By on June 23, 2017

cheech-chong high driving

Tragically, the Insurance Institute for Highway Safety has correlated the legalization of recreational marijuana use with more automobile accidents. Pot smoking in Colorado, Oregon and Washington seems to have resulted in collision frequencies roughly 3 percent higher than what would have been expected without legalization, according to a recent analysis from the Highway Loss Data Institute.

While this certainly isn’t an endorsement for de-legalizing recreational marijuana use, it is a reminder to stay off the roads if you’re having your head changed. Operating a motor vehicle while baked can get you into a sticky-icky situation, and nobody wants you having a green out on the expressway. That said, risks associated with driving under the influence of marijuana are much less cut-and-dried than alcohol.

This is largely due to how difficult it is for researchers to test marijuana. Despite its growing legalization, marijuana is classified by the Drug Enforcement Agency as a Schedule 1 drug, and subject to the highest level of restriction. Researchers need approval from their institution and apply for a license from the DEA before conducting a study. The government also has only so much pot to dole out for research purposes and gives the majority of it to the National Institute on Drug Abuse — fair and balanced testing of whether or not getting high while driving is safe is a little lower on NIDA’s list of priorities. In this instance, the same might be suggested of the IIHS. 

Studies utilizing stimulators have indicated that driving stoned can degrade overall driving effectiveness. However, the results of several studies on the subject of marijuana’s effect on reaction time have proven contradictory, especially in a real-world setting. While some studies have found that THC could more than double a crash risk, others, including a large-scale federal case-control study, were unable to establish any definitive link between marijuana use and crashes.

That’s further exacerbated when testing extends to someone who is no stranger to the herb. Despite the demonstrable neurophysiologic impairments associated with THC, studies involving experienced smokers driving on an outdoor course have shown almost no functional impairment under the influence of marijuana — except when combined with alcohol.

Reasons for the discrepancies between simulation and real-world testing have been hard to pin down. Many researchers have suggested that, unlike drivers under the influence of alcohol (who tend to underestimate their degree of functional deterioration), marijuana users typically overestimate their impairment and consequently employ compensatory strategies.

In one study where participants were given a dose of 7 mg of THC, roughly a third of a joint, drivers rated themselves as impaired even though their driving performance was not. A second test group with a blood-alcohol content of 0.04 percent, about two standard drinks and half the legal limit in most U.S. states, exhibited driving performance impairment even though they rated themselves as unimpaired.

However, the HLDI seems to indicate an appreciable risk factor that lab testing has been unable to replicate. Using neighboring states as comparative controls, the institute undertook a combined analysis of Colorado, Oregon and Washington before and after law changes. The control states were Idaho, Montana, Nevada, Utah and Wyoming. The collected data included collision claims filed between January 2012 and October 2016 for 1981 to 2017 model year vehicles. Analysts attempted to control for differences in driver population, insured vehicle fleet, the mix of urban versus rural driving exposure, unemployment status, and weather patterns.

“The combined-state analysis shows that the first three states to legalize recreational marijuana have experienced more crashes,” explained Matt Moore, senior vice president of HLDI. “The individual state analyses suggest that the size of the effect varies by state.”

Combined, it only accounts for a three percent increase among states that have legalized recreational marijuana usage — although cherry picking certain direct state-to-state comparisons yielded much higher results. For example, Colorado saw a whopping 14 percent increase when compared only to Nebraska.

“The combined effect for the three states was smaller but still significant at 3 percent,” Moore says. “The combined analysis uses a bigger control group and is a good representation of the effect of marijuana legalization overall. The single-state analyses show how the effect differs by state.”

It’s an admirable attempt, but some elements of the study are problematic. Firstly, there’s no assurance that drivers in other states weren’t also smoking marijuana illegally. Colorado, Oregon and Washington may be the first to legalize its usage but recreational cannabis is common in states where the law prohibits it. This also wasn’t a national assessment — it was a regional analysis specifically looking for disparities between several states. While the institute claimed to have made every effort to correct for variables, it couldn’t possibly have accounted for all of them and provided little analysis on individual states’ accident rates before and after legalization.

This is certainly enough to encourage motorists not to risk driving under the influence of marijuana but not enough to say with any degree of certainty that legalizing recreational marijuana use poses an increased risk to road safety. Unfortunately, the IIHS disagrees.

“Worry that legalized marijuana is increasing crash rates isn’t misplaced,” says David Zuby, executive vice president and chief research officer of the Insurance Institute for Highway Safety. “HLDI’s findings on the early experience of Colorado, Oregon and Washington should give other states eyeing legalization pause.”

