Geographic information systems and modeling techniques help show how local resources can benefit health.
Digital-mapping technology is all around us and provides an extraordinary tool for measuring our environments. But it’s not always simple to link those measurements to health, says Amy Auchincloss, who works at the intersection of two important trends: interest in how neighborhoods can influence health, and the growing potential of mapping technology.
Although researchers have long used maps to examine environmental health risks, the explosion of digital mapping in recent years has brought cutting-edge technology and vast amounts of data that can potentially be used to assess environmental exposures. Privacy concerns, combined with complex analytic issues, can make spatial analyses of health risk and outcomes difficult, however.
Spatial analyses require researchers to grapple with variations in scale, to fill in missing data, and to make use of spatial dependencies within statistical modeling—issues that tend to be oversimplified. For example, some health researchers have assigned neighborhood environmental exposures based on data that the Census Bureau collected for administrative purposes, which may not be the most relevant measure of health-related risk.
Auchincloss, an assistant research scientist in the Department of Epidemiology, has used sophisticated analytic methods to measure residential neighborhood exposures to healthy foods and physical activity resources. In a study in the January 2008 issue of Epidemiology, she used her carefully derived exposure methods to show that people who lived in an exercise-friendly neighborhood, or who had easy access to healthy food, had lower insulin resistance—a precursor to Type-2 diabetes. Resources that proved “protective” included safe streets, pleasant places to walk, and access to parks and gyms. The relationship between health and neighborhood environment appeared to be partially mediated by health-related behaviors (having a healthier diet, being more physically active) and by the consequences of health behaviors (lower body-mass index).
“Urban environments often develop in the absence of thinking about their relationship with public health,” Auchincloss says. She points out that many cities in the United States developed or “sprawled” without planning for walkability or public transportation. “This study supports work that urban and environmental planners and public-transportation advocates have been doing to design and alter residential areas to promote physical activity.”
Auchincloss notes an underlying motivation behind her work: to identify risks that perpetuate health disparities. Low-income neighborhoods, often populated by persons of color, “tend to be unsuitable environments for being physically active and for purchasing healthy food,” she says. “This study suggests these unsuitable environments may contribute to the high prevalence of Type-2 diabetes that we see in those communities.”
To Learn More
By Mary Beth Lewis
Send correspondence about this or any Findings article to the editor at firstname.lastname@example.org. You will be contacted if your letter is considered for publication.
Left: U.S. Census-based poverty data, as it appears on a map before and after using Auchincloss's spatial smoothing techniques.
- Diabetes is the sixth leading cause of death in the U.S.
- After controlling for individual risk factors, Auchincloss found insulin resistance (a precursor to Type-2 diabetes) to be 17 percent lower for persons living in an exercise-friendly area, as compared to persons living in a low-resource area
- Studies have shown that many people with pre-diabetes can successfully prevent or delay the onset of diabetes by losing five to seven percent of their body weight; this can be accomplished through 30 minutes or more of physical activity most days of the week and by following a low-calorie, low-fat eating plan