By Robyn D. Robertson & Heather Woods-Fry, Traffic Injury Research Foundation   |    October 2018    |    12 minute read

Innovations in vehicle safety have helped prevent and mitigate the consequences of risky driving behaviours in the past two decades. As a result, there is widespread anticipation that the transition from advanced driver assistance systems to semi-automated, or limited self-driving vehicles (LSDVs), has the potential to produce further and more rapid gains. Yet while these advances have contributed to substantial crash reductions, they have yet to produce the similarly anticipated reductions in crash costs.

Whether these declines in crash costs will materialize is a topic of debate for a few reasons. The first is that national polling data from the Traffic Injury Research Foundation (TIRF) suggests that those drivers most likely to be early adopters of LSDVs are also more prone to risk-taking behaviours. The second is that, although traditional road safety priorities such as speeding, non-use of seatbelts and alcohol-impaired driving have declined, two new issues have emerged as leading contributors to the crash problem. Distracted driving and marijuana-impaired driving have not previously been recognized as a major factor in crashes. Equally concerning is that these factors are unlikely to be mediated with the use of LSDVs.

New research on distracted driving has demonstrated the prevalence of this problem and underscored the pervasiveness of this behaviour. Neither initiatives to educate drivers and increase awareness about the risks, nor escalating penalties have significantly discouraged distraction among drivers. In fact, a published National Action Plan on distracted driving from TIRF reveals that distracted driving has in fact surpassed alcohol-impaired driving as the leading contributor to road fatalities in some jurisdictions. Moreover, the pending legalization of recreational marijuana has also prompted renewed concern about drug-impaired drivers. An analysis of crash data from the past decade shows that the presence of marijuana among fatally injured drivers has increased although new legislation will not come into effect until the last quarter of 2018.

These emerging issues warrant attention and concern as their potential implications on the road crash problem are substantial. A clear understanding of some of the latest research, data and policy implications related to semi-automated vehicles, distracted driving and drug-impaired drivers can help insurers better assess the risks and impending costs associated with each of these issues. This knowledge is essential to inform the development of policies and practices that can protect the health, safety and mobility of all Canadians.  

Semi-Automated Vehicles

There is much interest in automated vehicles (AVs) and their potential to dramatically reduce road crashes which most often occur due to driver error or condition. However, these benefits will only materialize if automated vehicles are implemented with appropriate educational strategies and safeguards that positively shape driver use of and reliance on them. Results from a 2016 national survey by TIRF with funding from the Toyota Canada Foundation showed that young male drivers who are more confident in the technology and drive more frequently are likely to be early adopters of it. However, this population of drivers is also generally less safety-conscious and more crash-involved. Conversely, the study also showed that aging Canadians, who are some of the safest and most experienced drivers, are less likely to rely on AVs.

Focus groups were undertaken by TIRF in 2018 to more closely examine the attitudes of older drivers and explore their reluctance to readily adopt the use of LSDVs also with funding from the Toyota Canada Foundation. Aging drivers can potentially achieve significant gains in health and mobility as a result of semi-automated vehicles, but only if they are willing to use them. Results of these focus groups reveal that their knowledge and beliefs about such vehicles influences the likelihood that they will rely on this technology to improve their safety (Robertson et al. 2018b). The most important findings emerging from these two studies is that there are opportunities to encourage aging drivers to become early adopters of LSDVs, to demonstrate the true safety potential of these vehicles, and improve outcomes associated with their widespread adoption. Other key findings include:

>Feelings of safety and knowledge are positively related to perceived ease of use and adoption of LSDVs.

>Older drivers are less likely to engage in risky driving behaviours with LSDVs or to become over-reliant on these technologies.

>Older drivers are interested in learning about LSDVs and very receptive to strategies and tools to help them use LSDVs in ways that maximize their safety benefits.

>Education for older drivers that focuses on the safety and performance of LSDVs delivered in a practical hands-on format can help our most vulnerable drivers reap the greatest benefits from this technology.

Notably, since older drivers are more sensitive to the inherent risks and limitations associated with LSDVs, they may be best-suited to test them in the real world. Ultimately, the use of LSDVs by a population of older drivers can be insightful regarding optimal strategies and conditions that are needed to safely integrate automated vehicles in the driving fleet for all drivers. This knowledge can help set standards regarding the level of education and skills that drivers of all ages must possess before using LSDVs.

