Community health assessment using self-organizing maps and geographic information systems

BackgroundFrom a public health perspective, a healthier community environment correlates with fewer occurrences of chronic or infectious diseases. Our premise is that community health is a non-linear function of environmental and socioeconomic effects that are not normally distributed among communities. The objective was to integrate multivariate data sets representing social, economic, and physical environmental factors to evaluate the hypothesis that communities with similar environmental characteristics exhibit similar distributions of disease.ResultsThe SOM algorithm used the intrinsic distributions of 92 environmental variables to classify 511 communities into five clusters. SOM determined clusters were reprojected to geographic space and compared with the distributions of several health outcomes. ANOVA results indicated that the variability between community clusters was significant with respect to the spatial distribution of disease occurrence.ConclusionOur study demonstrated a positive relationship between environmental conditions and health outcomes in communities using the SOM-GIS method to overcome data and methodological challenges traditionally encountered in public health research. Results demonstrated that community health can be classified using environmental variables and that the SOM-GIS method may be applied to multivariate environmental health studies.

[1]  N. Pearce,et al.  Environmental epidemiology: challenges and opportunities. , 2000, Environmental health perspectives.

[2]  Helen Couclelis,et al.  Geographies of the information society , 2003 .

[3]  A. McMichael,et al.  Assessing ecosystem health. , 1998, Trends in ecology & evolution.

[4]  N Krieger,et al.  Epidemiology and the web of causation: has anyone seen the spider? , 1994, Social science & medicine.

[5]  Alan Shiell,et al.  A simple guide to chaos and complexity , 2007, Journal of Epidemiology & Community Health.

[6]  A. Clarke,et al.  The causation of disease – The practical and ethical consequences of competing explanations , 2006, Medicine, health care, and philosophy.

[7]  W. Baxt.  Application of artificial neural networks to clinical medicine , 1995, The Lancet.

[8]  M. Marmot.  Improvement of social environment to improve health , 1998, The Lancet.

[9]  P. Elliott,et al.  Spatial Epidemiology: Current Approaches and Future Challenges , 2004, Environmental health perspectives.

[10]  Marie Lynn Miranda,et al.  GIS Modeling of Air Toxics Releases from TRI-Reporting and Non-TRI-Reporting Facilities: Impacts for Environmental Justice , 2004, Environmental health perspectives.

[11]  N. Pearce,et al.  [Traditional epidemiology, modern epidemiology and public health]. , 1996, Epidemiologia e prevenzione.

[12]  T. Greenhalgh,et al.  Complexity and clinical care , 2001, BMJ : British Medical Journal.

[13]  S. Syme,et al.  Rethinking disease: where do we go from here? , 1996, Annals of epidemiology.

[14]  Teuvo Kohonen,et al.  Self-Organizing Maps , 2010 .

[15]  Graham Dunn,et al.  Geographical epidemiology, spatial analysis and geographical information systems: a multidisciplinary glossary , 2007, Journal of Epidemiology and Community Health.

[16]  Daniel Wartenberg,et al.  Investigating disease clusters: why, when and how? , 2001 .

[17]  Rodolfo Saracci,et al.  Everything should be made as simple as possible but not simpler. , 2006, International journal of epidemiology.

[18]  A. McMichael.  From hazard to habitat: rethinking environment and health. , 1999, Epidemiology.

[19]  P. Törönen,et al.  Analysis of gene expression data using self‐organizing maps , 1999, FEBS letters.

[20]  A. J. McMichael,et al.  Integrated assessment of potential health impact of global environmental change: Prospects and limitations , 1997 .

[21]  S. Levin.  Ecosystems and the Biosphere as Complex Adaptive Systems , 1998, Ecosystems.

[22]  Tonny J. Oyana,et al.  Exploration of Geographic Information Systems (gis)-based Medical Databases with Self-organizing Maps (som): a Case Study of Adult Asthma , 2005 .

[23]  T. Poggio,et al.  Multiclass cancer diagnosis using tumor gene expression signatures , 2001, Proceedings of the National Academy of Sciences of the United States of America.

[24]  J. Mesirov,et al.  Molecular classification of cancer: class discovery and class prediction by gene expression monitoring. , 1999, Science.

[25]  Menno-Jan Kraak,et al.  Geovisualization to support the exploration of large health and demographic survey data , 2004, International journal of health geographics.

[26]  Neil Pearce,et al.  Complexity, simplicity, and epidemiology. , 2006, International journal of epidemiology.

[27]  Mikko Kolehmainen,et al.  Insulin resistance syndrome revisited: application of self-organizing maps. , 2002, International journal of epidemiology.

[28]  Ken Sexton,et al.  Assessing Cumulative Health Risks from Exposure to Environmental Mixtures—Three Fundamental Questions , 2007, Environmental health perspectives.

[29]  Marc P. Armstrong,et al.  Exploring the Use of Buffer Analysis for the Identification of Impacted Areas in Environmental Equity Assessment , 1997 .

[30]  David Waltner-Toews,et al.  Evolving Models of Human Health Toward an Ecosystem Context , 1999 .

[31]  Sara Irina Fabrikant,et al.  The European Information Society: Leading the Way with Geo-information, Proceedings of the 10th AGILE Conference, Aalborg, Denmark, 8-11 May 2007 , 2007, AGILE Conf..

[32]  John R. Nuckols,et al.  Using Geographic Information Systems for Exposure Assessment in Environmental Epidemiology Studies , 2004, Environmental health perspectives.