Visualize: Insights that power innovation

From seed to smoke: Driving under the influence

By Lucian McMahon October 25, 2017

Verisk explores the marijuana DUI riskThis is the fourth installment of our series on recreational and medical marijuana, From Seed to Smoke: Risks Across the Marijuana Supply Chain. Read the first article in the series. 

Exploring the risks

In this article, we’ll discuss some of the risks related to people using marijuana, including driving under the influence and employment-related concerns.

But wait, what is marijuana?

Let’s first define our terms. Marijuana is a type of plant that, if consumed properly, can give the user a “high,” generally induced by the psychoactive chemical THC. There also exist marijuana-related products that may not induce a psychoactive response to the user, such as cannabidiol (CBD). CBD is a non-psychoactive cannabinoid compound found in marijuana and hemp with apparent therapeutic effects.

Those states permitting recreational and/or medical marijuana regulate the purchase and use of marijuana by consumers.

Recreational marijuana

In eight states and DC, so-called “adult,” “personal,” or “retail” marijuana and marijuana-related products may be used for non-medical (“recreational”) purposes, subject to certain limitations. In general, marijuana stores can only sell marijuana products to adults 21 years or older - and many states limit the per diem quantity of recreational marijuana or marijuana-related products that a single individual may purchase. Additionally, states typically enforce restrictions on driving under the influence of marijuana, limit the amount of marijuana an adult may possess, and bar the use of marijuana products in public (with some exceptions).

Medical marijuana

Additionally, in many states, so-called “medical marijuana” and/or marijuana-related products may be purchased and used for medicinal or therapeutic purposes, subject to certain limitations.

Comprehensive medical marijuana: According to the National Conference of State Legislatures (NCSL), 29 states as well as Washington, D.C., have  so-called “comprehensive medical marijuana” programs. The NCSL defines a “comprehensive” program as, in part, as one that offers protection from criminal penalties for medical marijuana use; authorizes some kind of cultivation and distribution system for marijuana products; and permits the use of a variety of types of marijuana products, including the smoking or vaporizing of such products, when appropriate.

Limited access medical marijuana: An additional 16 states have so-called “limited access” medical marijuana programs, which the NCSL defines as programs that permit the use of “low THC, high cannabidiol (CBD)” products "in limited situations or as a legal defense.” These programs tend to be more restrictive than their comprehensive counterparts. For example, Missouri’s CBD law, in part, limits use to those with “intractable epilepsy;” limits the CBD product to 5 percent or more of CBD and “no more than three-tenths of one percent of THC;” and prohibits home cultivation of marijuana products.

Unlike state regulations governing retail marijuana that permit any adult aged 21 or older to purchase marijuana products, only individuals with qualifying medical conditions may purchase marijuana or marijuana-related products (such as CBD-infused products) from a licensed dispensary establishment. What constitutes a qualifying medical condition varies by state. Comprehensive and limited medical marijuana programs generally permit certain licensed physicians to “recommend,” “authorize,” or “certify” marijuana to patients with permitted qualifying conditions.

Marijuana use and 'stoned' driving

Marijuana and marijuana-related products that contain THC are intoxicants. It therefore stands to reason that marijuana use may impair driving abilities. Indeed, the U.S. National Institute on Drug Abuse (NIDA) states that marijuana “significantly impairs judgment, motor coordination, and reaction time, and studies have found a direct relationship between blood THC concentration and impaired driving ability.”

With marijuana use comes accidents: Some studies have found that marijuana use is correlated with an increase in the frequency and severity of car accidents. In May of last year, the American Automobile Association (AAA) published its findings from a number of studies commissioned to examine the relationships between marijuana use, impaired driving, and safety concerns in the state of Washington. The state legalized recreational marijuana use in 2012. One study found that fatal crashes “involving drivers who recently used marijuana” doubled since the state legalized the drug. Out of the 3,031 drivers involved in fatal crashes between 2010 and 2014, 10 percent had “detectable THC in their blood at the time of the crash.” The proportion of fatal crashes involving marijuana doubled from 8.3 percent in 2013 to 17 percent in 2014.

In June of this year, the Highway Loss Data Institute (HLDI) announced the results of its first analysis of collision claim frequencies in states that have legalized recreational marijuana, including Colorado, Oregon, and Washington. HDLI found, in part, that collision claim frequencies in the three states “are about 3 percent higher overall than would have been expected without legalization.” HLDI examined collision claims data between January 2012 and October 2016 using various controls, including neighboring states without legal recreational marijuana. The analysis found the following:

  • "After retail marijuana sales began in Colorado, the state’s estimated collision claim frequency increased 14 percent more than in nearby Nebraska, Utah, and Wyoming. 
  • Washington’s estimated increase in claim frequency was 6 percent higher than in Montana and Idaho
  • Oregon’s estimated increase in claim frequency was 4 percent higher than in Idaho, Montana, and Nevada.” 

However, the announcement noted that the estimated effect of legalization depended on the comparison state: “results for Colorado vary from a 3 percent increase in claim frequency when compared with Wyoming to a 21 percent increase when compared with Utah.”

