A “big data” approach to understanding neighborhood effects in chronic illness disparities

In the United States, race and ethnicity, place of residence, and socioeconomic status—which includes income, education, and employment—predict health and illness. Independent, negative health effects are more likely to afflict those who live in areas where a large proportion of residents are socioeconomically disadvantaged.

With funding from the Social Sciences Annual Institute (SSAI), a joint initiative of the U-M Office of Research and Rackham Graduate School, and from the MCubed Diamond initiative, this project will address challenges associated with gathering and analyzing large, diverse population health data sets.

Start date: 5/1/2015
End date: 4/30/2016 

Read More

Initiatives to reduce neighborhood-based health disparities are often hindered by gaps in the country’s population health information infrastructure. These gaps include a lack of accepted indicators of community health and data that are not fully exploited or effectively linked.

Associate Professor of Information Tiffany Veinot will use this project to address challenges associated with gathering and analyzing large, diverse population health data sets by leveraging emerging big data sources such as social media sites, GPS-based data from personal devices, and citizen-created maps.

The study will also include input from organizations that can provide existing data sources, like crime and EMS data, that are routinely collected but not often used for health research. The data will be collected from the 10-county Metropolitan Detroit Combined Statistical Area, which includes the counties of Lapeer, Livingston, Macomb, Oakland, St. Clair, Wayne, Genesee, Monroe, Washtenaw, and Lenawee.

With the data, Veinot and her team of co-investigators will analyze ecological effects of neighborhoods, and seek to find relationships among risk factors such as housing conditions, safety, access to health-enhancing services, and other categories commonly found to contribute to negative health outcomes.  

The project will also assess the feasibility of using GPS-based data collection techniques for measuring chronic illness management activities. Researchers will collect information from participants equipped with a smartphone and GPS tracking device in order to identify places where a person spends a great deal of time. 

The data collected from GPS devices will then be used to create measures of personal exposure to neighborhood risk factors. This portion of the study could provide vital first steps toward more precise modeling of the relationship between neighborhood characteristics and individual chronic illness management behavior.

Veinot is the principal investigator on this project and will be working with an interdisciplinary team of faculty members from schools at U-M:

  • Veronica Berrocal, assistant professor in the School of Public Health Biostatistics Department (MCubed Diamond collaborator)
  • Phillipa Clarke, research affiliate in the Population Studies Center and research assistant professor in the Survey Research Center
  • Robert Goodspeed, assistant professor of urban planning in the Taubman College of Architecture and Urban Planning (MCubed Diamond collaborator)
  • Daniel Romero, assistant professor and Presidential Research Fellow in the School of Information



Office of Research and Rackham Graduate School Social Sciences Annual Institute award, University of Michigan: $49,777

The Social Sciences Annual Institute, a joint initiative of the Office of the Vice President for Research and the Rackham School of Graduate Studies, supports social science research that crosses traditional disciplinary boundaries and advances innovation in the social sciences. SSAI provides seed funds to support research that is not yet likely to be funded through traditional channels.

This project has been approved for additional funding from the MCubed Diamond program. MCubed Diamond is funded by The Alfred P. Sloan Foundation and the Gordon and Betty Moore Foundation as part of the Science of Data Science (SODS) project.  The goal of SODS is to advance the state of data science and build institutional capacity to support it.