David Zuby sounds like a total bummer of a dude.

cheechandchong, paramount pictures

[Images: Paramount Pictures]

Get the latest TTAC e-Newsletter!

47 Comments on “Study Links Legalized Recreational Marijuana to Increased Crash Rates...”

  • avatar

    “I think we’re parked, man.”

  • avatar

    My momma talkin’ to me tryin’ to tell me how to live
    But I don’t listen to her ’cause my head is like a sieve
    My daddy, he disowned me ’cause I wear my sister’s clothes
    He caught me in the bathroom with a pair of pantyhose

    My basketball coach, he done kicked me off the team
    For wearin’ high-heel sneakers and actin’ like a queen

    The world’s comin’ to an end, I don’t even care
    As long as I can have a limo and my orange hair
    And it don’t bother me if people think I’m “funny”
    ‘Cause I’m a big rock star and I’m makin’ lots of money
    Money, money, money, money, money, money
    Ahhh! Ha ha ha ha ha ha ha…

    I’m so bloody rich! Ha ha ha ha
    I own apartment buildings and shopping centers! Ha ha ha ha
    And I only know three chords! Ha ha ha ha

  • avatar

    Friday early evening humor:

    Funniest scene from Super Troopers –

  • avatar

    “Pot smoking in Colorado, Oregon and Washington seems to have resulted in collision frequencies roughly 3 percent higher than what would have been expected without legalization, according to a recent analysis from the Highway Loss Data Institute.”


    Because *everyone I know who likes weed* used it before it was legal; it’s just cheaper and easier now.

    Is it that the same old stoners are more high now?

    Or is that 3% all the new amateurs?

    • 0 avatar
      Matt Posky

      Statistically it’s more likely they are newbies. Some studies have suggested that pot smoking has doubled within the last decade. That, or people are much more willing to confess to it.

    • 0 avatar

      Everyone you know that likes weed was probably a bit more paranoid about smoking weed and then hitting the road than they are now.

      • 0 avatar

        Actually they should be more paranoid now, at least in my state. Before legalization there was no standard for impairment and they didn’t regularly test for it after crashes. Now there is a standard and it testing after an accident is much more common.

        • 0 avatar

          This is a good thing, no down side to being forced to accept responsibility for your actions .

          I still love watching “Up In Smoke” every time it comes around on cable .


  • avatar

    The government wasted money on a study that everyone already knew the answer to. What’s next, a study to see if lack of sleep causes drowsy drivers?

    • 0 avatar

      Unfortunately, far too many people didn’t know the answer. Well, they didn’t want to believe it anyway.

    • 0 avatar

      The IIHS is an NGO.
      Their credibility is suspect. They have and agenda, that of the Insurance companies-see the first word in their title. They fought repeal of the ridiculous 55 national speed limit tooth and nail. I all fairness their agenda and actions are a mixed bag. I do like that they are paying attention to lighting performance lately.
      In some respects if someone is a dyed in the wool elected representative government hater, the IIHS is your dream come true. They are influencing automakers to exceed certain government standards. I believe the offset crash test that automakers are scrambling to meet is not a government regulation.

      • 0 avatar

        Agreed. IIHS and my interests only align for the crash tests. The legislative agenda is not, and never has been, in the motorist’s best interests. The only reason IIHS tests are important are because of the regulatory capture of NHTSA-if you buy a worldwide car, then NCAP tests are out there, but otherwise, IIHS crash porn.

      • 0 avatar

        Of course they have an agenda. If I had a captive income source like the auto insurance industry and was sitting on a big pile of money that I wanted to keep or increase I certainly would be looking into the negative impact of recreational drug use.

      • 0 avatar

        This. Re-read this paragraph:

        “In one study where participants were given a dose of 7 mg of THC, roughly a third of a joint, drivers rated themselves as impaired even though their driving performance was not. A second test group with a blood-alcohol content of 0.04 percent, about two standard drinks and half the legal limit in most U.S. states, exhibited driving performance impairment even though they rated themselves as unimpaired.”

        This is the same industry that has historically seized on every excuse from being unmarried, to driving a 2-door sedan, to rolling through a right turn on red at a camera intersection, as an excuse to jack up consumers’ insurance rates, then rammed through laws in various states to make it a criminal offense if you didn’t buy their product. You can see the profit-motive wheels turning here, yet again.

        • 0 avatar

          “This is the same industry that has historically seized on every excuse from being unmarried, to driving a 2-door sedan, to rolling through a right turn on red at a camera intersection, as an excuse to jack up consumers’ insurance rates, then rammed through laws in various states to make it a criminal offense if you didn’t buy their product. You can see the profit-motive wheels turning here, yet again.”

