How similar are two-unit bicycle and motorcycle crashes?
Introduction
Bicyclists, motorcyclists and pedestrians are often referred to as vulnerable road users in the road safety literature because the likelihood that they will be seriously injured if a collision occurs is higher than for motor vehicle occupants. Vulnerable road users comprise the majority of traffic fatalities in most low and middle-income countries (Naci et al., 2009) and in 2010 accounted for about a quarter to a half of traffic fatalities in high income countries such as the United States (28.6%) (NHTSA, 2012), Australia (32.0%) (DIT, 2012b) and Great Britain (49.7%) (DFT, 2012). In these three countries, motorcyclists and pedestrians make up the bulk of vulnerable road user fatalities, with pedal cyclists comprising only 3–6% of fatalities. However, pedal cyclists represent between a quarter (US) (NHTSA, 2012) and a third (Australia) (Henley and Harrison, 2012) of all vulnerable road users with non-fatal traffic injuries.
There are many similarities among the three groups of vulnerable road users as well as real differences. All three modes serve as both recreation and transport, have poor data and similar contributing factors to injury. In addition, most adult pedestrians, pedal cyclists and motorcyclists are also car drivers for whom walking, cycling or motorcycling is not their main mode of transport.
In common with pedestrians, injuries to bicycle and motorcycle riders result in higher injury costs (Hitchens and Palmer, 2012) than injuries to car occupants. The vulnerability of bicycle riders is particularly evident in crashes with motor vehicles. For bicyclists, only 6–8% of Emergency Department presentations result from collisions with vehicles (Scott et al., 2005) compared with at least 22% of hospital admissions for on-road crashes (Henley and Harrison, 2012) and more than 80% of on-road fatalities (DIT, 2012a). The higher travel speeds of motorcycles mean that they are vulnerable in single-vehicle crashes as well as in collisions with motor vehicles. Thus, while 80% of Australian on-road bicycle fatalities involved motor vehicles, only 58% of on-road motorcycle rider fatalities involved other motor vehicles (DIT, 2012a). In a German study, Otte et al. (2012) compared injury outcomes in multi-vehicle crashes involving pedestrians, pedal cyclists and motorized two-wheelers. Overall, pedestrians were the most severely injured, followed by motorized two-wheelers then bicyclists. The lower injury severity of bicyclists was related to the lower speed of the collision partner, compared to pedestrian crashes. They comment that the higher speeds of the motorcyclists contributed to the severity of their injuries.
Bicycle and motorcycle crashes have generally been analyzed in isolation, but it is expected that similar factors may be important for both types of crashes because both are minority road users in comparison with dual track vehicles, physically smaller, less visible, lack physical protection, are less stable, and more affected by road surface irregularities. The limited conspicuity of bicycles and motorcycles and consequently drivers failing to see them or give-way to them have been identified as contributing factors in many studies (e.g., Haque et al., 2012, Horswill et al., 2005, Pai, 2011b, Pai et al., 2009), which implies that there might be similarities between bicycle and motorcycle crashes, particularly at give-way situations. Other authors have suggested that drivers’ perception of motorcycles as less threatening than other large vehicles may also contribute to them failing to give way (see Pai, 2011a). Since bicycles are physically smaller and have limited conspicuity, these factors could be true for bicycles as well.
Poor availability of data is a problem for understanding both bicycle and motorcycle crashes. The numerators of crash risk (numbers of persons in crashes or injured) are substantially under-reported for bicyclists and motorcyclists. US and European studies indicate that only 11% (Stutts et al., 1990) to 13% (Veisten et al., 2007) of bicycle crashes are recorded in police statistics and the data are skewed to serious injury crashes and those that involve motor vehicles (Stutts et al., 1990). US, European and Australian comparisons show about twice as many hospitalized motorcyclists in health data as in police data (Henley and Harrison, 2012, NCIPC, 2012, NHTSA, 2012). The extent of under-reporting is greater in less serious bicycle crashes in many countries (see ITF, 2012). The denominators used in risk calculations often relate to per head of population, per license or registration or per distance traveled. Distance traveled appears to be conceptually a better denominator, but the availability of this data for motorcycles is patchy at best and its accuracy is sometimes disputed (see Haworth, 2003). Data regarding the distances traveled by bicyclists and pedestrians, and the extent to which this travel occurs on roadways, are very sparse (Aultman-Hall et al., 2012).
