Survivability Factors for Cyclists Hit by Motor Vehicles: Evidence from the Canadian National Collision Database

Author(s): Demers

Slidedeck Presentation:

7B - Demers


Sixteen years of police-reported data from Transport Canada's National Collision Database (NCDB) was analyzed with a view to identify and quantify various factors that can impact the survivability of cyclists involved in a motor vehicle collision. Our goal is to fill a possible gap in the knowledge base on cyclist accidents and cyclist safety by studying cyclist fatalities. The large data sample allows us to measure and analyze on a relative basis the mortality risk for individual cyclists with an unprecedented degree of precision and confidence, while controlling for a range of important person-level and incident-level characteristics. Focusing on cyclist fatalities, we model the mortality risk for cyclists using probabilistic regression models. The NCDB data presents unique challenges, including a variable selection problem (on account that there is a large number of covariates available in the dataset) and a missing data problem (due to data quality issues in terms of missing or inconsistently coded observations). We outline empirical strategies to deal with these hurdles. The analysis tends to show that collision survivability depends largely on the cyclist's age and helmet usage. Survivability increases with age up to age 21, peaks for cyclists aged 21 to 34 and falls after age 35. Controlling for age and other factors, there is evidence that a bicycle helmet reduces the risk that a cyclist fatality will occur by approximately 35%. Survivability in general and the life-saving benefits of bicycle helmets in particular do not depend on the sex of the cyclist. Predictably, head-on and rear-end collisions tend to be more deadly. Certain environmental and situational variables like strong winds and traffic control devices also appear to impact survivability, independently of the type of collision or the characteristics of the cyclist. There might be opportunities to sensitize cyclists in various age groups about the risks they are exposed to while cycling using tailored education campaigns. Enforcement may be required to reinforce the importance of wearing a bicycle helmet. Engineering solutions that prevent or better protect cyclists involved in head-on and rear-end collisions would be most valuable. Unfortunately, because police-reported data only covers cases that became known to police, we do not expect the NCDB sampling to be random in a statistical sense. Moral hazard, risk compensation and behavioral adaptation mechanisms might also lead some cyclists to engage in more risky behaviors. We tury to address these methodological concerns. Cycling is a popular mode of transportation, both as a leisurely activity and a commuting option. In the context of the Vision Zero approach, reducing bicycle crash-related deaths should be a key component of any comprehensive injury prevention, traffic engineering, road safety initiative or enforcement initiative. Our analysis was based on publicly available data. However, the approach we propose could translate directly onto the more comprehensive 'confidential' dataset held by Transport Canada. Future research efforts could rely on this more detailed data.