Commentary on" GIS in Public and Environmental Health: Visualisation, Exploration and Modelling." by Anthony Gatrell
Department of Pediatrics
831 83 Östersund, Sweden.
GIS and Public Health
Great leaps of science are often made at the borders where different fields of research meet. GIS in health research may prove to be a good example of this. In many public health and epidemiology projects GIS has recently emerged as an innovative and important component, sometimes even an essential tool. It is easy to determine spatial relationships between disease occurrence and other information that is geo-referenced differently from the disease data.
It is obvious that GIS has a lot to offer the health sciences ( Cliff and Hagget 1988, Gatrell and Löytönen 1998). On the other hand GIS involves concepts and analytic techniques that can lead to uncritical use of the technology. Epidemiology, statistics and geographic information science combined can bring important improvements to medicine and health research. More links should be formed in multidisciplinary research areas working with GIS techniques (Goodchild 1992).
There is little doubt that GIS provides powerful tools for visualising and linking data in public health surveillance. More important is however the epidemiologic analytical capabilities of the present and future GIS. Methodologies for analytical confirmatory analyses are essential. The current state of spatial epidemiology is still mainly descriptive from a statistical point of view. There is, in contrast to other areas of epidemiology, little inference in terms of confidence intervals and hypothesis testing. Thus the main importance of GIS-related methods are still as hypothesis generating. The gap between GIS and appropriate methods for statistical analysis is however currently being filled by an increasing number of software packages.
Environmental epidemiology is a field of increasing importance and improved techniques. The epidemiologist Sarrachi formulated a strategy for future environmental epidemiology, in 1978, bringing in to focus three important issues: 1. Improvements in exposure assessment. 2. Tackling the problem of the combined effect of multiple exposures. 3. Integration of experimental and epidemiologic evidence. Here the issue of exposure assessment is crucial. The questions of timing of exposure and its relationship to induction or promotion of disease including the human individual variation of susceptibility as well as latency periods and interaction with other factors are essential.
There has been a trend in modern epidemiology towards more individual-level focused research. This has raised considerable criticism during the last years (Pearce 1996). In an intense debate it has been claimed that an orientation in the direction of more multidisciplinary approaches to understand the causation of disease in populations is necessary.
In his paper, "GIS in public and environmental health: visualisation, exploration and modelling" Anthony Gatrell presents three classes of methods: Visualisation, exploration and modelling where the first two are closely related and brought together under the concept "exploratory visualisation". Visualization can be used in novel ways to explore the results of traditional statistical analysis. Displaying the locations of outliers and influential values on maps and showing variation in values over space and time can add a great deal to epidemiologic research. Although such tools are being developed and explored, they would benefit greatly from a closer and more seamless link between statistical packages and GIS. The second general class of GIS methods addresses exploratory spatial analyses that allow the analyst to sort meaningfully through spatial data, identify "unusual" spatial patterns, and formulate hypotheses to guide future research. The...