Traffic deaths: leading cause of death among Americans ages 5 to 34.

December 1, 2011 1:54 PM

Study shows medical marijuana laws reduce traffic deaths. 'A groundbreaking new study shows that laws legalizing medical marijuana have resulted in a nearly nine percent drop in traffic deaths and a five percent reduction in beer sales.' 'The study is the first to examine the relationship between the legalization of medical marijuana and traffic deaths.'

'Daniel Rees, professor of economics at the University of Colorado Denver co-authored the study with D. Mark Anderson, assistant professor of economics at Montana State University.'

'The economists analyzed traffic fatalities nationwide, including the 13 states that legalized medical marijuana between 1990 and 2009. In those states, they found evidence that alcohol consumption by 20- through 29-year-olds went down, resulting in fewer deaths on the road.'

The strong implication being that the effect is through young people substituting alcohol for marijuana. Yet, how to square that with a new study suggesting that Legalizing Medical Marijuana Does Not Increase Use Among Youth?

'“Although we make no policy recommendations, it certainly appears as though medical marijuana laws are making our highways safer,” Rees said.'

posted by VikingSword (74 comments total)

Drive stoned, not drunk!
posted by astapasta24 at 2:00 PM on December 1, 2011 [1 favorite]

And outrageously high gas prices also result in fewer accidents.
posted by Postroad at 2:01 PM on December 1, 2011 [2 favorites]

Correlation, meet Causation. Causation, Correlation.
posted by Danf at 2:02 PM on December 1, 2011 [12 favorites]

Causation, correlation, etc, but they are basically hypothesizing that people drink and drive less because they are smoking out under their medical needs marijuana and driving instead?

Are there really that many people who need prescription marijuana out there who are alternatively irresponsible drinkers?
posted by jabberjaw at 2:02 PM on December 1, 2011

I tap two mana and summon "Spurious Reasoning".

I also support the legalization of medical marijuana.
posted by DWRoelands at 2:03 PM on December 1, 2011 [5 favorites]

Relax, all right? My old man, he's a television repairman. He's got this ultimate set of tools. I can fix it.
posted by Cool Papa Bell at 2:05 PM on December 1, 2011 [5 favorites]

There was just an Explainer article Slate that talks about driving drunk vs driving stoned.
posted by Weeping_angel at 2:09 PM on December 1, 2011

Yet more evidence that we humans really, really like our drugs. If we can't use this kind of drug, we'll use that kind of drug. Whatever works.
posted by zardoz at 2:09 PM on December 1, 2011 [2 favorites]

Oops- an Explainer article in Slate. Also, kudos for the title DWI Versus DW-High
posted by Weeping_angel at 2:11 PM on December 1, 2011

List of countries by traffic-related death rate

Legality of cannabis by country

Someone put these two together, please.
posted by vidur at 2:12 PM on December 1, 2011 [1 favorite]

In Orange County, they have a Drive-Thru medical marihuana business.
posted by Bunny Ultramod at 2:13 PM on December 1, 2011 [1 favorite]

I wonder if this would hold up if there were public, social venues for smoking pot, like there are for alcohol.

(It's my understanding that pot doesn't have the same effect on driving ability, too - that basically you can learn to drive stoned and once you do you don't drive any worse than when you're sober - but I'm not sure how much that's a factor here compared to the fact that there just aren't pot bars.)
posted by restless_nomad at 2:14 PM on December 1, 2011 [1 favorite]

In the driving simulators, pot smokers drove significantly slower than the drunk drivers, even with researchers reminding them to speed up. They also gave the car in front of them a lot more room and were less likely to pass. Alcohol, on the other hand, increases risk-taking behavior. Drunk drivers drive faster, tailgate, and pass recklessly.