          Ding, ding, ding!

          I started to edit your quote to shorten it while preserving the message, but every word in there is important.

          As the kids say on their internet these days, “THIS.”

        • 0 avatar

          You reminded of the 2-door nonsense. Out of curiosity, in the past I’ve asked agents when insuring coupes whether the price would change if the exact same car were in the 4 door version and the answer was always yes. Once, the difference was a mind-boggling 10%. 10%, for have two big doors instead of four small ones. And no, it wasn’t for crash protection, as the cost of the liability was affected.

  • avatar

    Probably the most important conclusion, which went unmentioned in the article, is that driving under the influence of marijuana doesn’t make a big difference. Probably pets or kids in the car, or cell phones, raise accident rates more.

    • 0 avatar

      This absolutely wrong.
      And it makes a generalized statement that make me nutty and stirs nonsense into a discussion.
      Pot makes everybody react differently.
      I know.
      And different kinds of pot are as equally individually effective and each person.

      I have some pot that makes you think it was stepped on with LSD!

      And there are those that can and those that cannot. People are different skill wise. Some people can walk and chew gum…others can barely walk at all.

      And anybody who says pot doesn’t affect motor skills or is equal to kids in the back seat are insane,ignorant of today’s pot or purposefully misleading.

      And again I’m gonna say this…we are all so liberal with this until OUR family, friend or self is injured in an accident that involved an impaired pot smoker.

      Then you will feel the pain of today’s runaway PC madness and the lack of personal protection and legal options.

  • avatar

    Disclaimer – never smoked weed.

    This is yet another ‘study’ that claims legalizing weed causes more wrecks. This ‘study’ ASSUMES it was weed.

    If it’s true, show me what actually CAUSED those extra 3% of wrecks.

    This group’s study is a veiled attempt to increase our premiums.

  • avatar

    (James Spader voice)
    3%, is that all?
    3%… Huh.

    Echoing bachewy, this just shows that 3% more people that had wrecks also had weed in their system. Unless there is proof they were stoned at the time, AND that they were negligent (such as rolling a stop sign and hit someone, or were backing out without looking and getting hit, etc.), then its worth a look.

    But then again, 3%? People were predicting panic in the streets as stoners ran down children playing Pac Man in their Reliant in their head. And all we got was a 3% increase of some vague definition of supposedly weed-related traffic accidents.

    So much for Reefer Madness on I-5.

  • avatar

    Fake news.

  • avatar

    Are the States where they legalized marijuana and the States they used as a control for their group really comparable? Washington and Colorado both have large metro areas, and even Portland has a larger population than Last Vegas, the only large city in the control group. I have a feeling it’s a bit easier for accidents to go up in large cities that are constantly getting more crowded.

  • avatar

    Subscribed just to read the comments……


  • avatar

    The IIHS doesn’t even know legalizing Pot put at least 3% more drivers on the road, at any given time, thanks to the whole new industry with sub-industries catering to the needs of growers, smokers (don’t forget your edibles!), stores, dispensaries, etc, and all different kinds of personnel/agents/lawyer/CPAs/reps/pros/advisors including on the state/city/county side, of the booming business.

    • 0 avatar

      Seriously, if they are looking at absolute numbers rather than per mile traveled numbers, that is either big screw up or perhaps done deliberately for propaganda purposes. I don’t have numbers to back it up , but seat by the of my pants so to speak, traffic volume here in CO has gone up far more than 3% a year since legalization in ’14.

  • avatar

    In Colorado, the number of fatal and injury crashes combined dropped 0.2% from 2016 Jan-April to 2017 Jan-April.

  • avatar

    All I care about is if the stoners respect the slower traffic keep right/keep right except to pass laws. Can I get an amen? Or as they used to say in Rock Ridge, rrrreverend!

  • avatar

    Dave’s STILL not here man !

  • avatar

    Would we hear about it if a fully competent study showed a decrease in accidents when drivers are under the influence of weed? A 3% increase is getting pretty close to an improvement.

  • avatar

    Shocked, I tell you! SHOCKED!

  • avatar

    That study is flawed in its execution for sure. However, would driving under the influence of weed cause driving skills to degrade? I can reflect back to the 80s when I would drive home at night stoned. Remember, we are not talking about highway driving, but local two lane streets, and some double yellow line streets with limits about 45 or so to get an idea. I found that my sense of hearing was certainly heightened but I also found that I tended to over-react a bit, especially to things that would surprise you such as an animal running out in front of the car. Or I would get a bit paranoid as to when to signal for a turn when there was a copish looking car behind me. Since the route taken home was often the same, I memorized landmarks as points to signal so I would not have to be so nervous. So I do believe there is merit in the idea that stoners are less of a threat than drinking but I do not think getting behind the wheel while stoned is a good idea at all.

