The University of Alaska Anchorage (UAA) team has combined the 500 cities dataset with information on natural disaster occurrence to assess the impact of exposure to natural disasters on health and well-being. The team documented their work and findings by creating a research diary that outlined their approach to using the 500 Cities data, cataloging their successes and obstacles for other groups interested in conducting similar research. Moving forward, the team will use the results of their analysis to evaluate the impact of natural disasters in remote Alaskan communities where these data are lacking.

HOW THEY USED 500 CITIES DATA

To begin evaluating the relationship between health outcomes and natural disasters, the University of Alaska Anchorage’s Institute for Circumpolar Health Studies research team conducted an in-depth literature summary centered on different components of the relationship between climate change and health to select health outcomes from the 500 Cities dataset that research shows are most sensitive to natural disasters.

After diving into the literature, UAA selected three 500 Cities health outcomes to evaluate. Of 27 possible indicators, the team chose to evaluate how natural disasters correlate with three stress-related health outcomes: high blood pressure, poor mental health, and poor physical health.

WHAT THEY LEARNED

The University of Alaska Anchorage team noticed that the 500 cities dataset is often used in isolation, city by city, to assess health disparities within a small area. However, their project used the 500 cities dataset at a national scale and links the health outcome data to a nationwide environmental dataset.

That said, the team encountered challenges during their analysis and decided to document their results in a series of project diaries, outlining their project approach and key lessons learned that would be useful for people looking to conduct similar assessments with the 500 cities dataset. All of their research and more information about their project is available on their project website.

In terms of areas for future research, the UAA team is curious if access to evacuation routes or the presence of community disaster plans may influence the resilience of cities that have experienced severe storms or other natural disasters. Future studies could investigate whether communities with more highly developed evacuation strategies recover faster from weather-related environmental damage and have a higher quality of life.

WHAT THEY RECOMMEND FOR SIMILAR PROJECTS

The University of Alaska Anchorage team also explored a variety of natural-disaster data to identify the data best suited for their study, including National Oceanic and Atmospheric Administration storm events, Small Business Administration and Federal Emergency Management Agency data, insurance claims for natural disasters, FEMA flood zones, the Social Vulnerability Index dataset, and US census variables. After carefully reviewing available natural-disaster datasets, the team selected two indicators to inform their study: total insurance claims through the Small Business Administration and Federal Emergency Management Agency.

To evaluate these datasets in tandem, the UAA team had to consider geographic scale in their analysis when linking 500 cities data to other data at different scales (e.g., zip code, county, census tract). Next, the team outlined their methods for data compilation and analysis and presented preliminary results for the relationship between environmental and health outcomes. 

The UAA team has identified the following recommendations for organizations interested in pursuing similar approaches to using the 500 Cities data in their communities:

• Consult the 500 Cities codebook. It was helpful for our team to consolidate 500 Cities codebook information (such as the measure definitions) into one standalone file for reference.
• Standardizing geographic boundaries for analysis is key. When taking a deeper look at 500 Cities census-tract GIS shapefiles, we discovered that the shapes were different from US census-tract GIS shapefiles based on the distribution of residential areas. The need to connect data collected at different political and population-based boundaries (e.g., zip code, census tract) continues to demand consideration. Unlike many continuous environmental datasets, health and social variables are often collected and collated by political or census boundaries. To connect these datasets and examine the relationship between the environment and health, it is necessary to develop methods for merging these disparate data. We would emphasize the importance for future research teams to clearly state assumptions and methods used in connecting the 500 cities dataset with other data collected at different scales.

To learn more about University of Alaska at Anchorage’s work, visit UAA Institute for Circumpolar Health Studies on Facebook. If you are interested in learning more about this team’s project approach or have specific questions regarding replicating their work in your community, please feel free to contact the Principal Investigators of this project, Micah Hahn and Rebecca Van Wyck.