Hey man. What's your hurry, man? We'll get there. Chill out.
posted by emjaybee at 2:15 PM on December 1, 2011 [32 favorites]

Just think of the drop in traffic deaths if you included a coupon for half off pizza delivery with each dime bag.
posted by Foam Pants at 2:20 PM on December 1, 2011 [6 favorites]

In a related study the combination of cannabis use and cycling has been strongly correlated with having the best Saturday ever
posted by theodolite at 2:21 PM on December 1, 2011 [51 favorites]

Don't drive stoned, folks. Is it safer than driving drunk? I would say yes. It's not safe though.
posted by furiousxgeorge at 2:23 PM on December 1, 2011 [8 favorites]

the combination of cannabis use and cycling has been strongly correlated with having the best Saturday ever


Cannabis use + X = best Y ever, where

X {Kayaking in heavy waves at the beach, cruising through the woods on a lawn tractor, running naked through sand dunes}
Y {Tuesday afternoon, Friday evening, Sunday morning}
posted by CynicalKnight at 2:28 PM on December 1, 2011 [5 favorites]

Heckuva week you're having there, CynicalKnight!
posted by darkstar at 2:39 PM on December 1, 2011 [12 favorites]

from link: “Our research suggests that the legalization of medical marijuana reduces traffic fatalities through reducing alcohol consumption by young adults.”

I beg to differ, sir. By showing that states with fewer traffic fatalities tend to legalize marijuana, you've actually demonstrated conclusively that the people killed in traffic accidents are almost always pot-smoking hippies who vote.
posted by koeselitz at 2:40 PM on December 1, 2011 [3 favorites]

Posts about marijuana make everyone seem funnier.
posted by Ron Thanagar at 2:43 PM on December 1, 2011 [3 favorites]

Correlation, meet Causation. Causation, Correlation.

Causation, correlation, etc...

I tap two mana and summon "Spurious Reasoning".


On the one hand, I want to rail against people who heard "correlation does not equal causation" on the Internet a few too many times and decided They Really Got It and think that it means that no one can ever do science.

On the other hand, I want to know why this US-centric study was "published" by a private German labor institute.
posted by gurple at 2:51 PM on December 1, 2011 [7 favorites]

Man, the correlation != causation thing is tiresome. Realize that if a correlation does exist, barring a statistical fluke, there is some way to explain the relationship.

You're now implicitly throwing your hat behind one of two alternate explanations:
1. The causal relationship actually works the other way (e.g. less drinking causes medical marijuana to become legal)
2. A third factor that causes both less drinking and medical marijuana legislation.

Maybe medical marijuana legislation is passed in response to a dearth of drinking among 20-29 year-olds. Maybe less drinking and medical marijuana are both caused by some third factor that happens to only exist in those 13 states, the nature of which I would love for someone to elucidate. Sure. But are those really more reasonable explanations than actual causation?
posted by 0xFCAF at 2:55 PM on December 1, 2011 [21 favorites]

Isn't a medical marijuana dispensary a rather upstanding place? I'd imagine nobody buys pot stoned, while people definitely buy alcohol drunk, think a bar. Does medical marijuana negatively impact bars? You might even find some human attention issues, like worrying about the illegality causing accidents.

I prefer legalization marijuana through growing for personal use, along with decriminalization of personal sales, rather than the "big pot" creating medical marijuana route.
posted by jeffburdges at 2:56 PM on December 1, 2011 [1 favorite]

However, since those who oppose relaxing marijuana laws in any way claim that it will increase traffic fatalities, while this cannot prove that doing so leads to reduced auto deaths, it does show that it does not automatically result in an increase.
posted by Maias at 2:57 PM on December 1, 2011 [6 favorites]

The booze lobby won't take this lying down, passed out with vomit on it's shoes.
posted by Ron Thanagar at 2:58 PM on December 1, 2011 [11 favorites]

In summary, laboratory tests and driving studies show that cannabis may acutely impair several driving-related skills in a dose-related fashion, but that the effects between individuals vary more than they do with alcohol because of tolerance, differences in smoking technique, and different absorptions of THC. Driving and simulator studies show that detrimental effects vary in a dose-related fashion, and are more pronounced with highly automatic driving functions, but more complex tasks that require conscious control are less affected, which is the opposite pattern from that seen with alcohol. Because of both this and an increased awareness that they are impaired, marijuana smokers tend to compensate effectively for their impairment by utilizing a variety of behavioral strategies such as driving more slowly, passing less, and leaving more space between themselves and cars in front of them. Combining marijuana with alcohol eliminates the ability to use such strategies effectively, however, and results in impairment even at doses that would be insignificant were they of either drug alone. Case-control studies are inconsistent, but suggest that while low concentrations of THC do not increase the rate of accidents, and may even decrease them, serum concentrations of THC higher than 5 ng/mL are associated with an increased risk of accidents. Overall, though, case-control and culpability studies have been inconclusive, a determination reached by several other recent reviewers. Similar disagreement has never existed in the literature on alcohol use and crash risk.

