In this paper, we extend the spatially explicit survival model for small area cancer data by allowing dependency between space and time and using accelerated failure time models. Spatial dependency is modeled directly in the definition of the survival, density, and hazard functions. The models are developed in the context of county level aggregated data. Two cases are considered: the first assumes that the spatial and temporal distributions are independent; the second allows for dependency between the spatial and temporal components. We apply the models to prostate cancer data from the Louisiana SEER cancer registry.
KalbfleischJDPrenticeRL. The statistical analysis of failure time data, New York, NY: John Wiley, 1980.
2.
HendersonRShimakuraSGorstD. Modeling spatial variation in leukaemia survival data. J Am Statist Assoc2002; 27: 965–972.
3.
BanerjeeSWallMMCarlinBP. Frailty modeling for spatially correlated survival data, with application to infant mortality in Minnesota. Biostatistics2003; 4: 123–142.
4.
LiYRyanL. Modeling spatial survival data using semiparametric frailty models. Biometrics2002; 58: 287–297.
5.
HennerfeindABrezgerAFahrmeirL. Geoadditive survival models. J Am Statist Assoc2006; 101: 1065–1075.
6.
ZhangJLawsonAB. Bayesian parametric accelerated failure time spatial model and its application to prostate cancer. J Appl Stat2011; 38: 591–603.
7.
Wang S, Zhang J and Lawson AB. A Bayesian normal mixture accelerated failure time spatial model and its application to prostate cancer. Stat Methods Med Res. Epub ahead of print 2012.
8.
Onicescu G, Lawson AB, Zhang J, et al. Spatially-explicit survival modeling for small area cancer data. Manuscript submitted for publication. 2015.
HsuCEMasFSMillerJAet al.A spatial-temporal approach to surveillance of prostate cancer disparities in population subgroups. J Natl Med Assoc2007; 99: 5–7,72–80.
FarrowDCSametJMHuntWC. Regional variation in survival following the diagnosis of cancer. J Clin Epidemiol1996; 49: 843–847.
14.
DickmanPWGibberdRWHakulinenT. Estimating potential savings in cancer deaths by eliminating regional and social class variation in cancer survival in the Nordic countries. J Epidemiol Community Health1997; 51: 289–298.
15.
PostPNDamhuisRAvan der MeydenAP. Variation in survival of patients with prostate cancer in Europe since 1978. EUROCARE Working Group. Eur J Cancer1998; 34: 2226–2231.
16.
YuXQO'ConnellDLGibberdRWet al.Estimating regional variation in cancer survival: a tool for improving cancer care. Cancer Causes Control2004; 15: 611–618.
17.
HigdonDSpace and space-time modeling using process convolutions. In: AndersonCBarnettVChatwinPEl-ShaarawiA (eds). Quantitative methods for current environmental issues, London: Springer, 2002, pp. 37–56.
18.
CalderCA. A dynamic process convolution approach to modeling ambient PM2.5 and PM10 concentration levels. Environmetrics2007; 19: 39–48.
19.
CalderCAHollomanCHigdonDExploring space-time structure in ozone concentration using a dynamic process convolution model. In: GatsonisCKassRECarriquiryAet al. (eds). Case Studies in Bayesian Biostatistics, New York: Springer-Verlag, 2002; vol. 6: 165–176.
20.
Calder CA and Cressie N. Some topics in convolution-based spatial modeling. In: Proceedings of the 56th Session of the International Statistics Institute, August 2007.
21.
SpiegelhalterDJBestNGvan der LindeA. Bayesian measures of model complexity and fit (with discussion). J R Stat Soc Ser B2002; 64: 583–639.
22.
HastingsWK. Monte Carlo sampling methods using Markov chains and their applications. Biometrika1970; 57: 97–109.
23.
Bezanson J, Karpinski S, Shah VB, et al. Julia: A Fast Dynamic Language for Technical Computing, 2012. CoRR abs/12095145. http://arxiv.org/abs/1209.5145 (accessed 17 June 2015).
24.
R Core Team. R: A language and environment for statistical computing. R Foundation for Statistical Computing. Vienna, Austria. http://www.R-project.org/ (2015).
25.
GelmanACarlinJBSternHSet al.Bayesian data analysis, London: CRC Press, 2004.
26.
GelmanARubinDB. Inference from iterative simulation using multiple sequences. Statist Sci1992; 7: 457–511.
27.
WuI. Disparities in prostate cancer in African American men: What primary care physicians can do. Cleve Clin J Med2012; 79: 313–20.
28.
SiegelRMaJZouZet al.Cancer statistics, 2014. CA Cancer J Clin2014; 64: 9–29.
29.
OnegaTDuellEJShiXet al.Influence of place of residence in access to specialized cancer care for African Americans. J Rural Health2010; 26: 12–19.
30.
JemalAWardEWuXet al.Geographic patterns of prostate cancer mortality and variations in access to medical care in the United States. Cancer Epidemiol Biomarkers Prev2005; 14: 590–595.
31.
GregorioDIKulldorffMSheehanTJet al.Geographic distribution of prostate cancer incidence in the era of PSA testing, Connecticut, 1984 to 1998. Urology2004; 63: 78–82.
32.
DivaUBanerjeeSDeyDK. Modelling spatially correlated survival data for individuals with multiple cancers. Stat Modell2007; 7: 191–213.
33.
CarlinBPBanerjeeSHierarchical multivariate CAR models for spatio-temporally correlated survival data. In: BernardoJBDawidAPBergerJOWestM (eds). Bayesian Statistics 7, New York, NY: Oxford University Press, 2003.
34.
BanerjeeSDeyDK. Semiparametric proportional odds models for spatially correlated survival data. Lifetime Data Anal2005; 11: 175–191.
