Patients, doctors, and major health organizations like the Centers for Disease Control and Prevention (CDC) all agree that engaging in regular physical activity can support health and may play an important role in reducing the risk of chronic health conditions like type 2 diabetes, heart disease, depression and anxiety, and even dementia. Common physical activities like walking, cardio, and biking, are often a primary care physician’s first line of recommendations for patients managing or preventing chronic disease. However, these recommendations rest on the assumption that patients have access to safe, walkable streets, parks, and other communal outdoor spaces. All are aspects of our “built environment,” and they are not created equally.  

“At a time when COVID-19 infection and mortality rates are disproportionately higher among African Americans and Latinos in the USA, the downstream effects of inadequate built environments are particularly evident,” notes lead author Cathy Antonakos, of the University of Michigan School of Kinesiology’s Environment and Policy Lab, and her co-authors in a new paper published in the Journal of Transport and Health. 

IDEAL RRTC researcher Philippa Clarke, Ph.D. contributed to new analysis in two papers that take a closer look at our built environments—namely street “walkability—and the ways in which the accessibility of streets for pedestrians and cyclists that can encourage or discourage physical activity and potentially further entrench health inequities.

Physical Activity and Neighborhood Factors

The first study “Associations between body mass index, physical activity and the built environment in disadvantaged, minority neighborhoods: Predictive validity of GigaPan® imagery” uses a robotic tool called GigaPan to capture images of streets and stitch them together to provide documented “street-views” of a given place. GigaPan was used to document the built environments of two low-income, predominantly Black neighborhoods in Pittsburgh, Pennsylvania, and the research team evaluated the street data, auditing for over 54 attributes that were eventually categorized into three environmental exposure variables: 1.) Amenities for walking and biking like sidewalks and crosswalks; 2.) Presence of steep hills, vacant and boarded up properties; and 3.) Presence of parks, public recreational space, and lower housing density.

The researchers compared the neighborhood street variables with three markers of health to determine if the GigaPan method of assessing neighborhoods could be used to predict residents’ body mass index (BMI), frequency of walking, and average amount of moderate to vigorous exercise. Based on logistic regression modeling, the study concluded:

• Participants living in disadvantaged, mostly African American neighborhoods supportive of walking and cycling had lower BMI. The authors note that this association does not necessarily confirm a causal relationship, writing, “Participants in this study who value healthy lifestyles may have chosen to live in areas that support their desire to live a healthy lifestyle.”
• Participants were less likely to walk in environments with more recreation land use but less housing. The authors note that this association may not be accurate, due to the overall low frequency of facilities and lower inter-rater reliability in accurately identifying facilities based on street imagery. The authors also note that residents in the two neighborhoods studied may be more likely to walk for transportation rather than fitness and recreation. This distinction was not captured in the health survey and interviews conducted.

• None of the environmental exposure variables predicted residents’ moderate to vigorous exercise.

• Variable 2, the presence of steep hills, vacant/boarded lots, was not significantly associated with any of the measured health factors.

Overall, the research concludes that environmental exposures were reliably estimated from audits of high-resolution GigaPan® images. The findings suggest that GigaPan® should be considered for future studies of the built environment and health.The authors also conclude that  built environment changes in low-income, mostly African American communities that promote active transport may improve health. The study was led by Cathy Antonakos of the U-M School of Kinesiology, with Ross Baiers of the RAND Corporation, Tamara Dubowitz and Philippa Clarke, in the Department of Epidemiology at the U-M School of Public Health, and Natalie Colabianchi, at the U-M Institute for Social Research and the U-M School of Kinesiology.

Using Google Street View

The second study, “ Interrater Reliability of Historical Virtual Audits Using Archived Google Street View Imagery,” published in the Journal of Aging and Physical Activity assessed the reliability of historical Google Street View images as a useful data source for street or micro-level information about our built environment. A challenge to understanding the relationship between one's neighborhood built environment and healthy aging is in the quality and cost of environmental factors data. Macro level data exists through organizations like the U.S. Census, and micro-level data can be collected through researchers' on the ground documentation of neighborhoods. However both types of data sources have flaws. Macro-level data often does not provide the granularity needed to reliably assess walkability and street-level accessibility. Micro-level data is robust, but time consuming and often cost prohibitive. Both types of environmental data also provide a static look at environments that are constantly changing. Google Street View is a data source that has been widely utilized and reviewed, and this new study evaluated Google Street View images using a rigorous 78-item checklist that defined neighborhood and residential characteristics. The evaluation yielded high interrater reliability, with over 80% of checklist items having significant agreement across evaluators. The authors conclude, "Environmental assessment using archived virtual imagery has excellent reliability for factors related to residence access and many neighborhood characteristics. Archived imagery can assess past neighborhood characteristics, facilitating the use of historical environment data within existing cohorts."

Article Citations

Antonakos C, Baiers R, Dubowitz T, Clarke P, Colabianchi N. Associations between body mass index, physical activity and the built environment in disadvantaged, minority neighborhoods: Predictive validity of GigaPan® imagery. Journal of Transport & Health. 2020;17:100867. doi:10.1016/j.jth.2020.100867.

Harding AB, Glynn NW, Studenski SA, Clarke PJ, Divecha AA, Rosso AL. Interrater Reliability of Historical Virtual Audits Using Archived Google Street View Imagery. Journal of Aging and Physical Activity. 2020:1-8. doi:10.1123/japa.2019-0331 (in press)