The author goes on to say:

In the meantime, patients who smoke marijuana should be counseled to have a designated driver if possible, to wait at least three hours after smoking before driving if not, that marijuana is particularly likely to impair monotonous or prolonged driving, and that mixing marijuana with alcohol will produce much more impairment than either drug used alone. (source)
posted by AElfwine Evenstar at 3:02 PM on December 1, 2011 [3 favorites]

gurple: “On the one hand, I want to rail against people who heard ‘correlation does not equal causation’ on the Internet a few too many times and decided They Really Got It and think that it means that no one can ever do science.”

Sociology isn't science; and this study doesn't even rise to the level of sociology.

0xFCAF: “Maybe less drinking and medical marijuana are both caused by some third factor that happens to only exist in those 13 states, the nature of which I would love for someone to elucidate. Sure. But are those really more reasonable explanations than actual causation?”

For the love of god, yes. Social and political dynamics are not simple. They are filled will innumerable variables that all have to be accounted for. It's only as easy as drawing a straight line once you've covered every bit of ground in every situation.

The states that have legalized marijuana, for example, tend to have more liberals. Liberals as a group tend to drink less – not necessarily because of ideology or anything, it's just a correlation.

This is a very simple alternate explanation, but the truth is probably vastly more complicated. There probably isn't some one third factor, but dozens of other factors, that enter into this equation. Expecting simple, direct results when discussing the predilections and actions of millions upon millions of people is silly.
posted by koeselitz at 3:07 PM on December 1, 2011 [1 favorite]

"... are filled with innumerable..."
posted by koeselitz at 3:08 PM on December 1, 2011

Sociology isn't science; and this study doesn't even rise to the level of sociology.

Lovely. What, then, is your specific issue with the linear model they lay out on page 12? Or are you just broadly asserting that no scientific inquiry can be made into these kinds of data? If so, why not?
posted by gurple at 3:13 PM on December 1, 2011 [2 favorites]

Montana is full of liberals?

In any case, yes, it's complicated and there are many variables. The authors are pretty upfront about that - they looked at traffic fatalities because good records are kept on those, and the end of their report is full of tables with regression analyses and whathaveyou.

Don't attribute to the authors what some idiot wrote in a press release, is what I'm saying.
posted by rtha at 3:15 PM on December 1, 2011 [5 favorites]

Actually, koeselitz, I just jumped on you for that one sentence, but you pretty much laid out exactly "why not" in your second paragraph. OK, fair enough -- you don't think any scientifically defensible claims can be made using differences in data gathered between states, no matter what adjustments are made or what obtainable variables are taken into account.

That's a fantastically wacky claim, but at least I understand where you're coming from.
posted by gurple at 3:17 PM on December 1, 2011

"Montana is full of liberals?"

Montana has also recently realized that drive through beer stores might not be a good idea, that speed limits might be a healthy thing, that the laxest drunk driving laws in the country might have something to do with having the highest drunk driving related fatality rate in the country, and that fixing that might be a good idea. There might be some confounding variables in Montana.
posted by Blasdelb at 3:31 PM on December 1, 2011 [1 favorite]

People with glaucoma shouldn't be driving anyway.
posted by CheeseLouise at 3:33 PM on December 1, 2011 [2 favorites]

While not specifically medical marijuana, there have been plenty of these kinds of studies done internationally in the past; none of which showed an increase in accidents due to moderate use of cannabis when not combined with alcohol. As a side note, I have no understanding of how it's possible for at least some of the subjects who reported "having used marijuana from one to 10 times per month previously" to not feel, or at least taste, that they weren't smoking weed.

