The effect of spatial aggregation on performance when mapping a risk of disease

International Journal of Health Geographics

Marcello Pagano³  Al Ozonoff²  Caroline Jeffery¹ 

[1] Liverpool School of Tropical Medicine, Department of International Public Health, Monitoring and Evaluation Technical assistance and Research group, Liverpool L3 5QA, UK;Department of Pediatrics, Harvard Medical School, Center for Patient Safety and Quality Research, Boston Children’s Hospital, Boston, MA 02118, USA;Department of Biostatistics, Harvard School of Public Health, Boston, MA 02115, USA

关键词: Spatial epidemiology;    Simulations;    MAUP;    Scale effect;    Aggregation effect;    Spatial data;    Distance-based mapping;    Disease risk mapping;   

Others  :  804797
DOI  :  10.1186/1476-072X-13-9

 received in 2013-11-22, accepted in 2014-02-07,  发布年份 2014

【 摘 要 】

Background

Spatial data on cases are available either in point form (e.g. longitude/latitude), or aggregated by an administrative region (e.g. zip code or census tract). Statistical methods for spatial data may accommodate either form of data, however the spatial aggregation can affect their performance. Previous work has studied the effect of spatial aggregation on cluster detection methods. Here we consider geographic health data at different levels of spatial resolution, to study the effect of spatial aggregation on disease mapping performance in locating subregions of increased disease risk.

Methods

We implemented a non-parametric disease distance-based mapping (DBM) method to produce a smooth map from spatially aggregated childhood leukaemia data. We then simulated spatial data under controlled conditions to study the effect of spatial aggregation on its performance. We used an evaluation method based on ROC curves to compare performance of DBM across different geographic scales.

Results

Application of DBM to the leukaemia data illustrates the method as a useful visualization tool. Spatial aggregation produced expected degradation of disease mapping performance. Characteristics of this degradation, however, varied depending on the interaction between the geographic extent of the higher risk area and the level of aggregation. For example, higher risk areas dispersed across several units did not suffer as greatly from aggregation. The choice of centroids also had an impact on the resulting mapping.

Conclusions

DBM can be implemented for continuous and discrete spatial data, but the resulting mapping can lose accuracy in the second setting. Investigation of the simulations suggests a complex relationship between performance loss, geographic extent of spatial disturbances and centroid locations. Aggregation of spatial data destroys information and thus impedes efforts to monitor these data for spatial disturbances. The effect of spatial aggregation on cluster detection, disease mapping, and other useful methods in spatial epidemiology is complex and deserves further study.

【 授权许可】

   
2014 Jeffery et al.; licensee BioMed Central Ltd.

【 图 表 】

Figure 1.

Figure 2.

Figure 3.

Figure 4.

Figure 5.