Distracted Driving

Despite continued declines in fatalities due to road crashes in the past decade, deaths involving distracted driving have increased. A trend analysis published by TIRF reveals that distraction was a factor in 25% of fatal crashes in 2015 as compared to 16% in 2011 (TIRF September 2018). Data collection has significantly improved in the past seven years as more jurisdictions have added distraction-related variables to the crash form, and this has no doubt partially contributed to an increase in crashes. However, officers are also better able to identify distraction-related factors at crash events as more research about these behaviours has become available. In other words, the proportion of fatal crashes in which the cause is unknown has dramatically declined.    

Who is most likely to be injured or killed due to distracted driving?

In 2015, more than 350 people were killed in distracted driving collisions according to TIRF’s National Fatality Database (TIRF September 2018). Among those killed:

>The largest number of distraction-related fatalities involved distraction on the part of the surviving driver between 2000 and 2015. In other words, distracted drivers are more likely to kill other road users as opposed to being killed themselves. Fatally injured victims are either the non-surviving driver, passengers dying in a crash where at least one of the drivers was distracted, or pedestrians struck by a distracted driver.

>A larger percentage of fatally injured female drivers (18.6%) were distracted compared to 12.8% of fatally injured male drivers during this same time period.

>Drivers aged 35 to 44 years represented the largest percentage of fatally injured drivers who were distracted (17.3 %), followed closely by drivers aged 16 to 19 years (17.1 %) during the five-year period between 2011 and 2015.

Are there common characteristics of distraction-related crashes? 

Yes, the collision characteristics of distraction-related fatal crashes do not represent the typical pattern or characteristics found in other types of road crashes. From 2011 to 2015, crash data show that:

>The largest number of distraction-related fatalities occur during the late afternoon hours, from 3:00 pm to 5:59 pm, making afternoon rush hour a high-risk period for all road users.

>The largest percentage of distraction-related fatalities occur mid-week on Wednesday (27.1 %), with lower percentages on the weekend.

What initiatives exist to tackle distracted driving?

Distracted driving is a primary focus of road safety planning in jurisdictions across Canada. To tackle this issue, the coordination and exchange of information across groups of stakeholders and jurisdictions is essential. To facilitate partnerships, in 2015 TIRF, Drop It And Drive^® (DIAD) and The Co-operators formed the Canadian Coalition on Distracted Driving (CCDD) to encourage a coordinated approach. It represents a broad cross-section of 24 diverse stakeholders across the country. The CCDD has developed a National Action Plan to reduce distracted driving in Canada, and has launched a web-based information hub that contains the latest research on distracted driving, statistics, data, and laws and penalties in Canada. It has also produced a variety of educational tools and resources, such as the “Let’s Talk About Crashes” factsheet. As part of the Plan, the CCDD has also formed dedicated working groups with transport companies, insurers and health professionals to meet their individual needs related to this issue. Key objectives for insurers are to utilize consistent messaging, improve data collection, and assess the potential of new safety features. Activities that are underway from the CCDD in 2018 include the development of a business case to encourage all employers to adopt distracted driving policies in the workplace, and the development of a framework that describes the many types of new technologies that can potentially reduce distraction. This framework, which will be made available in the last quarter of 2018, highlights common questions that should be asked to assess the value of these technologies in the real world, and shares many examples of technologies that could help to address the problem if well-implemented.   

Marijuana-Impaired Driving

The consequences of the legalization of recreational marijuana on drug-impaired driving will likely be profound. Although laws prohibiting drug-impaired driving were strengthened in 2008 with the implementation of the Drug-Recognition Expert (DRE) program, the enforcement of these laws has been limited by an inadequate number of trained and certified police officers and limited acceptance of the program among courts.

With the passage of new impaired driving legislation in the form of Bill C-46, the current emphasis on behavioral impairment to detect drugs is expanded to include per se levels for THC* (the active ingredient in marijuana) at 2.5 nanograms/millilitres and 5 ng/ml, similar to the long-standing use of per se limits for alcohol at .05 and .08 mg/%. However, in sharp contrast to alcohol per se limits, the science to support specific per se limits for marijuana is hotly debated among leading experts and not well-established by science.

But what is known is that at least a portion of Canadians believe they are better and safer drivers after using marijuana and do not agree that using marijuana before driving is impairing. In fact, data from a new Road Safety Monitor public opinion poll on drugged driving in 2017 that will be published by TIRF in October 2018 shows that 17.6% of respondents believed that marijuana does not impair driving as much as alcohol, and younger drivers aged 16 to 24 years were more likely to believe this as compared to those aged 65 or older (Robertson et al. 2018a).

