A poisson regression approach for modelling spatial autocorrelation between geographically referenced observations

BMC Medical Research Methodology

Damien Jolley¹  Rory Wolfe¹  Mohammadreza Mohebbi¹ 

[1] Department of Epidemiology and Preventive Medicine, Faculty of Medicine, Nursing and Health Sciences, Monash University, Melbourne, Australia

关键词: Spatial analysis;    Socio-economic status;    Multilevel generalised linear model;    Disease mapping;    Dietary pattern;    Cancer incidence;   

Others  :  1139985
DOI  :  10.1186/1471-2288-11-133

 received in 2011-03-10, accepted in 2011-10-03,  发布年份 2011

【 摘 要 】

Background

Analytic methods commonly used in epidemiology do not account for spatial correlation between observations. In regression analyses, omission of that autocorrelation can bias parameter estimates and yield incorrect standard error estimates.

Methods

We used age standardised incidence ratios (SIRs) of esophageal cancer (EC) from the Babol cancer registry from 2001 to 2005, and extracted socioeconomic indices from the Statistical Centre of Iran. The following models for SIR were used: (1) Poisson regression with agglomeration-specific nonspatial random effects; (2) Poisson regression with agglomeration-specific spatial random effects. Distance-based and neighbourhood-based autocorrelation structures were used for defining the spatial random effects and a pseudolikelihood approach was applied to estimate model parameters. The Bayesian information criterion (BIC), Akaike's information criterion (AIC) and adjusted pseudo R², were used for model comparison.

Results

A Gaussian semivariogram with an effective range of 225 km best fit spatial autocorrelation in agglomeration-level EC incidence. The Moran's I index was greater than its expected value indicating systematic geographical clustering of EC. The distance-based and neighbourhood-based Poisson regression estimates were generally similar. When residual spatial dependence was modelled, point and interval estimates of covariate effects were different to those obtained from the nonspatial Poisson model.

Conclusions

The spatial pattern evident in the EC SIR and the observation that point estimates and standard errors differed depending on the modelling approach indicate the importance of accounting for residual spatial correlation in analyses of EC incidence in the Caspian region of Iran. Our results also illustrate that spatial smoothing must be applied with care.

【 授权许可】

   
2011 Mohebbi et al; licensee BioMed Central Ltd.

【 图 表 】

Figure 1.

Figure 2.

Figure 3.

Figure 4.

