When is a drunk driver actually DRUNK?
November 15, 2010 7:14 AM   Subscribe

What's the science underlying increasingly severe "blood alcohol" laws, particularly with regard to "good" drivers?

Driving while drunk constitutes a menace to humanity. No argument there. But I am wondering if there's any science that suggests an otherwise good driver (no record of accidents, no record of speeding, recklessness) is suddenly more likely to cause vehicular mayhem if they happen to have a low-to-moderate amount of alcohol in their system (blood alcohol content somewhere between .05 and .08).

I understand that any driver who's had a few drinks is not operating with the same "skill level" they'd be at if they were cold sober. But is a "good driver" actually more likely to have accidents (ie: would they not just slow down, drive with more care, realizing that they are operating at a compromised skill level just as a "good driver" adapts when they realize they're tired and/or distracted)?

FOR EMPHASIS: I'm not talking about already bad drivers here (record of prior accidents and/or recklessness). And I'm not talking about DRUNKS here (ie: someone who couldn't walk or think straight).
posted by philip-random to Science & Nature (17 answers total) 1 user marked this as a favorite
 
There is good evidence that some of the functions significantly involved in driving car, particularly at night, e.g. depth perception, peripheral vision, and glare recovery, can be affected by BAC concentrations as low as 0.06.

It doesn't really seem to matter just how good your driving record is: if you are impaired, you are at an increased risk of an accident. It's that simple. You are not entirely in control of your faculties. A person who is normally very careful will probably still be more careful with a little bit of alcohol in them than a habitually reckless driver would be with same BAC. But neither is as careful as they would have been without the booze.

So yes, there is good reason to think that alcohol consumption makes everyone more likely to have an accident, across the board, but it's probably worth thinking about that increase as a percentage than as an absolute increase in risk. So if you're, say, twice as likely to have a wreck if with a BAC of, 0.1 (just to make up some numbers), but person A only had a 1% chance of having a wreck and person B had a 5% chance, person B is still more dangerous, but both are more dangerous than their normal activities.
posted by valkyryn at 7:22 AM on November 15, 2010 [2 favorites]


Here you go.

The short of it is that even very small amounts of alcohol produce measurable negative effects in reaction time and judgment.

Driving is a tremendously dangerous activity at the best of times and it's not at all unreasonable to make laws against doing it while your judgment is impaired. (And as for your tired/distracted examples, many places already have laws against driving after too long a period without sleep or driving while performing certain other activities like using a cell phone.)

And yes, the good driver with 0.05 BAC is probably much less of a risk than the stone sober bad driver. And so, it is unfair that the former gets punished but not the latter. But it's better to have laws that are easy to apply in these situations. Also, I think it's reasonable to become skeptical of the broader judgment of people who decide they are a good enough driver to drive safely with a couple of drinks in them.

Incidentally, I say all this as someone who loves drinking and loves driving. But I don't feel upset about not being allowed to combine these interests.
posted by 256 at 7:42 AM on November 15, 2010 [1 favorite]


Have a look here.
It's a study by R F Borkenstein, from 1964, although more modern studies have similar results. Note the "Borkenstein dip".

Disclaimer: UK website which was lobbying against a reduction from 0.08 to 0.05, which has now been scrapped.
posted by derbs at 8:56 AM on November 15, 2010


Drinking impairs a lot of basic functions that valkyryn already listed. It's important to remember that other drivers (those who are not inebriated) drive with the expectation that everyone else is not impaired. So if they need to slam on the brakes, they expect the driver behind them to react in a timely fashion and not rear-end them. Someone who's been drinking (even at low BAL) won't react as quickly and is more likely to get in an accident because of this. It's not about how careful your are. And driving drunk doesn't necessarily make you less careful. But it does decrease your awareness and your ability to react to quickly developing situations on the road.
posted by smokingmonkey at 9:41 AM on November 15, 2010


In my experience, it's the 'good' drivers who drink-drive the most.

I've known two professional drivers who habitually, i.e. several times a week, would drink 4-5 pints then drive home. This always confused the hell out of me, as they were facing the same risks as everyone else but with the added possibility of destroying their entire career. Massive arrogance, I suppose.

