Google Street View Derived Built Environment Indicators and Associations with State-Level Obesity, Physical Activity, and Chronic Disease Mortality in the United States

• Published in: International journal of environmental research and public health 2020-05, Vol.17 (10), p.3659
• Main Authors: Phan, Lynn, Yu, Weijun, Keralis, Jessica M, Mukhija, Krishay, Dwivedi, Pallavi, Brunisholz, Kimberly D, Javanmardi, Mehran, Tasdizen, Tolga, Nguyen, Quynh C
• Format: Article
• Language: English
• Subjects: Accuracy; Application programming interface; Body mass index; Built Environment; Cardiovascular disease; Cardiovascular diseases; Chronic Disease; Chronic illnesses; Construction; Data collection; Diabetes; Diabetes mellitus; Environment Design; Environments; Exercise; Humans; Indicators; Mortality; Mortality - trends; Neighborhoods; Obesity; Parks & recreation areas; Physical activity; Physical fitness; Public health; Regression analysis; Residence Characteristics; Roads & highways; Structured Query Language-SQL; Teenagers; United States - epidemiology; Urban areas; Urban environments
• ISSN: 1660-4601; 1661-7827; 1660-4601
• DOI: 10.3390/ijerph17103659

Previous studies have demonstrated that there is a high possibility that the presence of certain built environment characteristics can influence health outcomes, especially those related to obesity and physical activity. We examined the associations between select neighborhood built environment indicators (crosswalks, non-single family home buildings, single-lane roads, and visible wires), and health outcomes, including obesity, diabetes, cardiovascular disease, and premature mortality, at the state level. We utilized 31,247,167 images collected from Google Street View to create indicators for neighborhood built environment characteristics using deep learning techniques. Adjusted linear regression models were used to estimate the associations between aggregated built environment indicators and state-level health outcomes. Our results indicated that the presence of a crosswalk was associated with reductions in obesity and premature mortality. Visible wires were associated with increased obesity, decreased physical activity, and increases in premature mortality, diabetes mortality, and cardiovascular mortality (however, these results were not significant). Non-single family homes were associated with decreased diabetes and premature mortality, as well as increased physical activity and park and recreational access. Single-lane roads were associated with increased obesity and decreased park access. The findings of our study demonstrated that built environment features may be associated with a variety of adverse health outcomes.