Drugs and Fatal Collisions: What the Canadian Numbers Really Tell Us

A new report from the Traffic Injury Research Foundation (TIRF) contains a startling statistic: in 2023, 41.2% of the road deaths in its Canadian database involved at least one driver who tested positive for drugs. Alcohol was involved in 29.3%. The findings are contained in TIRF's Drug Use in Fatal Collisions in Canada | 2000-2023 fact sheet.

That does not mean that drugs impaired a driver in 41.2% of fatal collisions. It is an important distinction, and one that is easy to miss when statistics about drug-impaired driving are reported.

So what does the TIRF research actually tell us?

Drugs have become a major part of the fatal-collision picture

TIRF's fact sheet examines information from its National Fatality Database. In 2023, 571 road deaths involved at least one driver who tested positive for drugs, compared with 487 in 2019 and 230 in 2000.

In TIRF's analysis, a fatality is classified as drug-related when at least one driver involved in the collision tests positive for a drug. The drugs can include cannabis, illicit drugs, prescription medications and over-the-counter medications.

The 571 fatalities represented 41.2% of all fatalities in the database in 2023. That compares with 35% in 2019 and 10.7% in 2000.

TIRF also reports that since 2013, a larger percentage of road deaths in its database has involved drugs than alcohol, distraction or speeding. In 2023, drugs were identified in 41.2% of road deaths, compared with 29.3% involving alcohol.

That represents a significant change from the traditional road-safety focus on alcohol when impaired driving is discussed.

A positive test does not prove impairment

This is where the numbers need some explanation.

TIRF is clear that the presence of a drug does not necessarily mean that the person was impaired by it. A positive test does not establish that the drug caused the collision, nor that it was the primary or only contributing factor.

A collision can have several contributing factors. Speed, alcohol, distraction, fatigue, road conditions, vehicle condition and driver behaviour can all be involved at the same time.

There is also an important difference between detecting a substance and determining whether it affected the driver's ability to operate a vehicle safely.

That distinction is particularly important with cannabis. THC, the principal psychoactive component of cannabis, does not provide a simple equivalent to a blood-alcohol concentration. A measurable amount of THC in a blood sample cannot by itself tell us exactly how impaired a particular person was when they were driving.

None of this makes the TIRF findings unimportant. It simply means that we should describe what the data actually measure rather than turn a positive toxicology result into proof of impairment.

The testing has also changed

There is another reason to be cautious when comparing the beginning and end of TIRF's 24-year data series.

Testing of fatally injured drivers was much less complete in the earlier years. TIRF reports that the drug-testing rate was about 49% before 2010, compared with 82.5% from 2011 onward.

That means some of the apparent increase over time may be due to better detection and more complete information.

The more recent data are therefore more informative than the earliest years, while the long-term trend still shows that drugs have become increasingly prominent in the fatal-collision picture.

Who is testing positive?

The characteristics of the drivers in the more recent data are interesting.

Between 2019 and 2023, 53.7% of fatally injured drivers who were tested were positive for at least one drug.

The difference between men and women was relatively small. A positive result was found in 54.1% of fatally injured male drivers and 52.2% of female drivers.

The highest percentage was among drivers aged 25 to 34, at 63.8%. The corresponding figures were 42.1% for drivers aged 16 to 19 and 43.8% for those aged 65 and older.

These are characteristics of drivers who died in collisions and were tested. They do not tell us how likely a driver in the general population is to use drugs or drive while impaired.

When are drug-positive fatal collisions happening?

The timing of these collisions is also noteworthy.

Between 2019 and 2023, the highest percentage of drug-positive drivers in fatal collisions occurred between 3:00 and 5:59 in the morning, when 47.7% tested positive. The figure was 41.9% between 9:00 p.m. and midnight.

Drug-related fatalities were also more common on Saturdays than on Mondays.

Single-occupant vehicles stand out as well. TIRF found that 50.6% of fatalities involving a single occupant were classified as drug-related, compared with 35.6% involving vehicles with two or more occupants.

These findings do not tell us why the collisions occurred, but they identify circumstances that deserve attention when road-safety programs are developed.

What about cannabis?

Cannabis is particularly interesting because its use became legal for adults in Canada in October 2018.

TIRF compared cannabis-positive results among fatally injured drivers before and after legalization. The average was 22.4% before legalization and 27% afterward.

It would be tempting to interpret that difference as evidence that legalization caused an increase in cannabis-related fatal collisions. The TIRF report does not make that conclusion, and the data do not establish it.

Cannabis-positive results increased from 22% in 2013 to 30.4% in 2020, but then declined to 25.2% in both 2022 and 2023. TIRF says more monitoring is needed before conclusions can be drawn about the longer-term effect of legalization.

There is also a practical reason not to try to establish a personal “safe” THC level for driving. The Province of British Columbia says there is no standard or safe time to wait before driving after using cannabis, and notes that a person may not feel impaired while still being under the influence of cannabis.

The practical answer is therefore much simpler than the science: if you have used cannabis, don't drive.

And what about British Columbia?

There is an important limitation for those of us who live in B.C. British Columbia is not included in TIRF's figures because B.C. coroner data for 2023 were not available when the fact sheet was prepared. TIRF therefore excluded B.C. data from its analysis for all of the years covered by this report.

That means the 41.2% figure should not be described as a B.C. statistic.

B.C. does have its own relevant data. The BC Coroners Service Motor Vehicle Incident Deaths, 2013-2023 report records 331 motor-vehicle incident deaths in B.C. in 2023. Its available data for 2013 through 2021 found that alcohol and/or drugs were a contributing factor in 33% of MVI deaths. The report notes that alcohol- and drug-related data for 2022 and 2023 were not yet available because those investigations had not all been completed.

The B.C. figure should not be compared directly with TIRF's 41.2%. The sources cover different periods and use different definitions and methods. The B.C. figure includes deaths where alcohol and/or drugs were determined to be a contributing factor, while TIRF's figure is based on fatalities involving at least one drug-positive driver and excludes B.C. from the analysis.

Even with those limitations, the B.C. data provide useful context. They show that alcohol and drugs have also been identified as contributing factors in a substantial proportion of motor-vehicle deaths in this province.

So what should we make of the numbers?

I think there are two conclusions that can reasonably be drawn from the report.

First, drugs are clearly an important part of Canada's fatal-collision picture. The number and proportion of fatalities involving at least one drug-positive driver have increased substantially, and in recent years the proportion has exceeded the proportion involving alcohol in TIRF's database.

Second, we should be careful about what we call that evidence. A positive drug test is not the same thing as proof of impairment, and it is not proof that the drug caused the collision.

Those two statements are not contradictory.

We can take the presence of drugs in fatal collisions seriously without pretending that toxicology can answer questions that it cannot.

For drivers, the practical message doesn't require a toxicology report. Cannabis, illegal drugs and some prescription and over-the-counter medications can affect the skills required for safe driving. If a medication or other substance can impair you, don't drive until you are confident that you are safe to do so. If you have used cannabis, B.C. guidance is straightforward: don't drive.

And if you are a passenger, remember that you have a choice too. Getting into a vehicle with an impaired driver puts you at risk even though you haven't taken the drug yourself.

The TIRF numbers give us a reason to pay attention to drugs and driving. They also give us a good reminder that responsible use of statistics means understanding exactly what they do — and do not — tell us.

Learn More

• Impairment by Prescription Medications
• Cannabis & Collisions: Dr. Jeff Brubaker
• Care and Control of a Motor Vehicle While Impaired