Data Merging and Visualization to Identify Associations between Environmental Factors and Disease Outbreaks

This paper describes data merging and visualization techniques for epidemiological and environmental surveillance data. The ultimate goal is to learn about associations between specific environmental factors and disease outbreaks. In such studies, environmental and clinical surveys often occur on different timelines. As such, data merging for the purpose of correlating the two data series can be difficult and subjective. Furthermore, scientists are often interested in exploring chronological lags between the series, making merging more complicated. Visualization of the data series by means of overlaid scatterplots and other multi-dimensional graphics, and exploratory quantification of possible lags and correlations is an important first step in building predictive models. We illustrate data merging with PROC SQL to merge environmental and clinical data with chronological lags. We use graph template language (GTL) to demonstrate data visualizations and correlations that enabled us to identify potential associations between cases of the disease and environmental variables, with a variety of possible lags. Results included in this paper were produced using SAS 9.3 on a Windows XP platform, using Base SAS, SAS/STAT software, and SAS/GRAPH. SAS 9.2 or later is required for ODS graphics extensions.