These data suggest that much more work is needed to help Canadians understand the potential risks associated with driving while impaired by marijuana. In addition, tracking of attitudes and practices, as well as more research into this topic is essential to increase knowledge and inform the development of effective countermeasures to address it.

What are the known effects of marijuana on driving?

Although low doses of marijuana produce mild to moderate impairment in cognitive and psychomotor abilities, larger doses produced significant impairment in cognitive, psychomotor and driving performance. The driving measures that are impaired include variability in speed maintenance, variability in lane maintenance (i.e., weaving), increased variability in headway, and increased reaction times (Hart et al. 2001; Sexton et al. 2000). Marijuana can affect a driver’s ability to concentrate on the driving task, to maintain appropriate headway with the vehicle ahead, and cause drivers to drift outside of their lane. Reaction time is also affected as marijuana can impair the driver’s ability to react to unexpected events (Hartman et al. 2016). Crash data shows that marijuana-impaired drivers have increased odds of being in a fatal crash (Li et al. 2013; Asbridge et al. 2012; Starkey et al. 2017) and have higher odds of being found responsible for a crash (Drummer et al. 2003; Drummer et al. 2004).

How prevalent is marijuana-impaired driving?

An increasing percentage of fatally injured drivers in Canada tested positive for marijuana in recent years according to data from TIRF’s National Fatality Database. While these data indicate presence as opposed to impairment which cannot be determined, results are concerning (TIRF October 2018).

>Among those drivers tested for drugs, 20.9% of fatally injured drivers were positive for marijuana in 2015. This percentage increased from 12.4% in 2010.

>A majority (82.9%) of fatally injured drivers were tested for drugs in Canada in 2014. Almost half (42.4%) of fatally injured drivers that were tested for drugs in 2014 were positive and almost half of those drivers were positive for marijuana (44.7 %).

>In 2014, there were more fatally injured drivers that tested positive for drugs than for alcohol (42.4 % versus 29.6%).

>From 2010 to 2014, the percentage of fatally injured drivers that tested positive for alcohol decreased from 37.6% to 28.4%.

Self-report data also demonstrates the increasing prevalence of drug-impaired driving in Canada. Data on drugged driving behaviours and attitudes from the TIRF Road Safety Monitor (RSM) series (Robertson et al. 2018a) show that:

>Almost three percent (2.9%) of Canadian drivers said that they had driven a motor vehicle within two hours of using marijuana or hashish at least one time during the previous 12 months in 2017.

>Approximately 2.6% of Canadian drivers reported driving within two hours of using marijuana in 2016, and 1.6% reported doing so in 2013.

Age. Crash data from 2014 show that fatally injured drivers aged 16 to 19 and 20 to 34 years old were the largest percentage of drivers that tested positive for marijuana (31.1% and 29.7% respectively) (TIRF December 2017). Self-report data from 2017 show that 49.7% of drivers who admitted to driving a motor vehicle within two hours of using marijuana were between the ages of 25 to 44 years old and 22.2% were between 16 to 24 years old (Robertson et al. 2018a).

Sex. Crash data show that 20.2% of fatally injured male drivers tested positive for marijuana versus 11.9% of fatally injured females in 2014 (TIRF December 2017). In addition, self-report data from 2017 show that 5.1% of males versus 0.8% of females admit to driving a motor vehicle within two hours of using marijuana (Robertson et al. 2018a).

Are there tests that can detect marijuana in drivers?

Yes, police officers can use a variety of tests to detect impairment in drivers due to marijuana.

Behavioural impairment tests. Police agencies are increasing the number of officers trained to complete the Standard Field Sobriety Tests (SFSTs), a scientifically-validated battery of tests shown to detect behavioral impairment. The SFST is a three-test battery that includes the horizontal gaze nystagmus (HGN), the walk-and-turn, and the one-leg stand tests. If results of the SFST provide reasonable grounds to suspect drug-impairment, a Drug Recognition Expert (DRE) is summoned to conduct a 12-step, standardized evaluation designed to assess physical, cognitive and medical indicators. DREs are trained to detect impairment in drivers due to drugs other than, or in addition to, alcohol. DRE certification requires scientific knowledge of the effects of different types of drugs on the body and officers are required to undergo extensive training to be qualified to conduct these evaluations. However, due to barriers associated with obtaining certification as well as cost of the program, the number of trained officers in Canada is low. Currently it is estimated that there are just 600 trained and certified DREs nationally.