During the study, some subjects were given actual marijuana cigarettes, and some were given a placebo, with neither the investigator nor the subject knowing which they had smoked. Another administrator kept track of who was given which type of cigarette.

posted by gman at 3:37 PM on December 1, 2011 [1 favorite]

Sociology isn't science

Really? I was under the impression that is was pretty widely accepted as a social science.
posted by AElfwine Evenstar at 3:37 PM on December 1, 2011

This study cannot be dismissed with lazy 'correlation != causation' accusations. Let me explain you why.

They controlled for:

Unemployment

Income

Miles driven

Decriminalized marijuana

Graduated Driver Liscence

Primary seatbelt law

Secondary seatbelt law

BAC 0.08 law

Administrative revocation law

Zero Tolerance Drunk Driving Law

Beer tax

and whether the state has any speed limit of 70

This means, they accounted for the impact of all these changes, and they are looking at the residual change left. The question is, did something in particular affect the little bit of change left, or was it just noise?

Then they compared the state which have passed a MML law with those which haven't.

They found that, on the whole, the little bit of change left

always moves just after MML law passes

always moves in the same direction

always moves roughly the same amount

and doesn't move if MML law didn't pass.


If you want to claim this is a coincidence, you need to find something which has changed -- by coincidence -- in exactly those exact 13 states, at the exact moment of the passage of the MML law, which hasn't changed in any of the other states, which has some reasonable mechanism of affecting in concert:

Achool consumption

The number of crashes involving alcohol

but without affecting the number of crashes not involving alcohol.

and that thing is not MML, and it is not any of the controls I just listed.

That seems like a tall order.

This is essentially the closest you will ever get to a controlled study, in the social sciences. It very rare to get a chance to collect evidence of this high quality.

posted by gmarceau at 3:40 PM on December 1, 2011 [88 favorites]

Awesome! As time goes on, the safety of marijuana vs. alcohol keeps going up...
posted by agregoli at 3:43 PM on December 1, 2011

If you made a pill that caused people to be better at driving drunk and made the pill widely available, would you not expect drunk driving accidents to decrease?

You would not need to change anything else, same drinking habits, same driving habits.

Turns out that habitual stoners are better at driving drunk.
posted by Ayn Rand and God at 3:43 PM on December 1, 2011 [4 favorites]

gurple: “That's a fantastically wacky claim, but at least I understand where you're coming from.”

Well – look, to be honest, it's the stuff on page 12 (the central methodology for drawing the direct conclusion) that I mostly have a problem with, because it doesn't actually take into account a host of other variables that I think are probably important. But there are some better parts of the study. To their credit, they seem to have tried to come at it from a lot of different angles. Toward the end of the study, they use other methods to try to suggest the connection, building up a variety of sources and connections.

It's really not fair of me to reject this whole study out of hand, anyway. I apologize.

me: “Sociology isn't science”

AElfwine Evenstar: Really? I was under the impression that is was pretty widely accepted as a social science.”

My own beliefs don't really enter into this. I don't think sociology is science, but that's not part of this conversation. As such, that comment of mine was a derail. Sorry. I don't need to go spouting invective for no reason.
posted by koeselitz at 3:43 PM on December 1, 2011 [3 favorites]

Yet, how to square that with a new study suggesting that Legalizing Medical Marijuana Does Not Increase Use Among Youth?

Offhand, I'd say one could square it by suggesting that youth are smoking the same amount of pot but getting it from legal sources, perhaps at one remove. Not to assert that's actually the case, of course, but that scenario would seem to fit the data.

Correlation, meet Causation. Causation, Correlation.

Obligatory xkcd link

Drive stoned, not drunk!

If memory serves me right, in his autobiographical novel Junky, William S. Burroughs claimed that marijuana was beneficial and that laws against it ridiculously draconian, but that it rendered one absolutely unfit for driving a car. YMMV, of course.
posted by Gelatin at 3:47 PM on December 1, 2011

Correlation, meet Causation. Causation, Correlation.

I was going to respond to this by making some of the same points gmarceau did above. But since he covered it so well, I'll just say -- wow, am I tired of pseudo-sophisticated stats 101 dismissals of evidence by people who aren't even bothering to read the linked paper.
posted by zipadee at 3:50 PM on December 1, 2011 [12 favorites]

it's the stuff on page 12 (the central methodology for drawing the direct conclusion) that I mostly have a problem with, because it doesn't actually take into account a host of other variables that I think are probably important.

