×

Modelling risks in disease mapping. (English) Zbl 1108.62121

Summary: We propose a strategy of analysis of mortality data with the aim of providing a guideline for epidemiologists and public health researchers to choose a reasonable model for estimating mortality (or incidence) risks. Maps displaying the crude mortality rates or ratios are usually misleading because of the instability of the estimators in low populated areas. As an alternative, many smoothing methods have been presented in the literature based on Poisson inference. They account for the extra-Poisson variation (overdispersion), frequently present in the homogeneous Poisson model, by incorporating random effects.
We recommend to test for the potential sources of extra-Poisson variation because, depending on them, the models which fit better the data may be different. Overdispersion can be mainly due to spatial autocorrelation, unstructured heterogeneity or to a combination of these two, and also, when studying very rare diseases, it can be due to an excess of zeros in the data. In this article, different situations the analyst may encounter are detailed and appropriate procedures for each case are presented. The alternative models are illustrated using mortality data provided by the Statistical Institute of Navarra, Spain.

MSC:

62P10 Applications of statistics to biology and medical sciences; meta analysis
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] Mollié A., Markov chain Monte Carlo in practice (1996)
[2] Dean CB., Journal of the American Statistical Association 87 pp 451– (1992) · doi:10.1080/01621459.1992.10475225
[3] Böhning D., Computer-assisted analysis of mixtures and applications (2000)
[4] Manton KG, Biometrics 37 pp 259– (1981) · doi:10.2307/2530416
[5] Clayton DF, Biometrics 43 pp 671– (1987) · doi:10.2307/2532003
[6] Tsutakawa RK., Journal of the American Statistical Association 83 pp 37– (1988) · doi:10.1080/01621459.1988.10478562
[7] Marshall RJ., Applied Statistics 40 pp 283– (1991) · Zbl 0825.62872 · doi:10.2307/2347593
[8] Tsutakawa RK., Biometrics 1 pp 69– (1985) · Zbl 0607.62136 · doi:10.2307/2530644
[9] Besag J, Annals of the Institute of Statistical Mathematics 43 pp 1– (1991) · Zbl 0760.62029 · doi:10.1007/BF00116466
[10] Leroux BG, Statistical models in epidemiology, the environment and clinical trials (1999)
[11] Dean CB, Biometrics 57 pp 197– (2001) · Zbl 1209.62274 · doi:10.1111/j.0006-341X.2001.00197.x
[12] Schlattmann P, Statistics in Medicine 12 pp 1037– (1993)
[13] Militino AF, Statistics in Medicine 20 pp 2035– (2001) · doi:10.1002/sim.821
[14] Breslow NE, Journal of the American Statistical Association 88 pp 9– (1993)
[15] Leroux BG., Statistics in Medicine 19 pp 2321– (2000) · doi:10.1002/1097-0258(20000915/30)19:17/18<2321::AID-SIM572>3.0.CO;2-#
[16] Meza JL., Journal of Statistical Planing and Inference 112 pp 43– (2003) · Zbl 1024.62046 · doi:10.1016/S0378-3758(02)00322-1
[17] Wakefield JC, Spatial epidemiology, methods and applications (2000)
[18] Maiti T., Journal of Statistical Planning and Inference 69 pp 339– (1998) · Zbl 0935.62124 · doi:10.1016/S0378-3758(97)00165-1
[19] Eberly LE, Statistics in Medicine 19 pp 2279– (2000) · doi:10.1002/1097-0258(20000915/30)19:17/18<2279::AID-SIM569>3.0.CO;2-R
[20] Laird NM, Journal of the American Statistical Association 82 pp 739– (1987) · doi:10.1080/01621459.1987.10478490
[21] Farrell P, The Canadian Journal of Statistics 25 pp 75– (1997) · Zbl 1063.62512 · doi:10.2307/3315358
[22] Ugarte MD, Biometrical Journal 45 pp 410– (2003) · doi:10.1002/bimj.200390022
[23] MacNab YC., Biometrics 59 pp 305– (2003) · Zbl 1210.62028 · doi:10.1111/1541-0420.00037
[24] Dempster AP, Journal of the Royal Statistical Society 39 pp 1– (1977)
[25] Böhning D, Disease mapping and risk assessment for public health (1999)
[26] Schlattmann P, Statistics in Medicine 15 pp 919– (1996) · doi:10.1002/(SICI)1097-0258(19960415)15:7/9<919::AID-SIM260>3.0.CO;2-W
[27] Böhning D., Environmetrics 14 pp 431– (2003) · doi:10.1002/env.598
[28] Biggeri A, Computational Statistics & Data Analysis 41 pp 617– (2003) · Zbl 1429.62493 · doi:10.1016/S0167-9473(02)00189-5
[29] Wakefield JC, Bayesian Statistics 6 (1999)
[30] Besag J, Biometrika 82 pp 733– (1995)
[31] Besag J., Journal of the Royal Statistical Society Series B 36 pp 192– (1974)
[32] Dean CB, Computational Statistics and Data Analysis 45 pp 235– (2004) · Zbl 1429.62425 · doi:10.1016/S0167-9473(02)00324-9
[33] Best NG, Bayesian Statistics 6 (1999)
[34] Lawson AB., Statistical methods in spatial epidemiology (2001) · Zbl 1016.62125
[35] MacNab YC, Statistics in Medicine 19 pp 2421– (2000) · doi:10.1002/1097-0258(20000915/30)19:17/18<2421::AID-SIM579>3.0.CO;2-C
[36] Ugarte MD, Stochastic Environmental Research and Risk Assessment 19 pp 33– (2005) · Zbl 1096.62127 · doi:10.1007/s00477-004-0202-8
[37] Van den Broek J., Biometrics 51 pp 738– (1995) · Zbl 0825.62377 · doi:10.2307/2532959
[38] Lachenbruch OA., Statistical Methods in Medical Research 11 pp 297– (2002) · Zbl 1121.62626 · doi:10.1191/0962280202sm289ra
[39] Yau, KKW, Biometrical Journal 45 pp 437– (2003) · doi:10.1002/bimj.200390024
[40] Lee AH, Biometrical Journal 43 pp 963– (2001) · Zbl 0989.62063 · doi:10.1002/1521-4036(200112)43:8<963::AID-BIMJ963>3.0.CO;2-K
[41] Ugarte MD, Biometrical Journal 46 pp 526– (2004) · doi:10.1002/bimj.200310061
[42] Ridout M, Biometrics 57 pp 219– (2001) · Zbl 1209.62079 · doi:10.1111/j.0006-341X.2001.00219.x
[43] Self S, Journal of the American Statistical Association 82 pp 605– (1987) · doi:10.1080/01621459.1987.10478472
[44] Heilbron D., Biometrical Journal 36 pp 531– (1994) · Zbl 0846.62053 · doi:10.1002/bimj.4710360505
[45] Schwarz G., Annals of Statistics 6 pp 461– (1978) · Zbl 0379.62005 · doi:10.1214/aos/1176344136
[46] Akaike H., IEEE Transactions on Automatic Control 19 pp 716– (1974) · Zbl 0314.62039 · doi:10.1109/TAC.1974.1100705
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.