Roadside saliva device. Some jurisdictions will also soon implement the use of roadside oral fluid tests to collect saliva from drivers suspected of marijuana-impaired driving. The Dräger DrugTest 5000® is the first device to be approved by the federal government. It has high validity and accuracy and has been extensively tested in many countries around the world, including Canada. However, the use of the device in extremely cold weather still presents an issue. Oral fluid screening devices are non-invasive, easy to use, and have been validated to detect the presence of marijuana in saliva. However, the widespread use of such devices will be affected by the high cost associated with the technology.

Conclusions

The most pressing road safety issues that will challenge us in the next decade relate to the safe implementation of semi-automated vehicles, distracted driving and marijuana-impaired driving. Traditional strategies used to address historical road safety priorities, such as the implementation of laws, policies, and enforcement combined with penalties will likely be insufficient on their own to change driver behaviours. This is because AVs and distraction impact a much larger segment of drivers than has been the case with other road safety issues. We are simply unable to bring enough resources to bear to police the roads everywhere at all time. This means a much larger and more heterogeneous population must be targeted, and some of them will be more receptive to some types of countermeasures than others.

In terms of drug-impaired driving, this will present an equally formidable challenge due to the widespread misperception that the use of marijuana by drivers is safe, or at least safer than other types of impairment. As such, convincing these drivers that this behaviour is in fact risky will be a more daunting task.

More positively, although current data suggest concern is warranted about these issues, there are also immediate opportunities to begin to address them. More agencies are seeking to coordinate, partner and share knowledge and initiatives in recognition of their magnitude and complexity. New partners are entering the road safety field, bringing with them expertise, different perspectives and new ideas that enable us to be more effective and efficient. At the same time, improvements in data collection and new research are revealing ways that we can influence behaviour and shift the tide as with the case of aging drivers and AVs.

Going forward, insurers can work cooperatively with their counterparts in the industry and in other sectors to improve data collection and increase the standardization and coordination of data indicators that are collected. This will provide national measures of the prevalence of these issues and ensure comparability between jurisdictions. In the absence of this approach, a national picture on these issues cannot be established and progress on these issues cannot be tracked over the long-term. In addition, insurers can strive to provide the public with consistent facts and messaging about these issues, and in turn re-shape social norms to emphasize the unacceptability of risks posed by these top road safety issues.

* The term “marijuana” refers to the cannabis plant that contains more than 100 cannabinoids. The primary psychoactive component of cannabis is delta-9-tetrahydrocannabinol, commonly known as THC. THC and its psychoactive metabolite, 11-hydroxy-THC or 11-OH-THC, and primary inactive metabolite, 11-nor-9-carboxy-THC or THC-COOH are frequently measured in biological fluids to document marijuana intake.

REFERENCES/FURTHER READING:

Asbridge, M., Hayden, J. A., & Cartwright, J. L. (2012). Acute cannabis consumption and motor vehicle collision risk: systematic review of observational studies and meta-analysis. British Medical Journal. 2012:344. doi: https://doi.org/10.1136/bmj.e536 BMJ 2012;344:e536

Canada. Parliament. House of Commons. “Impaired Driving Act” Bill C-46, 42nd Parliament, 1st Session, 2017. Ottawa: Public Works and Government Services Canada - In Committee (House), October 31st 2017.

Dräger DrugTest® 5000. Retrieved from: https://www.draeger.com/en-us_us/Alcohol-And-Drug-Detection/Products/Breath-Alcohol-and-Drug-Testing/Drug-Testing-Devices/DrugTest-5000

Drummer, O. H., Gerostamoulos, J., Batziris, H., Chu, M., Caplehorn, J. R., Robertson, M. D., & Swann, P. (2003). The incidence of drugs in drivers killed in Australian road traffic crashes. Forensic Science International, 134(2), 154-162.

Drummer, O. H., Gerostamoulos, J., Batziris, H., Chu, M., Caplehorn, J., Robertson, M. D., & Swann, P. (2004). The involvement of drugs in drivers of motor vehicles killed in Australian road traffic crashes. Accident Analysis & Prevention, 36(2), 239-248.

Hart C.L., van Gorp, W., Haney, M., Foltin, R.W., Fischman.  M.W. (2001). Effects of acute smoked marijuana on complex cognitive performance. US National Library of Medicine, National Institutes of Health. 2001 Nov;25(5):757-65. DOI:10.1016/S0893-133X(01)00273-1

Hartman, R. L., Richman, J. E., Hayes, C. E., & Huestis, M. A. (2016). Drug Recognition Expert (DRE) examination characteristics of cannabis impairment. Accident Analysis & Prevention, 92, 219-229.

