Pedestrian Deaths in Poorer Neighborhoods Report

The number of pedestrians killed along the nation’s roadways has slowly climbed in recent years, even as total traffic fatalities declined.

A Governing analysis of more than 22,000 traffic accidents occurring between 2008 and 2012 finds that pedestrians are killed at disproportionately higher rates in the nation's poorer neighborhoods.

While some regions are safer than others, large cities continue to record a significant numbers of pedestrian fatalities each year. Several localities have responded by making urban landscapes more walkable, stepping up traffic enforcement and launching other initiatives aimed at improving pedestrian safety.

Download report PDF
Read magazine story
Interactive map of fatal accidents
Neighborhoods with highest fatality rates

Results Summary

Governing compiled data on locations of all fatal pedestrian accidents reported in the National Highway Traffic Safety Administration’s Fatality Analysis Reporting System from 2008-2012. Using each accident’s geographic coordinates, the total number of accidents was computed for Census tracts, which vary in size, but are similar to neighborhoods.

Within metro areas, low-income census tracts recorded pedestrian fatality rates approximately twice that of more affluent neighborhoods. Examining census tracts’ poverty rates yielded a similar pattern. Metro-area tracts below the national poverty rate of 15 percent registered 5.3 deaths per 100,000 residents over the five-year period. By comparison, poor neighborhoods where more than a quarter of the population lived in poverty had a rate of 12.1 deaths per 100,000 people.

The following tables show 2008-2012 pedestrian death rates for all census tracts within metro areas:


Census Tract Poverty Rate 2008-12 Deaths Per 100K
≤ 5% 3.8
>5-10% 5.5
>10-15% 7
>15-20% 8.3
>20-25% 9.9
>25-30% 11.2
>30% 12.6
Census Tract Per Capita Income 2008-12 Deaths Per 100K
High Income
Middle Income
Low Income
(Less than $21,559)

Local Area Data

In general, poorer neighborhoods were found to have higher numbers of pedestrians killed per capita than other areas within the same jurisdictions. For many localities, the disparity was particularly large. Low-income tracts had deaths rates more than double that of high-income tracts in Broward County, Florida, and Wayne County, Michigan, for example. This wasn’t true of all cities, though. Census tracts in most of New York’s boroughs and Philadelphia recorded roughly similar fatality rates across income and poverty levels.

While poorer tractas tend to have the highest pedestrian fatality rates, middle-income communities recorded death rates only slightly different from wealthier neighborhoods in most larger counties.

Of the 104 metro areas examined with at least a half million residents, all but four recorded higher per capita pedestrian death rates for poor census tracts (with poverty rates greater than 25 percent) than their metro area total.

The following map shows poorer neighborhoods with the highest per capita pedestrian death rates. Larger markers represent metro areas with higher pedestrian death rates for Census tracts where more than a quarter of the population lives in poverty. (Click to open full-screen interactive map in new window.)

Within all the nation’s metro areas, approximately 7.2 pedestrians per 100,000 residents died from 2008 through 2012. The following table lists all metro areas with populations exceeding a half million, ranked by five-year total per capita death rates:

* Indicates fewer than 10 deaths occurred in these tracts

Pedestrian fatality statistics are also available at the county level for all counties with at least a half million residents. For other smaller individual counties, per capita rates are greatly affected by fewer total fatalities and smaller low-income populations.

Open County Pedestrian Deaths Tool

Where Pedestrians Are Killed

The following tables summarize locations of fatal pedestrian accidents, as reported in NHTSA’s Fatality Analysis Reporting System. Totals cover all accidents nationwide from 2008 through 2012.

Relation to Junction Deaths % of Total Deaths
Non-Junction 15,232 69
Intersection-Related 3,007 14
Intersection 2,736 12
Driveway Access/Alley 446 2
Entrance or Exit Ramp-Related 252 1
Through Roadway 205 1
Other/Unknown 147 1
Route Type Deaths % of Total Deaths
Municipality Local Street 6,336 28.8
State Highway 5,293 24.0
U.S. Highway 3,444 15.6
County Road 2,609 11.8
Interstate 2,208 10.0
Township Local Street 1,193 5.4
Other/Unknown 782 3.6
Local Frontage Road 160 0.7

Roadway Function Class Deaths % of Total Deaths
Rural: Interstate 699 3.2
Rural: Local Road 1,386 6.3
Rural: Major Collector 1,075 4.9
Rural: Minor Arterial 907 4.1
Rural: Minor Collector 275 1.2
Rural: Other Principal Arterial 1,542 7.0
Urban: Collector 979 4.4
Urban: Interstate 1,603 7.3
Urban: Local Road 3,250 14.8
Urban: Minor Arterial 3,226 14.6
Urban: Other Freeways or Expressways 1,062 4.8
Urban: Other Principal Arterial 5,862 26.6
Unknown 159 0.7


Governing compiled data on all pedestrian fatalities in the National Highway Traffic Safety Administration’s Fatality Analysis Reporting System (FARS) occurring between 2008 and 2012. Records in the database included each accident's latitude and longitude coordinates, except for about 300 records (less than 2 percent of all fatalities). A spatial analysis was then performed using the FARS data and Census tract boundary shapefiles to determine the tract in which each accident occurred.

Economic and demographic estimates for tracts were obtained from the U.S. Census Bureau’s 2008-2012 American Community Survey. Census tracts were divided into three per-capita income categories, with an equal number of tracts for each tertile (High income: $31,356+; Middle income: $21,559-$31,355; Low income: <$21,559). A separate poverty category was determined using the percentage of a tract’s population below the poverty level. The Census Bureau does not publish poverty and income estimates for about 500 census tracts (less than 1 percent of all tracts), most of which have no residents. These tracts were excluded from the analysis.

Pedestrian death rates were computed for all Census tracts within metropolitan statistical areas, or about 60,000 total. Data summarizing accident locations covers all Census tracts.


More from GOVERNING Data