Beyond the Crash Report: Finding Additional Data on Pedestrian Crashes


This is a guest blog post by Ellen Zavisca, Principal Transportation Planner for Knoxville-Knox County Planning. Ellen has been a transportation planner with Planning and with the Knoxville Regional Transportation Planning Organization (TPO) since 2005.

When I speak with people about my work collecting and analyzing crash data, I often say “This is the tip of the iceberg. We just don’t know how big the iceberg is.” So, how to we figure out how much ice is actually looming below the surface?

If you’re a person who walks a lot, you’ve probably experienced a scary near-miss incident where you were almost hit by a car. If you had been hit, you could call the police and file a report. But what do you do about a near-miss? It’s valuable information to those of us trying to make our streets safer. I set out to learn more about near misses (and crashes) to better understand the data we already had about crashes reported to police.

The City of Knoxville completed a road diet on Cumberland Avenue in 2017, taking the road from four lanes to two lanes with a median and occasional turning pockets, plus wider sidewalks. Reported pedestrian- and bicycle-related traffic crashes are down 90% since the road re-opened to traffic.

I worked closely with an advocacy group – Bike Walk Knoxville – to draft a survey that would be shared via social media, email lists, and newsletters. We targeted bike groups and shops, and tried to reach walkers through our local transit system. We incentivized responses by entering everyone into a drawing for prizes, including a bike shop gift certificate and a one-month bus pass. I wanted to learn about where unreported incidents (including unreported crashes and near-misses) are happening, and also about the scale of those incidents, in comparison with crashes that are reported to police.

We received around 200 responses. Here’s a bit about what we learned:

  • For every crash reported to police, there was an additional crash not reported. All of the unreported crashes resulted in no injury, or just a minor injury, and respondents told us they didn’t want to take the time to file a police report.
  • For every crash reported to police, there were 30 near-miss incidents.
  • We heard from a lot more bicyclists than walkers. That’s probably because it’s easier to target bicyclists (through shops and clubs) than it is to target walkers. But we know from our crash database that walkers are hit more often – about 75 percent of incidents in our database of reported crashes involve people walking. And 95 percent of pedestrian/bicycle fatalities in our region are people killed while walking. So we need to find better ways to reach out to the walking public.
  • We heard about hot spots of near misses, and were able to share that information with local engineering staffs and law enforcement officials so that they can try to remedy those situations before they result in serious crashes.

An additional way I’m trying to learn about the scale of the iceberg is through hospital discharge data. What I’m trying to capture there are any incidents where someone is hit by a car while walking or bicycling, doesn’t file a police report, but later realizes they’re injured and goes to the hospital for treatment.

Initially, I sought to get hospital discharge data from our state health department to compare individual records with crash reports made to police. But privacy regulations make it impossible for them to share data with us in a way that could result in an individual being identifiable. Still, the state was intrigued by what we were trying to do, and offered to help. So I’m sharing part of our crash database with them, and they’re going to try to match crash reports with hospital discharges to see if there are any injuries from pedestrian/motor vehicle or bicycle/motor vehicle crashes that aren’t being reported to police. This will give us another glimpse at the full iceberg.

All of these efforts to supplement crash report data are intended to help make the case to elected officials and other influential community members that hazardous conditions for people walking and bicycling in our region are widespread and need to be addressed.

To download a copy of the survey used in this work, click here.