Why Every Bicycle Counts and What We Can Learn from Fatal Crashes

Earlier this year, the League of American Bicyclists quietly launched a new website called Every Bicyclist Counts. The site tracks bicyclist fatalities, and, it does this for several reasons.

First, it serves as a memorial for fallen cyclists. Every bicyclist matters and we want to honor the memory of those who have lost their lives bicycling, while also providing a place for grieving friends and family to pay tribute in the comments, if they wish.

Second, we recognize that we all have a lot to learn about the circumstances of fatal crashes. To improve safety on our roads, we need to better understand what leads to fatal crashes and other bicyclist deaths (they’re not all crashes).

Third, we want to try to improve the response to, coverage of, and follow-up to every one of these fatal crashes. We believe that if we can hold a spotlight to the police, justice system, and media response to these incidents, we may be able to improve the quality of such responses over time.

This month, we sent out an appeal to League members describing Every Bicyclist Counts and asking for support for this important memorial, data-gathering, and awareness-raising tool. We immediately heard from members who were appreciative of the project – and eager to find out more about what we’ve learned so far.

In recognition of the interest in new data on bicycling fatalities, we decided to share some of what we know up to this point. However, please treat these data as highly preliminary. Imagine a big “Draft – Data not final” watermark on the page behind all that follows.

Why Every Bicyclist Counts

Every death of a cyclist is a personal tragedy for the family and for the entire community of bicyclists. We want to honor that with this project. We also see a gaping hole in the data currently collected about bicyclist fatalities. Only the most basic information on fatal crashes (time of day, intersection/non-intersection, gender, age, etc.) is gathered by the federal Fatality Analysis Reporting System (FARS) and there is a long wait before the data is publicly available. We gather that information here too, using the same categories when possible for consistency. But by delving more deeply and gathering specific details on bicyclist fatalities, we are able to add more richness to the data. We also report on fatalities that take place off of roads and those that do not involve motor vehicles.

Methodology

We gather information on fatalities as we learn about them through web-alerts and notifications from the public (primarily cyclists and family members; email Elizabeth[at]bikeleague.org). It appears that we are collecting a majority of – but not all – fatalities. We hope that as word of the project spreads, we will capture more and more of the 600+ annual fatalities and fill in more details about the circumstances. We rely first on public documents – newspaper reports and obituaries, blogs, police reports – and supplement that with first hand reports. We plan to compile and analyze the data annually.

To date, we have recorded 148 fatalities. We gathered information on four early crashes to work out the kinks before starting in earnest in February 2011. The remaining 144 fatalities reported below took place between February 2011 and May 16, 2012. We do not catch every fatality (we will back-fill as we find out about more), but our data appears consistent with FARS records from past years. Our high crash times of day line up with the times reported in FARS. The proportion of male and female fatalities is also consistent (FARS 2010 reports 87% male, our data set reports 88% male).

Limitations: Despite the enormous value and new analysis we believe this project will bring to the bicycling community, we recognize that there are (at least) several limitations.

• No coverage of injuries and near-misses. As it is, we have taken on the significant task of trying to track down what amounts to nearly two crashes per day. It would be nearly impossible to do the same for the tens of thousands of non-fatal crashes.
• No exposure data. Without knowing how many people are riding under difference conditions, it’s impossible to know the relative risk of different circumstances. This problem haunts other bicycle risk analysis as well.
• Not scientific. It is not a census of all fatal crashes (though we are trying to make it as comprehensive as possible), nor is it a random sample. We report only the ones we find out about. In addition, it requires a person enter the information and make determinations about which categories best apply. (Links to all sources are available on EveryBicyclistCounts.org.)
• Dependence on public sources. Every Bicyclist Counts still depends, to some extent, on police reporting and media accounts. These are often flawed, incomplete, or biased. We believe that additional information from cyclists and families can help improve our data. The project may, in the long run, help improve the quality of future reporting.

Nonetheless, we felt we had to start somewhere. We hope that through this project we will add to our understanding of the causes and circumstances fatal crashes and, ideally, improve institutional reporting practices.

Preliminary findings

Please consider these findings as the first step to learning more about fatal crashes. We have much more to learn through this project. The proportions will be slightly different in our Every Bicyclist Counts annual reports, since they will be derived from different data sets. As described above, this preliminary dataset is cumulative, starting primarily late-2011.

Collision type

We were able to determine the nature of the crash (eg. right hook, rear end) in 125 of the 148 fatalities.  We were surprised to learn that more than a third (35%) of cyclists killed were hit from behind. This is a considerably higher share than we had previously thought. Additionally, one cyclist was killed after rear-ending an automobile.

The next largest category is the generic “failure to yield” – 18% by drivers, 10% by bicyclists. This category was borrowed from FARS. Going forward, we are going to further break down this category to understand the nature of the failure to yield. This is an area of rich potential to add to our understanding of fatal crashes.

There were an equal number of fatal head-on collisions as right hooks, 11 fatalities/9 percent in each case. This will be one to watch as our dataset expands. Sideswipes and T-hits accounted for 6 percent each.

There were other deaths that would not be captured in the federal FARS database of traffic fatalities. For example, one cyclist died of dehydration on a trail, another died after striking a bollard on a trail. Sad occurrences like these reinforce the need for proper preparation and caution on all rides.

Collision location

The FARS database tells us only if bicycling fatalities occur at an intersection (~40% in 2010) or not at an intersection (~60% in 2010). We wanted to know more.

Half of the recorded fatalities occurred on the road at a non-intersection location. An additional 16 percent occurred on a roadway shoulder. Nearly 20 percent took place at an intersection; one more was recorded as occurring at a driveway; two happened in crosswalks. (Not included here are two people who were killed walking their bikes.)

Twelve of the fatalities in the dataset occurred in bike lanes (or intersections where the cyclist had been in a bike lane). Fatal crashes also occurred on trails and sidewalks, and took place at a railroad crossing.

Other trends

Among the 81 fatal crashes where an additional factor was reported for the driver, 37 (46%) were operating in a careless or inattentive manner, 20 (25%) committed hit-and-runs, 11 (14%) involved alcohol and/or drugs. Other factors were “sun in eyes” (4), distracted (3), drowsy (2), and red-light running (1).

Among the 29 fatalities with cyclist-related factors, 8 (35%) involved wrong-way riding, with one more reported as wrong-way sidewalk riding, and one more same-direction sidewalk riding. The remaining factors were one-offs in the current data set, including an instance of alcohol/drugs, a fall, a lack of lights or reflective gear, weaving in the road, holding the back of a pick-up truck, striking a bollard, and a fatal dehydration.

Several factors correspond with the highest rates of cycling:

• The highest concentration of fatal crashes (43%) occurred between 3:00pm and 9:00pm
• The majority (88%) of those killed were male
• Large proportions occurred in California (24%) and Florida (18%) – note that winter months are over represented in this data set which began in earnest in February 2011
• The average age was 42; the median age was 45

The vast majority of the sources (104 of 148) did not report on the presence of a helmet.

As our data set expands, we some of these data will likely smooth out.

Take-aways

Discussions of data can seem cold. This project is driven by a desire to understand and prevent future bicyclist fatalities through education and street design. Behind every number there is a life. Just as every bicyclist counts, every death needs to be counted. If this resonates with you, we invite you to support this project with a donation.

Keep riding and be safe out there.