【 参考文献 】

• [1]McCullagh P, Nelder JA: Generalized Linear Models. London: Chapman and Hall; 1989.
• [2]Breslow NE, Clayton DG: Approximate inference in generalized linear mixed models. J Am Stat Assoc 1993, 88:9-25.
• [3]Waller L, Gotway C: Applied Spatial Statistics for Public Health Data. New York: Wiley; 2004.
• [4]Iran statistical yearbook. Tehran: Statistical Center of Iran; 2000.
• [5]Reconstruction and estimation of Golestan province population according to 2000 geographic boundaries Tehran: Statistical Center of Iran; 2003.
• [6]Reconstruction and estimation of Mazandaran province population according to 2000 geographic boundaries Tehran: Statistical Center of Iran; 2003.
• [7]Mahboubi E, Kmet J, Cook P, Day N, Ghadirian P, Salmasizadeh S: Oesophageal cancer studies in the Caspian Littoral of Iran:the caspian cancer registry. Br J Cancer 1973, 28:197-214.
• [8]Saidi F, Sepehr A, Fahimi S, Farahvash MJ, Salehian P, Esmailzadeh A, Keshoofy M, Pirmoazen N, Yazdanbod M, Roshan MK: Oesophageal cancer among the Turkomans of northeast Iran. Br J Cancer 2000, 83:1249-1254.
• [9]Mohebbi M, Mahmoodi M, Wolfe R, Nourijelyani K, Mohammad K, Zeraati H, Fotouhi A: Geographical spread of gastrointestinal tract cancer incidence in the Caspian Sea region of Iran: spatial analysis of cancer registry data. BMC Cancer 2008, 8:137. BioMed Central Full Text
• [10]Mohebbi M, Wolfe R, Jolley D, Forbes A, Mahmoodi M, Burton R: The spatial distribution of esophageal and gastric cancer in Caspian region of Iran: an ecological analysis of diet and socio-economic influences. International Journal of Health Geographics 2011, 10:13. BioMed Central Full Text
• [11]Mohebbi M, Nourijelyani K, Mahmoudi M, Mohammad K, Zeraati H, Fotouhi A, Moghadaszadeh B: Time of occurrence and age distribution of digestive tract cancers in northern Iran. Iranian J Publ Health 2008, 37:8-19.
• [12]Income and expenses survey in rural families in 1995 Tehran: Statistical Center of Iran; 1996.
• [13]Income and expenses survey in urban families in 1995 Tehran: Statistical Center of Iran; 1996.
• [14]Anderson T, (Ed.): An introduction to multivariate statistical analysis. New York: John Wiley & Sons; 1984.
• [15]Esteve J, Benhamou E, Raymond L: Descriptive Epidemiology. Volume Iv. Lyon: IARC Scientific Publication; 1994.
• [16]Moran P: Notes on continuous stochastic phenomena. Biometrika 1950, 37:17-23.
• [17]Cressie N: Statistics for Spatial Data, rev. edn. New York: Wiley; 1993.
• [18]Walter SD: The analysis of regional patterns in health data: I. Distributional considerations. Am J Epidemiol 1992, (136):730-741.
• [19]Langford IH, Leyland AH, Rasbash J, Goldstein H: Multilevel modelling of the geographical distributions of diseases. J R Stat Soc Ser C Appl Stat 1999, 48:253-268.
• [20]Clayton D, Kaldor J: Empirical Bayes estimates of age-standardized relative risks for use in disease mapping. Biometrics 1987, 43:671-681.
• [21]Besag J, York J, Mollié A: Bayesian image restoration with two applications in spatial statistics. Ann Inst Stat Math 1991, (43):1--20.
• [22]Wolfinger R, O'Connell M: Generalized linear mixed models a pseudo-likelihood approach. Journal of Statistical Computation and Simulation 1993, 48:233-243.
• [23]Cameron A, Windmeijer F: R2 measures for count data regression models with applications to health-care utilization. Journal of Business and Economic Statistics 1996, 14:209-220.
• [24]SAS/STAT 9.2 User's Guide, Chapter 95: The VARIOGRAM Procedure SAS Publishing; 2008.
• [25]Littell R, Milliken G, Stroup W, Wolfinger R: SAS system for mixed models; Chapter 11: Spatial Variability. 2nd edition. Cary, NC: SAS Institute, Inc; 2006.
• [26]Rasmussen S: Modelling of discrete spatial variation in epidemiology with SAS using GLIMMIX. Computer Methods and Programs in Biomedicine 2004, 76:83-89.
• [27]Monestiez P, Dubroca L, Bonnin E, Durbec JP, Guinet C: Geostatistical modelling of spatial distribution of balaenoptera physalus in the Northwestern Mediterranean sea from sparse count data and heterogeneous observation efforts. Ecological Modelling 2006, 193:615-628.
• [28]Kuha J: AIC and BIC comparisons of assumptions and performance. Sociological Methods & Research 2004, 33:188-229.
• [29]Best NG, Arnold RA, Thomas A, Waller LA, Conlon EM: Bayesian models for spatially correlated disease and exposure data. In Bayesian statistics 6, eds. Edited by Bernardo JM, Berger JO, Dawid AP, Smith AFM. Oxford: Clarendon Press; 1999:131-156.
• [30]Clayton D, Bernardinelli L: Bayesian methods for mapping disease risks. In Methods for Small area studies in geographical and environmental epidemiology. Edited by Cuzick J, Elliott P. Oxford: Oxford University Press; 1992:205-220.
• [31]Guthrie KA, Sheppard L, Wakefield J: A hierarchical aggregate data model with spatially correlated disease rates. Biometrics 2002, 58:898-905.
• [32]Wakefield J, Salway R: A statistical framework for ecological and aggregate studies. Journal of the Royal Statistical Society Series A 2001, 164:119-137.
• [33]Goovaerts P: Geostatistical analysis of disease data: accounting for spatial support and population density in the isopleth mapping of cancer mortality risk using area-to-point Poisson kriging. International Journal of Health Geographics 2006, 5:52. BioMed Central Full Text
• [34]Goovaerts p: Kriging and semivariogram deconvolution in the presence of irregular geographical units. Mathematical Geology 2008, 40:101-128.
• [35]Jang M, Lee Y, Lawson A, Browne W: A comparison of the hierarchical likelihood and Bayesian approaches to spatial epidemiological modelling. Environmetrics 2007, 18:809-821.