/10.1002/sim.5504
A. Perperoglou, P.H.C. Eilers, Penalized regression with individual deviance effects. Comput. Stat. 25(2), 341ā361 (2010). https://doi.org/10.1007/s00180-009-0180-x
M. Plummer, JAGS version 4.3.0 user manual (2017). https://CRAN.R-project.org/package=CARBayes
R Core Team, R: A language and environment for statistical computing. R Foundation for Statistical Computing (2018). http://www.R-project.org/
S. Richardson, Statistical methods for geographical correlation studies, in Geographical and Environmental Epidemiology, ed. by P. Elliot, J. Cuzick, D. English, R. Stern (Oxford University Press, Oxford, 1996), pp. 181ā204
A. Riebler, S.H. SĆørbye, D. Simpson, H. Rue, An intuitive Bayesian spatial model for disease mapping that accounts for scaling. Stat. Methods Med. Res. 25(4), 1145ā1165 (2016). https://doi.org/10.1177/0962280216660421
D. Ruppert, M.P. Wand, R.J. Carroll, Semiparametric Regression (Cambridge University Press, Cambridge, 2003)
D.J. Spiegelhalter, N.G. Best, B.P. Carlin, V. der Linde, Bayesian measures of model complexity and fit (with discussion). J. R. Stat. Soc. Ser. B (Stat. Methodol.) 64 (4):583ā640 (2002). https://doi.org/10.1111/1467-9868.00353
S. Sturtz, U. Ligges, A. Gelman, R2WinBUGS: a package for running WinBUGS from R. J. Stat. Softw. 12(3), 1ā16 (2005). https://doi.org/10.18637/jss.v012.i03
Y.-S. Su, M. Yajima, R2jags: using R to run āJAGSā. R package version 0.5-7 (2015). https://CRAN.R-project.org/package=R2jags
A. Vehtari, A. Gelman, J. Gabry, Practical Bayesian model evaluation using leave-one-out cross-validation and WAIC. Stat. Comput. 27(5), 1413ā1432 (2017). https://doi.org/10.1007/s11222-016-9709-3
L.A. Waller, B.P. Carlin, Disease mapping, in Handbook of Spatial Statistics, ed. by A.E. Gelfand, P.J. Diggle, P. Guttorp, M. Fuentes (CRC Press, Boca Raton, 2010)
S. Watanabe, Asymptotic equivalence of Bayes cross validation and widely applicable information criterion in singular learning theory. J. Mach. Learn. Res. 11, 3571ā3594 (2010)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
Ā© 2020 The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG
About this chapter
Cite this chapter
Cramb, S., Duncan, E., Baade, P., Mengersen, K.L. (2020). A Comparison of Bayesian Spatial Models for Cancer Incidence at a Small Area Level: Theory and Performance. In: Mengersen, K., Pudlo, P., Robert, C. (eds) Case Studies in Applied Bayesian Data Science. Lecture Notes in Mathematics, vol 2259. Springer, Cham. https://doi.org/10.1007/978-3-030-42553-1_10
Download citation
DOI: https://doi.org/10.1007/978-3-030-42553-1_10
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-42552-4
Online ISBN: 978-3-030-42553-1
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)