Canonical correlation analysis: potential for environmental health planning

There is a challenging need to identify the relationships between environmental quality and health status. It may be especially important to be able to isolate key variables which can be consolidated into a few indices of environmental conditions as they are related to health. Such indices might be used to identifying associations among groups of variables, such as specific geographic area. The indices may also provide insights into environmental health relationships which are worthy of further epidemiological investigation. Canonical correlation analysis is a multivariate statistical technique which provides a means of identifying associations among groups of variables, such as health and environmental measures. The technique can produce weighted indices of environmental conditions as they are related to health within a city or region. This paper describes what canonical correlation is, and outlines how it might be used for these purposes. An illustrative application based on data collected for Philadelphia, Pennsylvania is also presented.