Authors: Arie Manangan*, CDC
Topics: Hazards and Vulnerability, Geographic Information Science and Systems, Climatology and Meteorology
Keywords: Health Vulnerability
Session Type: Paper
Start / End Time: 5:20 PM / 7:00 PM
Room: Napoleon B1, Sheraton 3rd Floor
Presentation File: No File Uploaded
Methods: We constructed three vulnerability indices using various techniques based on previous studies: 1) Principal Components Analysis (PCA), 2) percentile ranking, and 3) overlay analysis. To determine if these heat vulnerability indices vary with actual cases of heat-related illness, we calculated a Spearman’s rank correlation coefficient using heat-related death data from the state of Georgia from 2002-2009, at the county level. To compare the strength of each of the indices, we calculated a linear regression model using each of the vulnerability indices with the heat-related illness data. Results: Our results show that a heat-vulnerability index based on an overlay analysis approach detects a statistically significant correlation with actual cases of heat-related illness (p-value = 7.4e-10). Alternatively, a percentile ranking approach also shows a significant statistical relationship (p-value= 1.0e-07), but a comparison of the two methods shows that the linear regression model of overlay analysis approach has greater significance (i.e. smaller p-value) and tighter confidence intervals, when looking specifically at heat-related illness. Conclusion: Our results suggest that a heat-vulnerability index based on an overlay analysis approach does co-vary with the geographic distribution of actual cases of heat-related illness, and may be an effective tool for identifying geographic regions that may be more vulnerable to the health effects of extreme heat exposure.