But is marijuana really to blame?: However, not all studies show the correlation between marijuana use and a more frequent and severe risk of car accidents. A recent study published in the American Journal of Public Health found no significant increase in vehicle crash fatality rates in Washington and Colorado after recreational marijuana legalization. The study examined crash fatality data between 2009 and 2015 in both states and eight additional control states. The researchers “compared year-over-year changes in motor vehicle crash fatality rates (per billion vehicle miles traveled) before and after recreational marijuana legalization with a difference-in-differences approach that controlled for underlying time trends and state-specific population, economic, and traffic characteristics.” The study found the rates “not statistically different from those in similar states without recreational marijuana legalization.”

Furthermore, a National Center for Biotechnology Information study found that marijuana’s effects on impairment varies greatly between individuals thanks to tolerance, dosage, THC absorption, and method of consumption. This particular study states that the evidence is inconclusive on whether marijuana increases the risk for accidents – but goes on to note that mixing alcohol and marijuana “is greater than the risk of driving under the influence of either alone.”

The US National Highway Traffic Safety Administration's (NHTSA) 2015 study echoed these findings. According to NHTSA, “analyses incorporating adjustments for age, gender, ethnicity, and alcohol concentration level did not show a significant increase in levels of crash risk associated with the presence of drugs. This finding indicates that these other variables (age, gender, ethnicity, and alcohol use) were highly correlated with drug use and account for much of the increased risk associated with the use of illegal drugs and with THC.” In fact, a 2016 study published in Addiction found that previous risk estimates of driving under the influence of THC were found to have been too high.

One study that found evidence suggesting an association between recent marijuana use and increased accident risk noted that epidemiological complications make it difficult to know when marijuana is to blame for an accident. This is largely due to the nature of marijuana intoxication itself.

Marijuana intoxication levels can vary significantly

The length of a user’s intoxication from marijuana varies widely by consumption method, potency of the product, and the user’s physiological characteristics, according to AAA. Smoking may induce immediate onset of intoxication, with intoxication lasting several hours; oral consumption may induce intoxication as long as two hours after ingestion, with longer-lasting intoxicating effects.

Identifying marijuana intoxication is further complicated by THC’s persistence in a user's body for long periods after intoxication. AAA notes that THC can be “detectible in blood or urine for weeks, with high variability between people […] Simply detecting any THC does not therefore indicate impairment; relying upon any detectable THC or the presence of a THC metabolite has been a flaw in prior research on driver culpability in crashes.”

In fact, at least one court has reportedly addressed the issue of potential DUIs for drivers not actually driving under the influence. According to a December 2016 Washington Times article, the Arizona Court of Appeals has decided that medical marijuana patients arrested for driving under the influence of marijuana in Arizona can contest the charges by arguing they were not too impaired to drive. The article reported that the Court found that the burden of proof falls on police officers to prove that the alleged offender was too impaired to operate a motor vehicle, regardless of the presence of marijuana in the alleged offender’s system found during a blood test. (See the full decision here.)
The Arizona Medical Marijuana Act reportedly allows medical marijuana patients to have marijuana in their systems; furthermore, the appeals court cited an earlier Arizona Supreme Court decision, which ruled that Arizona’s medical marijuana law “affords an affirmative defense for those patients who can show, by a preponderance of the evidence, that the concentration of marijuana or its impairing metabolite in their bodies was insufficient to cause impairment.”

Marijuana limits vary by state

Different states that permit marijuana use may regulate driving under the influence in different ways.

Some states that allow marijuana use enforce so-called per se drugged driving laws, establishing a maximum allowable limit of THC in the bloodstream of a driver. For example, Washington State enforces a limit of 5 nanograms per milliliter of THC in a user’s bloodstream; amounts detected above that limit could constitute an offense of driving under the influence.

But some have reportedly argued that this limit is low enough to implicate regular medical and recreational users of marijuana because of THC’s potentially long persistence. In fact, a recent study examined “whether data from drivers arrested for suspected driving under the influence (DUI) supported any particular quantitative threshold for a per se law for THC.” Of particular interest, all of the concentrations thresholds examined in the study “would have misclassified a substantial number of drivers as impaired who did not demonstrate impairment on the [Standardized Field Sobriety Test] (SFST), and would have misclassified a substantial number of drivers as unimpaired who did demonstrate impairment on the SFST.” These findings led the researchers to generally conclude that quantitative thresholds for THC “cannot be scientifically supported” in regards to impaired driving.

Other states, such as Oregon, do not enforce a specific limit; instead, Oregon law requires police officers to base their determination of a drugged driver on observations, apparently in response to the complications of detecting quantifiable marijuana intoxication.

(For more information on state DUI laws for marijuana, see this National Conference of State Legislatures webpage.)

There are efforts currently underway to test for marijuana intoxication while operating a motor vehicle that will take THC persistence into account. For example, Stanford University announced that affiliated researchers have developed a potential tool for detecting marijuana intoxication by screening for the presence and concentration of THC in the saliva of a potentially intoxicated driver. The researchers apparently hope that saliva “may correlate with impairment better than THC in urine or blood.”

Evolving issues

The extent to which marijuana intoxication can increase driving risks, how marijuana intoxication can be measured, and how regulators can best address these considerations are all evolving issues. The ISO Emerging Issues team continues to research and track potential considerations related to marijuana. You can find more of our research on ISOnet (ISOnet login required).

Stay tuned for our next installment, when we’ll look at considerations related to marijuana use and employment-related issues, including workers’ compensation.

Lucian McMahon, CPCU, ARM-E, AU-M, is a product development specialist with the ISO Emerging Issues team. You can contact Lucian at