Research Papers

Characteristics of left- and right-turning vehicle-pedestrian crashes and what can be done about them

Filename 5A_Arason_FP.pdf
Filesize 687 KB
Version 1
Date added June 18, 2019
Downloaded 4 times/fois
Category 2019 CARSP XXIX Calgary
Tags Research and Evaluation, Session 5A
Author/Auteur Arason, Erdelyi, Chan, Phillips, Brubacher
Stream/Volet Research and Evaluation

Slidedeck Presentation:



Background/Context: Intersection crashes involving turning vehicles hitting pedestrians is an issue that is not fully understood or given sufficient attention.

Aims/Objectives: We sought to compare the frequency of left- and right-turning single vehicle-pedestrian crashes versus other single vehicle-pedestrian crash configurations. We also sought to identify crash countermeasures that can respond to these crash situations.

Methods/Targets: We used police crash reports in British Columbia between January 2004 and December 2015 to determine the count and proportion of crashes, injuries, and fatalities for single-vehicle versus pedestrian crashes. We explored several factors that contribute to pedestrian injury or fatality, including (1) driver pre-collision action (left or right turn), (2) accident location (at/between signalized or non-signalized intersection) (3) road type (number of lanes, divided/undivided, speed limit), and (4) driver impairment or distraction. We calculated odds ratios through univariate logistic regression. We also undertook literature and inter-jurisdictional reviews to identify corresponding crash countermeasures.

Results/Activities: Over the total study period there were 3,878 crashes with a serious pedestrian injury or fatality. Of these crashes, road conditions were typically dry (62.8%) and clear (49.5%), most occurred in a 50 km/h speed zone (68.4%), many occurred during daylight hours (45.1%), and just over half occurred at an intersection (52.3%).

Of single-vehicle serious injury and fatal pedestrian crashes at intersections, the three most common crash configurations involved a left-turning vehicle (36.0%), a vehicle going straight (48.2%), and a right-turning vehicle (9.4%). The largest portion of these crashes occurred at a signalized control when the driver had a green light (37.4%), followed by situations where there was no traffic control (26.8%). Of all factors examined, the three highest unadjusted odds ratios for fatality or serious injury were driver impairment (OR= 2.37, 95% CI: 1.92, 2.91), the presence of three or more traffic lanes compared to one lane (OR=1.95, 95% CI: 1.51, 2.55), and nighttime (9 PM - 6 AM) compared to daytime (12 PM - 6 PM) conditions (OR= 1.71, 95% CI: 1.49, 1.97).

Discussion/Deliverables: Serious injury and fatality left-turn vehicle-pedestrian crashes are almost four times more likely to occur than right-turn ones. Most pedestrians when stuck were proceeding in a legal manner underscoring the need for road design solutions.

Indeed, many low-cost solutions to this problem exist at intersections including: changing signalization from permissive to protected-only phasing, banning the right-turn-on-red, leading pedestrian intervals, pedestrian scrambles, tighter turning radius, better lighting, innovative protected intersection designs, and many other measures.

Our study relied on police-reports. However, police are not required to attend all motor vehicle crashes, including pedestrian-involved ones, and as a result there is under-reporting. Other research using non-police data (e.g., hospital or ambulance data) would be beneficial to further understand the problem.

Conclusions: Serious injury and fatal left-turn vehicle-pedestrian crashes happen at a rate almost four times that of right-turn vehicle-pedestrian crashes. Each of these turning configurations, however, comprise a problem and each, paradoxically, has many low-cost solutions available that have often been unimplemented or under-implemented."