What other variables exactly? Keep in mind that the state and year fixed effects, as well as the state specific time trends, will capture a huge number of other variables that are not explicitly controlled for. They capture, for instance, lasting cultural differences between a 'hippie' state like California or Colorado that legalized medical marijuana and a more conservative state like Alabama or whatever. They also capture linear time trends in traffic fatalities that differ between states, as well as any national trends in traffic accidents.
posted by zipadee at 3:55 PM on December 1, 2011

If memory serves me right, in his autobiographical novel Junky, William S. Burroughs claimed that marijuana was beneficial and that laws against it ridiculously draconian, but that it rendered one absolutely unfit for driving a car. YMMV, of course.

Like the study suggests, a few different factors do come into play there, among them potency, how soon after smoking you drive, and how you administer the cannabis. 10 gravity hits of kind half an hour before getting behind the wheel is a phenomenally bad idea; a bowl of standard shwag a couple hours beforehand, probably not really impairing. But hell no, I would never smoke and drive. My driving sucks enough as it is.
posted by Marisa Stole the Precious Thing at 3:57 PM on December 1, 2011

One more thing.

If you want to claim this is a coincidence, you need to identify a factor which, in addition to affecting (as I was saying)

• Achool consumption
• The number of crashes involving alcohol
• but without affecting the number of crashes not involving alcohol.

it also need to affect

• the number of crashes on weekend and nights
• without affecting the number of crashes during weekday daylight.

What could possibly do this other than MML? Or is it just all a big coincidence?

Think of it this way, if this was a murder mystery, nobody here would be yelling, "please let the poor MML go, clearly all this must be a coincidence."
posted by gmarceau at 4:07 PM on December 1, 2011

@Vidur:

Quick and dirty spreadsheet. -0.008; no correlation.

(No warranty is introduced here. or extroduced.)
posted by nickrussell at 4:12 PM on December 1, 2011 [1 favorite]

Yet, how to square that with a new study suggesting that Legalizing Medical Marijuana Does Not Increase Use Among Youth?

This new study is perfectly consistent with the old one. Check Table 3 on page 29 of the new study. Like the old study, their data also indicates that marijuana use did not increase among youth in RI.

The main effect the new study found on crashes they found came from older demographics.
posted by gmarceau at 4:12 PM on December 1, 2011

I'm reading this thread on my phone.

Maybe I should pull over to the shoulder.
posted by box at 4:16 PM on December 1, 2011 [3 favorites]

nickrussell, thanks!
posted by vidur at 4:20 PM on December 1, 2011

Hey man...... am I drivin' ok?
posted by Liquidwolf at 4:24 PM on December 1, 2011

Maybe I should pull over to the shoulder.

Well at least make another post when you get home so we know you're alright.
posted by sammyo at 4:35 PM on December 1, 2011 [1 favorite]

As someone who spent far too many years riding a motorcycle, people try to defend driving stoned pas harmless piss me off. Just because it's not as bad as driving drunk does not mean it's not being reckless with 2 tons of steel. Aggressive driving is not the only thing that's a problem, being able to noticed unexpected events (like say that pedestrian jaywalking, or that motorcycle that is turning left etc) and being able react quickly in a crisis are important as well and from my experience both of those are hurt by being stoned.

If you are stoned don't fucking drive ok?
posted by aspo at 4:55 PM on December 1, 2011 [6 favorites]

I agree that no one should drive impaired, but there are HUGE variances in "stoned," and not all instances of smoking marijuana lead to Cheech-and-Chong-like stupification.
posted by agregoli at 5:07 PM on December 1, 2011 [1 favorite]

Yeah, and most people can drink a beer and drive right afterwards. (I'd say everyone, but I've known a few people who really couldn't.)
posted by aspo at 5:11 PM on December 1, 2011 [1 favorite]

laws legalizing medical marijuana have resulted in a nearly nine percent drop in traffic deaths and a five percent reduction in beer sales.'