    The biggest problem from an enforcement point of view is how to tell if the person is under the influence. You can get pulled over and test positive for THC even if you have not smoked/ate THC for a few days. In fact, long term stoners may never test clear for THC even if they stopped for a couple of months. A vexing problem for sure.

    • 0 avatar

      Enforcing on the suspicion of risk factor, such as testing for blood alcohol level, is very different from enforcing on the basis of observed or evident violation of traffic laws.

      Perhaps since/if being under the influence of marijuana is proven to be such a low risk, if a risk at all, enforcement could be on the basis of evident impairment or proof of violation of traffic laws.

      Seeking proof of dui involving alcohol is valid because we know this is a big risk factor. But seeking proof that a cell phone is in the car and penalizing on that basis actually makes more sense than seeking proof of marijuana influence and with no other proof evident. There should have to be proof of impairment, not suspicion of influence.

      Driving with a burned out taillight does not raise suspicion of influence and is not proof of impairment, while wandering across lanes fairly raises proof of impairment.

      And where are the voices of the personal freedom / anti-government / cash-grab traffic fines on this one?

      Would it be ok to be stoned and at the wheel of a car operating autonomously?

  • avatar

    I would venture to guess that cell phone use by the driver far exceeds the increase in accidents that is presumed in this article to be caused by THC .

  • avatar

    Here we agree as yesterday yet one more typical prius driver glued to this texting, blew a stop sign in front of me and simply ignored me as i followed him a few blocks tooting my horn every time he did it again .

    This next to a Middle School with Children present .


  • avatar

    There does exist just such a device for TV sets ~ sold in England and I tried hard to get one here but never could make it happen .


    • 0 avatar

      There are cell phone jammers…available from the UK. I wanted one when I rode the commuter train.

      • 0 avatar

        The FCC busted a guy who put a cell jammer in his car. The problem was that he took out the cell towers as he passed, and every day a particular tower would crash….he was a commuter. Once a pattern was established, some simple direction finding at the correct time had him caught.

        You want to clear out your area of cell blabbers, but if you take out the Tower, you have just caused a huge problem…..for everyone in that cell…911 calls too.

        I’ve sat at enough lights where you literally have to wait 5-7 seconds PER CAR before they notice the light changed…you can see in the rearview mirrors that heads are down. I’ve sat through one short light in my area many times simply because everyone had to finish the text before driving.

        3% is well within the usual statistical variations….the IIHS is generally NOT a reliable source, they are a partisan lobby….

  • avatar
    Big Al From 'Murica

    So insurance rates will be raised in states that legalize and there is the Crux of the matter…money for insurance companies based on a dubious connection. I am not pro legalization btw, but this is stupid.

  • avatar

    Here come the nanny-state prohibitionists and another attempt to try and thwart legal marijuana in the U.S.

    I’ve driven stoned plenty of times and I have NO accidents on my record.

  • avatar

    I was driving down a road at 55-60 and came up on a car doing 35-40.Followed him until I could pass and when I did me and my wife looked over and the guy was sitting in his seat like it was a barcalounger with a joint in his mouth.
    Same guy drunk would be going 75-80 and weaving all over the road.
    One thing about grass it makes you want to concentrate when you drive,booze just makes you stupid.
    Perfect world don’t do either and drive but have you looked around and paid attention.I’m surprised everyone’s not shooting up

  • avatar

    Cynically, this “study” appears with an Attorney General of the United States that is toughening laws on marijuana use.

  • avatar

    Great story. Thanks, Matt!

Read all comments

Recent Comments

  • namesakeone: Interesting article(s), but one correction: The bottom picture, of the black Matador coupe, is actually...
  • chicklet: I was 30 and came upon a used 5k coupe de ville in 1982, triple white and had to have it. I lived in NYC...
  • Imagefont: So all it needs is a 400kWH battery. Easy. But you might have to spend a few days at the KOA because I...
  • dwford: I was told when I sold cars that the people who paid the most were the happiest, while the people who...
  • ollicat: You guys are right. This is 125 miles under ideal conditions. Take your motorhome out west and climb from...

New Car Research

Get a Free Dealer Quote

Who We Are

  • Adam Tonge
  • Bozi Tatarevic
  • Corey Lewis
  • Jo Borras
  • Mark Baruth
  • Ronnie Schreiber