Environmental and Ecological Statistics

, Volume 12, Issue 3, pp 259–273 | Cite as

Local likelihood disease clustering: development and evaluation



This paper illustrates a method based on local likelihood (LL) for detecting disease clusters. The approach is based on estimating a lasso distance for each region: within which regions are considered to be clustered. An important advantage in implementing this approach is that it does not require any special Monte Carlo Markov Chain (MCMC) algorithm, e.g., reversible jump MCMC, which is essential in hidden Markov model approach. Another advantage is that extending the model to incorporate covariates is straightforward. We illustrate three ways of doing this by using Eastern Germany lip cancer data. By using simulated data, we have made a comparison with the BYM model [Besag et al. (1991) Annals of the Institute of Statistical Mathematics, 43, 1–59] and the mixture model [Lawson and Clark (2002) Disease Mapping and Risk Assessment for Public Health, Chapman and Hall]. We also did a limited examination of the ability of the LL model to recover true relative risk under different priors for lasso parameter. In order to check the edge effects, which has been overlooked in many spatial clustering models for disease mapping but deserves special attention as it lacks observable neighbors, we have adapted here a simple approach to observe the changes in relative risks when the edge regions are omitted.


absolute difference prior BYM model CAR prior covariate dependence disease clustering edge effect local likelihood mixture model 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Besag, J., York, J., Mollie, A. 1991Bayesian image restoration with two applications in spatial statistics (with discussion)Annals of the Institute of Statistical Mathematics43159Google Scholar
  2. Cressie, N.C., Lawson, A.B. 2000Hierarchical probability models and Bayesian analysis of mine locationsAdvances in Applied Probability32315330CrossRefGoogle Scholar
  3. Green, J., Richardson, S. 2002Hidden Markov models and disease mappingJournal of the American Statistical Association9710551070CrossRefGoogle Scholar
  4. Kauermann, G., Opsomer, J.D. 2003Local likelihood estimation in generalized additive modelsScandinavian Journal of Statistics30317337CrossRefGoogle Scholar
  5. Kunsch, H.R. 1987Intrinsic autoregressions and related models on the two-dimensional latticeBiomerika74517524Google Scholar
  6. Lawson, A.B. 1993Discussion contribution on the Gibbs sampler and other Markov chain Monte Carlo methodsJournal of the Royal Statistical Society, Series B556162Google Scholar
  7. Lawson, A.B. 1995Markov chain Monte Carlo methods for putative pollution source problems in environmental epidemiologyStatistics in Medicine1424732486PubMedGoogle Scholar
  8. Lawson, A.B. 2000Cluster modelling of disease incidence via RJMCMC methods: a comparative evaluationStatistics in Medicine1923612376PubMedGoogle Scholar
  9. Lawson, A.B. 2002Spatial mixture relative risk models applied to disease mappingStatistics in Medicine21359370PubMedGoogle Scholar
  10. Lawson, A.B. (2005) Cluster detection: a critique and a Bayesian proposal. Statistics in Medicine. To appearGoogle Scholar
  11. Lawson, A.B. and Clark, A.B. (1999) Markov chain Monte Carlo methods for putative sources of Hazard and general clustering. In Disease Mapping and Risk Assessment for Public Health, A.B. Lawson, D. Bohning, E. Lessafre, A. Biggeri, J.F. Viel and R. Bertollini (eds), Wiley. 119–141Google Scholar
  12. Lawson, A.B. and Denison, D. (2002) Spatial cluster modelling: an overview. In Spatial Cluster Modelling, A.B. Lawson and D. Denison (eds), Chapman and Hall. 1–19Google Scholar
  13. R Development Core Team (2003) R: a language and environment for statistical computing. In R Foundation for Statistical Computing, Vienna, AustriaGoogle Scholar
  14. Spiegelhalter, D., Best, N., Carlin, B., Linde, A. 2002Bayesian measures of model complexity and fit (with discussion)Journal of the Royal Statistical Society, Series B64583639Google Scholar
  15. Spiegelhalter, D., Thomas, A., Best, N., and Lunn, D. (2003) WinBUGS User Manual. In MRC Biostatistics Unit, Institute of Public Health, Cambridge, UKGoogle Scholar
  16. Sun, D., Tsutakawa, R.K., Kim, H., He, Z. 2000Spatio-temporal interaction with disease mappingStatistics in Medicine1920152035CrossRefPubMedGoogle Scholar
  17. Vidal-Rodeiro, C.L. and Lawson, A.B. (2005) An evaluation of the edge effects in disease map modelling. Computational Statistics and Data Analysis49, 45–62Google Scholar
  18. Yu, B., Mykland, P. 1998Looking at Markov sampler through Cusum path plots: a simple diagnostic ideaStatistics and Computing8275286CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media,Inc. 2005

Authors and Affiliations

  1. 1.Department of Epidemiology & Biostatistics, Arnold School of Public HealthUniversity of South CarolinaColumbiaUSA

Personalised recommendations