【 参考文献 】

• [1]Koch T: The map as intent: variations on the theme of John Snow. Cartographica: Int J Geogr Inform Geovisualization 2004, 39:1-14.
• [2]Snow J: Report on the Cholera Outbreak in the Parish of St. James, Westminster during the Autumn of 1854, the Cholera Inquiry Committee. London: Churchill; 1855.
• [3]Elliott P: Spatial Epidemiology: Methods and Applications. Oxford: Oxford University Press; 2000.
• [4]Lawson AB, Kleinman K: Spatial And Syndromic Surveillance for Public Health. New York: John Wiley and Sons; 2005.
• [5]Scott DW: Multivariate Density Estimation: Theory, Practice, and Visualization. New York: Wiley-Interscience; 1992.
• [6]Ozonoff A, Jeffery C, Pagano M: Multivariate disease mapping. Proceedings of the American Statistical Association, Biometrics Section [CD-ROM] ASA 2009.
• [7]Jeffery C, Ozonoff A, White LF, Nuno M, Pagano M: Power to detect spatial disturbances under different levels of geographic aggregation. J Am Med Inform Assoc 2009, 16:847-854.
• [8]Jones SG, Kulldorff M: Influence of spatial resolution on space-time disease cluster detection. Plos One 2012, 7(10):e48036. doi:10.1371/journal.pone.0048036
• [9]Waller L: Statistical power and design of focused clustering studies. Stat Med 1996, 15:765-782.
• [10]Gotway CA, Young LJ: Combining incompatible spatial data. J Am Stat Assoc 2002, 97:632-649.
• [11]Diggle PJ: Overview of statistical methods for disease mapping and its relationship to cluster detection. In Spatial Epidemiology: Methods and Applications. Edited by Elliott P, Wakefied JC, Best NG, Briggs DJ. Oxford: Oxford University Press; 2000:87-103.
• [12]Wakefield J, Kelsall J, Morris S: Clustering, cluster detection, and spatial variation in risk. In Spatial Epidemiology: Methods and Applications. Edited by Elliott P, Wakefied JC, Best NG, Briggs DJ. Oxford: Oxford University Press; 2000:128-152.
• [13]Olson KL, Grannis SJ, Mandl KD: Privacy protection versus cluster detection in spatial epidemiology. Am J Publ Health 2002, 2006:96.
• [14]Lawson AB: Tutorial in biostatistics: disease map reconstruction. Stat Med 2001, 20:2183-2204.
• [15]Kelsall JE, Diggle PJ: Non-parametric estimation of spatial variation in relative risk. Stat Med 1995, 14:2335-2342.
• [16]Kelsall JE, Diggle PJ: Spatial variation in risk of disease: a nonparametric binary regression approach. Appl Stat 1998, 47:559-573.
• [17]Jeffery C: Disease Mapping and Statistical Issues in Public Health Surveillance. Cambridge: Harvard University, Biostatistics; 2010.
• [18]Jeffery C, Ozonoff A, White LF, Pagano M: Distance-based mapping of disease risk. Int J Biostat 2013, 9(2):265-290.
• [19]Best N, Richardson S, Thomson A: A comparison of Bayesian spatial models for disease mapping. Stat Methods Med Res 2005, 14:35.
• [20]Kelsall J, Wakefield J: Modeling spatial variation in disease risk: a geostatistical approach. J Am Stat Assoc 2002, 97:692-702.
• [21]Lawson AB: Bayesian disease Mapping: Hierarchical Modeling in Spatial Epidemiology. London: Chapman & Hall; 2008.
• [22]Wakefield J: Disease mapping and spatial regression with count data. Biostatistics 2007, 8:158.
• [23]Waller LA, Gotway CA: Applied Spatial Statistics for Public Health Data. New York : Wiley-IEEE; 2004.
• [24]Lawson AB: Disease Mapping and Risk Assessment for Public Health. New York: Wiley; 1999.
• [25]Turnbull BW, Iwano EJ, Burnett WS, Howe H, Clark LC: Monitoring for clustering of disease: application to leukemia incidence in upstate New York. Am J Epidemiol (Suppl) 1990, 132:S136-S143.
• [26]Waller LA, Turnbull BW, Clark L, Nasca P: Spatial pattern analyses to detect rare disease clusters. In Case Studies in Biometry. Edited by Lange N, Ryan L, Billard L, Brillinger D, Conquest L, Greenhouse J. New York: Wiley; 1994:3-23.
• [27]Kulldorff M, Feuer EJ, Miller BA, Freedma LS: Breast cancer clusters in the Northeast United States: a geographic analysis. Am J Epidemiol 1997, 146:161-170.
• [28]Wand MP, Jones MC: Kernel Smoothing. Hall New York: Chapman; 1995.
• [29]Bellman R: Adaptive Control Processes: a guided tour. Princeton, N.J.: Princeton University Press; 1961.
• [30]Team RC: R: A Language and Environment for Statistical Computing. Computing: R Foundation for Statistical; 2013.
• [31]Lange N: Case Studies in Biometry. New York: Wiley; 1994.
• [32]Gotway CA, Young LJ: A geostatistical approach to linking geographically aggregated data from different sources. J Comput Graph Stat 2007, 16:115-135.
• [33]Ahrens C, Altman N, Casella G, Eaton M, Hwang JTG, Staudenmayer J, Stefanescu C: Leukemia clusters in upstate New York: how adding covariates changes the story. Environmetrics 2001, 12:659-672.
• [34]Bonetti M, Pagano M: The interpoint distance distribution as a descriptor of point patterns, with an application to spatial disease clustering. Stat Med 2005, 24:753-773.
• [35]Kulldorff M, Nagarwalla N: Spatial disease clusters: detection and inference. Stat Med 1995, 14:799-810.
• [36]Ozonoff A, Jeffery C, Manjourides J, White LF, Pagano M: Effect of spatial resolution on cluster detection: a simulation study. Int J Health Geogr 2007, 6:52. BioMed Central Full Text
• [37]Brillinger DR: Spatial-temporal modelling of spatially aggregate birth data. Surv Meth 1990, 16:255-269.
• [38]Brillinger DR: Examples of scientific problems and data analyses in demography, neurophysiology, and seismology. J Comput Graph Stat 1994, 3:1-22.
• [39]Tobler WR: Smooth pycnophylactic interpolation for geographical regions. J Am Stat Assoc 1979, 74:519-530.
• [40]Muller HG, Stadtmuller U, Tabnak F: Spatial smoothing of geographically aggregated data, with application to the construction of incidence maps. J Am Stat Assoc 1997, 92:61-71.