Integrating Geographic Information Systems and Spatial Analysis into Public Health Applications
by Jacqueline Montain © 2001

This research project includes a comparison of Geographic Information System (GIS) implementation at three selected public health organizations in Ontario. A series of recommendations based upon these experiences may be considered by other public health organizations interested in the use of such technologies. GIS and advanced spatial statistics are also utilized for the analysis of public health information.

Data for the project were obtained from the Ontario Ministry of Heath and Long Term Care Provincial Health Planning System database. The data were aggregated to the Statistics Canada Census Division (CD) level, and Standardized Mortality Ratio (SMR) values were computed for each of these geographic units. Spatial autocorrelation coefficients of Moran's I and Geary's C were then calculated to determine the extent of clustering in mortality due to ischemic heart disease, lung cancer, and cerebrovascular disease for census divisions during the years of 1996 and 1997. Some evidence of significant positive spatial autocorrelation was found in the SMR values for each of the conditions during the two years of analysis. There were however, differences in the results of I and C and measures of significance depending on the method of neighbour weighting scheme used.