Communities Severed by Roads Experience a Larger Number of Collisions

March 5, 2024

New York City neighborhoods with disrupted community connections due to traffic, roads, and transport infrastructure are experiencing an uptick in traffic collisions. This increase is seen both in total collisions, as well as those in which pedestrians or cyclists are injured or killed, according to a new study by researchers at Columbia University Mailman School of Public Health. The findings appear in the journal Environment International.

Roadway fatalities declined from 1980 to 2010, followed by a plateau in the 2010s, but then increased again in 2020 and 2021.

“Despite recent remarkable progress, road safety remains an urgent urban issue in New York and other U.S. cities. It is important to understand how the spatial configuration of the city enhances these road safety issues,” said Marianthi-Anna Kioumourtzoglou, ScD, associate professor of Environmental Health Sciences at Columbia Mailman School and senior author. “We wanted to investigate how the presence of road traffic and roads influence our health, beyond more commonly studied exposures such as air pollution and noise. Hopefully, our findings can help in the design of a less deadly and healthier city.”

Kioumourtzoglou and colleagues developed the first-ever index of community severance (Community Severance Index) to characterize the role of roads, traffic, and lack of pedestrian infrastructure in the disconnection of communities. They then examined the impact of community severance on road collisions.

The researchers used road traffic collisions in 2019 to estimate the association between community severance and traffic collisions, combining data on road infrastructure, road traffic activity, and pedestrian infrastructure. NYC-Open Data provided detailed information on the date and location of road traffic collisions. Spatial data to generate the index were obtained through the EPA and other providers, such as OpenStreetmap and the U.S. Department of Transportation.

They used an artificial intelligence algorithm to uncover patterns in the data, identifying many severed communities in the city with a high density of roads and traffic and a paucity of pedestrian space. They accounted for sociodemographic data, which could potentially influence the association between community severance and road traffic collisions. 

The researchers focused on road traffic collisions, as an approximation for road safety, because they can affect urban health—directly through injuries and death but also indirectly—for example, by influencing transportation choices. “Perception of lower road safety can lead people to choose to drive a car instead of cycling or walking in their day-by-day trips,” said Jaime Benavides, an associate research scientist in the Department of Environmental Health Sciences, who led the study.

The researchers were motivated to investigate how the predominant presence of roads and traffic shapes our health and behaviors and acts as a barrier, to the detriment of urban spaces for pedestrians. As one example, people who live on streets with higher traffic volume may interact less with their neighbors relative to those living on less congested streets.

Historically in the U.S., financial and political support has prioritized highways over public transit. Many urban highways were built directly through the neighborhoods of disadvantaged Black communities, severing their connections internally and/or segregating them from other communities.

“Our results highlight the necessity to further understand how the presence of road infrastructure and traffic influences human health in cities beyond most commonly studied traffic-related exposures such as air pollution and noise,” said Kioumourtzoglou. “The comprehensive indicator we developed for the presence of roads and traffic, the Community Severance Index, can be useful to identify areas that require targeted interventions, such as pedestrianization programs in New York City and other U.S. cities,” added Benavides.

Co-authors are Sabah Usmani and Vijay Kumar, Columbia Mailman School of Public Health.

The work was supported by the National Institute of Environmental Health Sciences (ES009089, ES028805, ES030616). The authors declare no known competing financial interests.

Media Contact

Stephanie Berger,