The aim of this paper is to explore the similarities and differences between bicycle and motorcycle crashes with other motor vehicles in order to assess the extent to which similar treatments may be effective for both bicycle and motorcycle crashes. If so, greater benefits in terms crash costs saved may be possible for the same investment in treatments. The comparisons focus on two-unit crashes of higher severity levels, given that single-vehicle crashes are less likely to be similar because of the much higher travels speeds of motorcycles than bicycles, and crashes of lower severity levels are prone to greater under-reporting, as discussed earlier. This paper proceeds to outline the four-stage methodological approach involving crash data filtering, descriptive analyses of crash and controller characteristics, modeling probabilities of bicycle and motorcycle crashes, and modeling at-fault characteristics of riders and drivers using regression models. Results from the descriptive analyses and regression models are then presented, followed by a discussion of the findings and their implications in developing targeted countermeasures. Limitations and conclusions of the research are finally presented.
Section snippets
Setting
This research was conducted in the State of Queensland, Australia. Queensland has 4.3 million inhabitants and a climate that varies from sub-tropical to tropical, allowing year-round bicycle and motorcycle riding. A recent national population survey estimated that about 26% of the Queensland population rode a bicycle in the previous month (ABC, 2012). However, in the 2011 Australian Census, only 1.2% of Brisbane residents traveled to work by bicycle (ABS, 2012). There were 162,000 motorcycles
Descriptive statistics
The characteristics of the 6761 two-wheeler crashes are summarized in Table 1, Table 2. Riders were more likely to be male (79.1% bicycle, 89.3% motorcycle) than the drivers in these crashes (57.6% drivers in bicycle crashes, 59.4% drivers in motorcycle crashes). Almost a quarter (22.6%) of the bicycle riders were children. About 10–13% of motorcycle riders and drivers were novices (learner, provisional or restricted licenses). Almost 8% of motorcycle riders were unlicensed.
Drivers of other
Discussion
This paper explored the degree of similarity of two-unit bicycle and motorcycle crashes to assess whether the same treatments might be effective for both types of crashes. The analyses revealed general similarities in the road characteristics and crash types, with intersection from adjacent approaches being the most common crash type for both bicycles and motorcycles. Consistent with earlier research (ACEM, 2008, Comelli et al., 2008, Haque et al., 2009, Hurt et al., 1981, Johnston et al., 2008
Limitations
The dataset did not include information on some factors which have previously been shown to be significant predictors of at fault status in motorcycle crashes, including seat belt use by the other driver (Schneider et al., 2012) and estimated vehicle speed (Kim et al., 2007). The lack of data for some of the high-risk behaviors may have prevented identifying relationships between them (as in Schneider et al., 2012), which may have resulted in effects being attributed to the wrong variable.
While
Conclusions
The crash types, patterns of fault and factors affecting fault are generally similar for two-unit bicycle and motorcycle crashes. This confirms the need to combat the factors contributing to failure of other drivers to yield right of way to two-wheelers, and suggests that some of these actions should prove beneficial to the safety of both motorized and non-motorized two-wheelers. The limited conspicuity of two-wheelers might have resulted in drivers failing to see and yield right of way to
References (57)
- et al.
Toronto bicycle commuter safety rates
Accident Analysis and Prevention
(1999) - et al.
Towards a conceptual model of motorcyclists’ risk awareness: a comparative study of riding experience effect on hazard detection and situational criticality assessment
Accident Analysis and Prevention
(2012) - et al.
Critical factors in fatal collisions of adult cyclists with automobiles
Accident Analysis and Prevention
(2010) - et al.
Causation mechanisms in car-to-vulnerable road user crashes: implications for active safety systems
Accident Analysis and Prevention
(2012) - et al.
An investigation on multi-vehicle motorcycle crashes using log-linear models
Safety Science
(2012) - et al.