To be fair, neither were in any accidents I knew of nor ever got caught. Horrendous behaviour regardless.
posted by Quantum's Deadly Fist at 10:05 AM on November 15, 2010


Well, it's impossible to give some sort of grade to all drivers and then allow them different blood alcohol limits according to their skill level. Personallu, I don't understand why puritanical America just do what most eastern European countries do, which is to ban driving if you have have any alcohol in your blood at all. Takes all the guesswork and human error out of deciding whether you're okay to drive or not.
posted by Dee Xtrovert at 10:11 AM on November 15, 2010


I don't think this question is about faculty impairment. If anything, it's the opposite. The question is not "can a person with X alcohol drive as well as that same person with no alcohol". The question is "can this person with X alcohol drive as well as a completely different person also with X alcohol". If there's a difference (not already due to metabolization and so forth) then presumably the better driver is compensating for impairment. Therefore, statements about how much alcohol is impairing a faculty are non-operative.

But in either question, looking at the data for how much your peripheral vision (or whatever) is reduced while drinking is suspect to me. Look at the recent data that banning texting while driving, or requiring helments on bikes, is either ineffective or even counterproductive. I'm not going to defense those particular items, my point is just that you can't look at what "obviously" would have an effect--you have to go and measure the actual effect.
posted by DU at 10:25 AM on November 15, 2010


Response by poster: Thanks for the answers so far. One thing that comes to mind as I read through them is that I neglected to use the word "statistics" and/or "statistical" in my question.

ie: is there statistical evidence to suggest that ... an otherwise good driver (no record of accidents, no record of speeding, recklessness) is suddenly more likely to cause vehicular mayhem if they happen to have a low-to-moderate amount of alcohol in their system (blood alcohol content somewhere between .05 and .08)?

And yes, this is a leading question. I live in a jurisdiction which has recently lowered the allowable blood alcohol limit from .08 to .05, with some particularly draconian punishments if one should fail a breath test (instant loss of license, big fines etc). That said, I'm not a regular driver and certainly not a drinking driver. I've just become fascinated at this sudden change of law which doesn't seem to have a lot of science to back it up.
posted by philip-random at 12:05 PM on November 15, 2010


The measurement problem in your question remains: how do you define or identify an otherwise "good" driver? I've never had an accident, I never speed, and I've never been in trouble with the courts. On the other hand I catch a train to and from work every day and drive a car perhaps three times a month.

Here are some related numbers on road deaths Victoria in Australia, indicating that a large minority of drivers who die are affected by alcohol.
posted by Fiasco da Gama at 12:32 PM on November 15, 2010


Oh, that. Legislators do that sort of thing all the time. Taking a "tough on crime" or "zero tolerance" approach to drunk driving is a politically popular thing to do, and it's supported by numbers far better than most of the other things they could come up with.

But presuming that legislation is based on some rational consideration of all relevant facts is silly. Legislators pass things to help themselves get re-elected. Appearing tough on crime is a pretty uncontroversial way of doing that, particularly here, where the opposition has to work really hard not to sound like that guy who is obviously drunk but insists that he's totally fine.
posted by valkyryn at 12:58 PM on November 15, 2010


Fiasco, that's an interesting chart. Assuming that most drink drivers are "only a little bit over (bloody idiots)" i.e. 0.05 - 0.10, the statistics show that they're not really causing too many crashes, it's the truly hammered ones over 0.15 that are the real risk, even though we can presume they're not too frequent on the road. So on a strictly risk base, 0.05 seems quite low.

Of course, it's easier to stop people getting to 0.15 if you set the bar at 0.05.
posted by wilful at 5:03 PM on November 15, 2010


Wilful, it certainly indicates that drivers who are very affected by alcohol die at a higher rate than those who are moderately affected by alcohol, but I'm not sure it indicates anything about those rates compared to drivers who are not affected at all (and as such doesn't really answer the OP's question).

Here's one from NZ which has the conclusion that:
The New Zealand risk of driver fatal injury during the main drinking times increases steeply with increasing BAC. At BAC=20 mg/dl, an average risk for drivers aged over 19 is twice his/her risk at BAC=0. At BAC=20 mg/dl, the estimated risk for the average teenaged driver is almost three times their non-drinking risk. Risks are significantly higher at all BAC levels for drivers aged under 20 and drivers aged 20-29 than for drivers 30 and over
So yes, driving skill and age/experience seem to be very important as well as BAC, but risk and BAC go hand-in-hand.