Really? Coz I only enjoy weed when combined with booze, especially beer.
posted by jonmc at 5:25 PM on December 1, 2011

Cool story bro.
posted by Marisa Stole the Precious Thing at 5:34 PM on December 1, 2011 [1 favorite]

Sociology isn't science; and this study doesn't even rise to the level of sociology.

I'm going to use this at parties.

Really? I was under the impression that is was pretty widely accepted as a social science

Well I have friends who claim to be scientists because they got MBA's and call their education "the science of management", but that doesn't necessarily mean its "science".

Science has to have:

1. testable concepts
2. reliance on accurate gathering of reliable measurable data.
3. be objective
4. based on practical investigations/empirical knowledge.

I've read sociology papers about the effects of alcohol on certain groups...and I've read sociology papers about 'the simpsons and their effects on post-modern tv' or some shit.

Truth is that sociology is huge...and its largely as scientific as the uni dept wants to make it. So yeah...there's a lot of krap there. But there also might be some science.

But yeah...for the most part, I don't really see it anywhere near science.

I mean seriously...even psychologists make fun of sociologists when they talk about "science".
posted by hal_c_on at 6:01 PM on December 1, 2011 [1 favorite]

In California, at least, it's not uncommon to go to a dispensary stoned. I've sworn it off, though, because it inevitably leads to standing at the counter for what seems like forever, completely unable to make any sort of purchasing decision whatsoever.
posted by You Can't Tip a Buick at 6:01 PM on December 1, 2011

If you are stoned don't fucking drive ok?

For real. Slower reaction time = more likely to die or worse...kill others.
posted by hal_c_on at 6:02 PM on December 1, 2011 [1 favorite]

In California, at least, it's not uncommon to go to a dispensary stoned. I've sworn it off, though, because it inevitably leads to standing at the counter for what seems like forever, completely unable to make any sort of purchasing decision whatsoever.

I've frequented a lot of dispensaries...from the valley to malibu to venice to LA (proper). I have NEVER seen anybody stoned as they come in. A lot of places have chill-out rooms (seriously, its like an opium den. I once saw 2 boxes of krispy kremes at one place, and the owner told me it only costs him a few bucks a day to get it, but he swears that most of his customers come in because they get a donut after getting stoned in the chillout room) where you can smoke...but I've never seen somebody already stoned getting mmj.

Although, at least a third of the people I've dealt with have already been stoned big time and had problems calculating how much 2 ounces cost.
posted by hal_c_on at 6:07 PM on December 1, 2011

^^"people i've dealt with" =dispensary employees
posted by hal_c_on at 6:07 PM on December 1, 2011

I'm more interested in the data that seems to indicate that increased marijuana use correlates with reduced methamphetamine use.

In other words, marijuana is a gateway drug, away from harder drugs.

Meanwhile, there have been a number of studies that suggest that cigarette smoking may be a gateway 'drug' to meth addiction. (There was another study on this announced in the news just a couple of weeks ago that I can't find at the moment).

So once again, why is it that marijuana is illegal, when cigarettes and alcohol are legal? Don't we have this backwards?
posted by eye of newt at 7:47 PM on December 1, 2011 [5 favorites]

I want to rail against people who heard "correlation does not equal causation" on the Internet a few too many times and decided They Really Got It and think that it means that no one can ever do science.

I spent ten years in the research field. I know a little something about it.
posted by DWRoelands at 8:13 PM on December 1, 2011

I don't see the connection DWRoelands.
posted by Ayn Rand and God at 8:49 PM on December 1, 2011

why is it that marijuana is illegal, when cigarettes and alcohol are legal? Don't we have this backwards?

I found that to be a sensitive topic here in the past.
posted by CynicalKnight at 9:15 PM on December 1, 2011

Are there really that many people who need prescription marijuana out there who are alternatively irresponsible drinkers?

Hey, how's it going?
posted by chaff at 10:04 PM on December 1, 2011

I'd like to point out a possibility that seems to miss mention. What if the thing is that small doses of weed make YOUNG drivers better drivers? Or, perhaps, the experience of driving high cause a general increase in caution that extends even when not high, especially amongst young drivers?