Modeling fault among motorcyclists involved in crashes
Accident Analysis and Prevention
(2009) - et al.
Characteristics of, and insurance payments for, injuries to cyclists in Tasmania, 1990–2010
Accident Analysis and Prevention
(2012) - et al.
The visual search patterns and hazard responses of experienced and inexperienced motorcycle riders
Accident Analysis and Prevention
(2010) - et al.
Bicyclist injury severities in bicycle–motor vehicle accidents
Accident Analysis and Prevention
(2007) - et al.
Visibility aids for pedestrians and cyclists: a systematic review of randomised controlled trials
Accident Analysis and Prevention
(2004)
Safety effects of permanent running lights for bicycles: a controlled experiment
Accident Analysis and Prevention
Injury protection and accident causation parameters for vulnerable road users based on German in-depth accident study GIDAS
Accident Analysis and Prevention
Motorcycle right-of-way accidents—a literature review
Accident Analysis and Prevention
Overtaking, rear-end, and door crashes involving bicycles: an empirical investigation
Accident Analysis and Prevention
A mixed logit analysis of motorists’ right-of-way violation in motorcycle accidents at priority T-junctions
Accident Analysis and Prevention
Attention and expectation problems in bicycle–car collisions: an in-depth study
Accident Analysis and Prevention
Age and experience in motorcycling safety
Accident Analysis and Prevention
Examination of factors determining fault in two-vehicle motorcycle crashes
Accident Analysis and Prevention
Bicycle accidents and injuries: a pilot study comparing hospital- and police-reported data
Accident Analysis and Prevention
Total costs of bicycle injuries in Norway: correcting injury figures and indicating data needs
Accident Analysis and Prevention
Drivers’ and cyclists’ experiences of sharing the road: incidents, attitudes and perceptions of visibility
Accident Analysis and Prevention
Using reflective clothing to enhance the conspicuity of bicyclists at night
Accident Analysis and Prevention
The effectiveness of the ‘ride-bright’ legislation for motorcycles in Singapore
Accident Analysis and Prevention
MAIDS. In-depth Investigations of Accidents Involving Powered Two Wheelers – Final Report 2.0
Innovative Data Collection for Pedestrians, Bicycles, and Other Non-motor Vehicle Modes
Motorcycling and Leisure: Understanding the Recreational PTW Rider
Cited by (31)
Exploring temporal instability effects on bicyclist injury severities determinants for intersection and non-intersection-related crashes
2024, Accident Analysis and PreventionIdentifying the heterogeneous effects of road characteristics on Motorcycle-Involved crash severities
2023, Travel Behaviour and SocietyUsing random forest to test if two-wheeler experience affects driver behaviour when interacting with two-wheelers
2023, Transportation Research Part F: Traffic Psychology and BehaviourCitation Excerpt :It is also possible that the driver looks and detects the TW but fails to process the information effectively (e.g. speed, distance) to avoid an accident (Räsänen and Summala 1998). Other trends for two-unit crashes involving TWs and other motor vehicles were reported by Haworth and Debnath (2013). For instance, Opposing vehicles turning (18.9 % vs 10.2 %) and Rear-end crashes (16.6 % versus 3.9 %) were more likely for motorcycles, while Vehicle leaving the driveway (19.7 % versus 7.6 %) was more common among bicycle crashes.
Bicycle crash contributory factors: A systematic review
2022, Safety ScienceCitation Excerpt :Whilst there is a growing body of research applying systems thinking to understand road trauma generally (e.g. McIlroy et al., 2019; Read et al., 2021) it is not clear whether this has transferred into the realm of cyclist safety and bicycle collisions. For example, studies examining bicycle crash contributory factors studies tend to focus on factors relating to the road users, vehicles and road environment involved (Haworth & Debnath, 2013; Moller et al., 2021; Prati et al., 2018). A recent systematic review on the factors which contribute to bicycle-motor vehicle collisions found that a majority of the included studies identified factors relating to road user’s behaviour and infrastructure.
- 1
Tel.: +61 7 3138 8417; fax: +61 7 3138 0111.