International Association of Chiefs of Police. (2014). Drug Evaluation and Classification Program (DECP). Drug Recognition Experts (DRE). The International Association of Chiefs of Police (IACP). Retrieved from: http://www.decp.org/experts/.

Li, G., Brady, J. E., & Chen, Q. (2013). Drug use and fatal motor vehicle crashes: a case-control study. Accident Analysis & Prevention, 60, 205-210.

NHTSA (2013). “Preliminary statement of policy concerning automated vehicles.” U.S. DOT

Robertson, R.D., Mainegra Hing, M., Woods-Fry, H., & Vanlaar, W.G.M. (2018a) Road Safety Monitor 2017 Drugs & Driving in Canada. Ottawa, ON. Traffic Injury Research Foundation.

Robertson, R.D., Woods-Fry, H., Vanlaar, W.G.M. Mainegra Hing, M. (2018b) Senior Drivers and Automated Vehicles: Knowledge, Attitudes, and Practices. Ottawa, ON. Traffic Injury Research Foundation. http://tirf.ca/TIRFCAD18JJ

Robertson, R.D., Bowman, K., Brown, S.W. (2017). Distracted Driving: A National Action Plan. Ottawa, ON.: Traffic Injury Research Foundation; and Drop It And Drive^®. http://tirf.ca/TIRFCAD17A

Robertson, R.D., Meister, S.R., and Vanlaar, W.G.M. (2016). Automated Vehicles: Driver Knowledge, Attitudes, and Practices. Ottawa, ON: Traffic Injury Research Foundation. http://tirf.ca/TIRFCAD16J

Sexton, B. F., Tunbridge, R. J., Brook-Carter, N., Jackson, P. G., Wright, K., Stark, M. M., & Englehart, K. (2000). The influence of cannabis on driving. TRL report, 477, 106

Starkey, N. J. & Charlton S.G. (2017). The prevalence and impairment effects of drugged driving in New Zealand. School of Psychology. University of Waikato, Hamilton.

 NZ Transport Agency research report 597. https://www.nzta.govt.nz/assets/resources/research/reports/597/597-The-prevalence-and-impairment-effects-of-drugged-driving-in-NZ.pdf

Stuster, J., Lim, E., Berning, A., and Agimi Y. (2011). Evaluation of the Effects of SFST Training on Impaired Driving Enforcement. National Highway Traffic Safety Administration. DOT HS 811 455.

Traffic Injury Research Foundation. (October 2018). Marijuana Use Among Drivers in Canada, 2000-2015. Ottawa ON. Traffic Injury Research Foundation. http://tirf.ca/TIRFCAD18QQ

Traffic Injury Research Foundation. (December 2017). Marijuana Use Among Drivers in Canada, 2000 – 2014. Ottawa, ON. Traffic Injury Research Foundation. http://tirf.ca/TIRFCAD17M

Traffic Injury Research Foundation. (September 2018). Distraction-Related Fatal Collisions, 2000-2015. Ottawa, ON. Traffic Injury Research Foundation. http://tirf.ca/TIRFCAD18RR

Traffic Injury Research Foundation. (December 2017). Let’s Talk About Crashes. Ottawa, ON. Traffic Injury Research Foundation & Drop It And Drive^®. http://tirf.ca/TIRFCAD17T

TIRF resources

Canadian Coalition on Distracted Driving (CCDD) E-hub – sponsored by The Co-operators: This web-based resource contains a wealth of information about distracted driving that can help stakeholders tackle priority issues. The content is relevant across sectors, disciplines and communities of practice with a vested interest in this issue and who are looking for solutions to address this issue. The E-Hub contains the latest research, educational programs and tools, current data and legislation in Canada, visual aids, other web-based resources and examples of practice.

Drug-Impaired Driving Learning Centre (DIDLC) – sponsored by Desjardins Insurance: The DIDLC is a fully bilingual, web-based educational resource and comprehensive, accessible tool created to inform the development of an evidence-based drug-impaired driving strategy. It was designed to meet the needs of a wide spectrum of diverse stakeholders who are seeking information about this priority topic. The DIDLC contains several modules structured in a question and answer format, similar to other TIRF educational programs: The Problem, Effects of Drugs on Driving, Laws & Penalties, and Tools & Technologies.