See, in the far past, I was young, believe it or not! And back in my youth, I was of the opinion my driving was improved by what I call a "light buzz". ("Stoned" is not the same at all. By definition, stoned means impaired. Would you call someone "drunk" after 1 beer?). In my specific case, my driving suffered from my being tense. Driving with a slight buzz taught me to relax better, and then I learned in general how to drive better.

But another thing about comparisons like this over time, my experience smoking pot today is completely different. Today, I'd not drive after smoking, at all (not that I drive anyway, prefer public transit). Funny enough, today I'd also never get stoned except by accident.

Oh, still another issue occurs to me, and that's the difference between driving with a buzz, and driving with a buzz while carrying a bag of dope. Oops! HUGE difference, if and only if, the stuff is illegal. Medical marijuana isn't illegal. You're don't have to get wound up and paranoid about every cop car you see.
posted by Goofyy at 12:27 AM on December 2, 2011 [1 favorite]

This is essentially the closest you will ever get to a controlled study, in the social sciences. It very rare to get a chance to collect evidence of this high quality.

To be fair to the people who didn't bother to read it, it is 35 odd pages long. Is this normal in the social sciences? And what happened to IMRAD?

I think they could have gone a long way to allaying the correlation != causation fears by demonstrating a dose-response relationship between the proportion of populace in each state at each age entitled to prescription marijuana and the reduction in deaths. I fact the opposite seems to be true, with older drivers, who have more marijuana entitlements being no safer, which is lazily put down to the fact that they're old and slow even before they get high.
posted by roofus at 6:49 AM on December 2, 2011

How does one get "stoned by accident"?

Just wondering because even being in a car clam baking the hell out it back in the day before I smoked produced zero results minus smelling like weed.
posted by handbanana at 6:50 AM on December 2, 2011

I once got remarkably stoned merely by attending a 311 concert.
posted by jeffburdges at 7:27 AM on December 2, 2011

Just jealous as all the times I've gotten high (too damn many to count) I had to smoke it, vape it, eat it. Wish I could cop a free buzz.
posted by handbanana at 7:38 AM on December 2, 2011

I once got very high thanks to the guys standing behind me at a PJ Harvey show. They were about a foot taller than my friend and I, and smoked for what seemed like the first hour of the show.

God love San Francisco, where smoking a cigarette inside at the Warfield or the Fillmore or etc. will get you tossed out, but light a joint or a bowl? You just get looks from people who wish they'd brought some, too.
posted by rtha at 8:40 AM on December 2, 2011

As a stimulant, weed effed me up more than any other drug on this planet (yes, including coke, meth, etc.). I was a fairly straight-edge, responsible rec drug user with the harder stuff as I got older.

Driving (actually, doing anything) on weed for me was "OH FUCK MY LIFE'S A MESSANDISTHATTHECOPSBEHINDME?!?!?DUDETHROWITOUTTHEWINDOW/PUT A TOWEL UNDER THE DOOR!!"

Yeah, that guy.

I'm sure your mileage varied, and I will forever resent humanity for my inability to enjoy that gorgeously-scented, readily-available anywhere herb . . .
posted by eggman at 1:22 PM on December 2, 2011

Don't drive stoned, folks. Is it safer than driving drunk? I would say yes. It's not safe though.

You are incorrect. I tend to be more aggressive and impatient behind the wheel when not properly dosed. When I am stoned I zone in, focus, and drive in a mellow and non-confrontational way.

I am a better and safer driver when I am properly medicated, there is no question about it. Anecdotal reports from friends and colleagues suggest that this is a common experience among pot smoking motorists.
posted by Meatbomb at 3:28 PM on December 2, 2011

If you want to claim this is a coincidence, you need to identify a factor

States with more people smart enough to vote for MML also have more people smart enough to get designated drivers? The kinds of attitudes in a community that would lead to approving MML also lead to people being less likely to drive drunk? Some kids get drunk as a way of escaping from having to deal with the intense cancer pain their parents are dealing with?

But I'm with koeselitz in thinking that it's really unlikely there is "a factor" that explains differences in these kinds of complex behaviors. I think social science is a real science, but it faces a lot more difficulties than other sciences in trying to make confident statements about the world. When you can't assign states randomly to a treatment / control group, there's needs to be a big asterisk next to your results.
posted by straight at 1:30 PM on December 5, 2011