Urban Wire To Prevent Pedestrian Fatalities, We Need Better Data
Peace Gwam, Eleanor Noble
Display Date

Media Name: gettyimages-1228815047_crop.jpg

Pedestrian fatalities have increased since the onset of the COVID-19 pandemic. Even as Americans drove 13 percent fewer miles in 2020 compared with 2019, speeding, risky driving, and larger cars resulted in a 21 percent increase in pedestrian deaths. These 6,721 pedestrians were disproportionally people of color, children or elderly Americans, and people with disabilities.

Legislators are now turning to data-driven interventions to address this crisis. At the federal level, the Infrastructure Investment and Jobs Act requires states to assess data on pedestrian fatalities and serious injuries and identify a program to mitigate risks. Nationwide, local jurisdictions are participating in Vision Zero, a network of communities using data-driven interventions with the goal of eliminating all traffic fatalities and severe injuries.

These interventions can lead to funding for proven methods to protect pedestrians. Unfortunately, data on pedestrian safety are often inconsistent, with differing methodologies that lead to gaps in data. Without accurate data, interventions to protect pedestrians may be ineffective.

What little do we know about pedestrian fatalities in the United States?

The United States counts pedestrian fatalities three ways. Each approach has a different methodology and gaps in its data collection process, which can lead to inaccurate estimates.

  • The federal Fatality Analysis Reporting System (FARS) database reports crashes in all states, Puerto Rico, and Washington, DC. It only includes crashes in public trafficways and can exclude high-collision areas like parking lots and driveways. FARS data can exclude children who are victims of backovers, for example.
  • State-level data collection uses police data to count pedestrian injuries and fatalities. Community members may be less likely to report a pedestrian injury or fatality if they fear or mistrust the police. A study in Washington, DC, found communities with the largest Black populations had the highest shares of unreported crashes.
  • Finally, hospitals collect data on pedestrian injuries and fatalities in their local network. These data are often not public, and their collection depends on individual hospitals’ standards and capacity. Hospital data are rarely used for public interventions, with the exception of San Francisco (PDF), which is working to link hospital data with police-recorded collision data.

These three data-collection systems often report conflicting numbers.

Take Washington, DC, for example. In 2017, FARS and Vision Zero police-recorded collision data reported 11 pedestrian fatalities. That year, DC’s four trauma centers (PDF) reported 18 pedestrian fatalities. In 2019, FARS reported 9 pedestrian fatalities in DC, and Vision Zero reported 12 pedestrian fatalities. Hospital data were not available in 2019.

How can we improve data on pedestrian safety?

Data-driven interventions are only as good as the quality of the data used to drive them. To begin to address the pedestrian fatality crisis, policymakers can consider the following steps to improve data collection and quality:

  1. Ensure data can be disaggregated to understand inequities: Questions about geographic, racial, or gendered disparities can only be answered if the data are accurate at the individual level. If disaggregated data are unavailable, be intentional and straightforward about the limitations of using proxy variables to understand disparities. Existing limited data show that pedestrian fatalities disproportionately affect communities of color, children, elderly people, and people with disabilities. To ensure these disparities don’t widen, better data will be key.
  2. Look beyond standard pedestrian fatality data sources: By linking unconventional data sources (like hospital reports, 911 calls, and fire or emergency medical services departments), transportation agencies can fill important data gaps and capture cases in which police were not involved.
  3. Consider alternative sources for public data collection: Surveys or crowdsourced pedestrian safety data can provide insight on where pedestrian injury or fatality undercounts may occur, as well as areas pedestrians perceive as unsafe and may avoid altogether.
  4. Validate your findings through sustained community engagement and easily accessible public data: Involving community members and advocates early in decisionmaking about data sources, methods, and key takeaways can help prevent discrepancies down the road.

The increase in pedestrian fatalities—and the disproportionate impacts on communities of color and people with disabilities—is alarming. Better quality data are a crucial first step in addressing this crisis.


The Urban Institute has the evidence to show what it will take to create a society where everyone has a fair shot at achieving their vision of success.

Show your support for research and data that ignite change. 

Research Areas Neighborhoods, cities, and metros
Tags Transportation
Policy Centers Metropolitan Housing and Communities Policy Center
Related content
Media Name: gettyimages-1057929832_crop.jpg
Policymakers can make roads safer for pedestrians by incorporating new and different data on vehicle size and pedestrian risk.
Media Name: gettyimages-1158658762_crop.jpg
By learning from the steps taken by other countries, the Biden administration’s Safe Streets for All program can use its funding to effectively work toward the promise in its name.