Having given the matter a bit of thought though, from the original question:
But is a "good driver" actually more likely to have accidents (ie: would they not just slow down, drive with more care
One of the major effects of alcohol that's not directly a matter of physical capability or eyesight or reflexes is the lowering of inhibitions, increased feeling of self-confidence, euphoria and decreased ability to assess risk. There's no reason to think drivers who have been drinking will take precautions against their own impairment—quite the opposite.
posted by Fiasco da Gama at 5:30 PM on November 15, 2010


well, to get meta on the question, a "good" driver is one who's not so stupid as to think that they are unimpaired by alcohol.

being "good" is a matter of attitude at least as much as it is a facility with hand-eye coordination. So a good driver is one that doesn't drink drive. I realise this is a bit "no true scotsman" fallacy...
posted by wilful at 5:44 PM on November 15, 2010


Response by poster: I put "good" in quotes because I didn't feel like going down a rabbit hole discussing a definition of "good", (a value judgment if there ever was one). As I suggested in my initial question, I consider a "good" driver not someone who necessarily obeys all the rules of the road but just someone who drives well (gets predictably and consistently from point A to point B without causing any havoc), regardless of what may or may not be in their blood, on their mind etc.

Fiasco da Gama: One of the major effects of alcohol that's not directly a matter of physical capability or eyesight or reflexes is the lowering of inhibitions, increased feeling of self-confidence, euphoria and decreased ability to assess risk. There's no reason to think drivers who have been drinking will take precautions against their own impairment—quite the opposite.

Except your New Zealand stats suggest that more mature drivers do take such precautions. That is, given the same amount of alcohol in their blood, the younger drivers are more likely to mess up (less able to assess risk and take precautions -- ie: drive slower). No?

So maybe I should have worded the question as:

What's the science underlying increasingly severe "blood alcohol" laws, particularly with regard to MATURE drivers with "good" driving records?
posted by philip-random at 6:36 PM on November 15, 2010


What's the science underlying increasingly severe "blood alcohol" laws, particularly with regard to MATURE drivers with "good" driving records?
I can tell you that easily, it's not about science: it's about designing a simple workable system of road rules. You can't (at least where I live) legislate for offences which are different for different people depending on age, or previous criminal history, or unmeasurables like "maturity", and you have to have an enforceable system of sanctions. If you have to increase the severity of laws in response to a perceived social problem like the road toll—about which there seems to be agreement—then you have to do it fairly, in a non-discriminatory way.

What you can do, as my State has in Australia, is have more stringent rules on specific classes of at-risk drivers (zero for L-platers, .02 for those on provisional licences, as well as lower speed limits, passenger restrictions, and fewer "points" to lose before licence restrictions come into effect).

But pity the poor night-shift copper pulling drivers over in the rain and having the same argument about .05 and speed and experience over and over again.
posted by Fiasco da Gama at 7:12 PM on November 15, 2010


I can tell you that easily, it's not about science: it's about designing a simple workable system of road rules.

+11. It's like people who complain about speed limits. "I can't believe this road is 80km/h! I have a good car! I am good driver! I should be allowed to do 100km/h!"

Well, that may be the case - but how the hell would you ever implement such an arrangement, much less police or enforce it? Particularly when the number of variables tends to increase sharply, as they have in your question. What are we going to do - get people drunk to 0.08 when they take their driver's test and see how they do?

"Congratulations, Mr Jones, here's your license to drive with a BAC of 0.12 on white spirits in your Volvo C70 between 4pm and 6pm while wearing your prescription lenses following a meal of at least 400 calories consumed not more than half an hour before you begin a journey that does not include an interstate."

"Gee willikers, Mr Jones! Can I take the drunk driving test?"

"Sorry, Jimmy. Not til you're mature."

posted by obiwanwasabi at 12:04 AM on November 16, 2010


Response by poster: Particularly when the number of variables tends to increase sharply, as they have in your question. What are we going to do - get people drunk to 0.08 when they take their driver's test and see how they do?

For the record, in my jurisdiction (British Columbia), I'm pretty sure there's already a different set of rules for younger drivers and blood-alcohol (zero tolerance up to a certain age, I think). Otherwise, my curiosity was prompted by a change (for everyone else) from .08 down to .05. My suspicion was that it was driven more by politics (playing to public opinion) than anything resembling hard science. Hence my question ...
posted by philip-random at 8:59 AM on November 16, 2010


« Older Can I lease and then finance the balloon payment...   |   Screen printed my fingertips off. Help! Newer »
This thread is closed to new comments.