Blood Alcohol

By Jim McVeagh 27/07/2010

I blame the feral response to National’s announcement – that they will not be dropping the legal blood alcohol limit to 0.05 g% – squarely on the media. I am certain that it is the media’s propensity always to give the maximum amount of airtime and newsprint to the most hysterical over-reaction that encourages people to use rabid hyperbole instead of rational argument. Blood on their hands? Are they for real? Scandalous! Thunders National Addiction Centre director, Doug Sellman who adds this gem:

“They are throwing that all away and saying, `We are quite relaxed about 30 deaths and 680 injuries and $230m’.

No, Prof. Sellman, National are saying “We suspect there is a great deal of bullshit in your figures”. I have to agree with them.

Estimates of the reduction in harm achieved by lowering the legal Blood Alcohol Concentration (BAC) from 0.08 to 0.05 are usually derived from studies like the ones in the WHO graph:

Assessing the relative risk of an accident will give you a rough idea of how many fewer accidents will occur. And from that you can work out lives saved. Possibly. This work has mostly been done using simulators in very controlled circumstances. The nice smooth looking curve is not so much a measure of the real-world risk, but a measure of the decreased speed of reaction and decision-making. It is therefore unsurprising that this produces such a neat graph – after all, we know that alcohol depresses reaction times in proportion to the BAC. It does not tell us whether accidents and fatalities would be reduced so attaching crash figures to this information is merely an exercise in math rather than a real-world scenario.

Zador’s well-known study (Zador PL, Krawchuk SA, Voas RB.  Alcohol-Related Relative Risk of Driver Fatalities and Driver Involvement in Fatal Crashes in Relation to Driver Age and Gender: An Update Using 1996 Data. J Stud Alcohol2000;61:387-95.) provides us with some actual crash data from the US databases. Zador showed that people with a BAC between 0.02 and 0.049 had a 2.5 times higher likelihood of a fatal crash and those with a BAC of 0.05 to 0.079 had a 6 times higher likelihood of a fatal crash. Unfortunately, as I have pointed out before, these are very wide bands, making it seriously difficult to determine the best place to set a maximum BAC. It is highly likely that the arbitrary nature of the bands has predetermined the arbitrary nature of a 0.05% limit. Again, it does not really address the issue of a safe BAC, but at least it indicates that more alcohol produces more fatal crashes.

A recent meta-analysis has suggested that no BAC may be considered “safe” for a complex activity such as driving. Having read the paper, I have no problem with this conclusion at all. The upshot is that, despite the emotive language and the dubious figures cited in the media, the debate around a reduced BAC limit is entirely missing the point. The question should be whether we should people to drink any alcohol and then drive.

This question is not as simple as it seems. Danyl at the Dim-post baldly puts it thus:

The problem is that the 30 to 60 people that will die during the two year research period don’t know who they are so they don’t know that their freedom has been compromised, while the many thousands of people who like to have a few drinks and then drive home do know they’ll be trivially inconvenienced by a reduction in the drink-drive limits.

Sarcasm aside, this argument overlooks the fact that we already allow dangerous people on our roads. The elderly and the young both have high accident rates even when sober. A teenager with a BAC of 0.5 has an increased risk of a fatal crash up to 17 times the rate of a sober driver. An adult with a BAC of 0.8% is not statistically more dangerous than a perfectly sober 17-year-old. Should we deny the young and the old the ability to drive themselves?

There are people who routinely speed and people who routinely tail-gate. There are those who have sleeping problems and drive tired and those who drive under the influence of pain killers and cough medicines. There are those who drive while adjusting their radios and those drive with children in the car. Some drive cars that are falling apart and some drive left-hand drive imports. For one or two of these drivers we have laws that may cause one to be fined, but the vast majority of these motorists drive with impunity, despite being far more dangerous than the average driver.

We choose to allow these people on the road because we strike a balance between increasing our danger and a person’s freedom to drive. Sometimes the risk is too great (people with no license, the recidivist drunk, the drunken youth) but usually we accept an increased risk of around 2-3 times greater than normal (most of the people mentioned above). The question therefore remains as to exactly what reduction in risk we will be achieving by lowering the BAC level to 0.05%. The data is not yet conclusive. Most countries who have lowered their BAC from 0.08 to 0.05 have experienced about a 8-12% improvement in accident statistics. As the lowering of the limit has invariably been accompanied with a raft of other measures and a police blitz on alcohol, it is hard to be sure exactly what this means. Joyce is right to want to gather better data so that we know whether such a reduction is actually worthwhile.

The decision is not as trivial as Danyl makes out. A BAC of 0.05% may put a large number of responsible citizens afoul of the law and achieve little or nothing in terms of road safety.


Related posts:

  1. Pushing the Limit
  2. Drinking to the Limit
  3. Booze Scoop

0 Responses to “Blood Alcohol”

  • You wrote:
    “after all, we know that alcohol depresses reaction times in proportion to the BAC”

    But the graph above does not seem to indicate that, with no increase in risk up to a BAC of 0.02. In fact, the black line indicates that people with a BAC of 0.03 are safer than sober drivers.

  • I don’t disagree with what you are saying but for me, the issue about lowering the BAC from 80 to 50 is not just about scientific information. I have dealt with hundreds of BAC and BrAC and urine AC calculations and case work over the last ten years and it is very much about what people think they can get away with drinking, i.e. the actual drinking culture.
    New Zealand is a hard drinking country and has so far resisted attempts to alter the national perception that drink driving is OK. Although the UK still has a BAC limit of 80, there is a stigma attached to drink driving that is not present in NZ. Also, let’s remember that the blood:breath ratio in England and Wales is different, which means that rather than a BrAC of 400, the law stipulates 350, which does make a tangible difference in what we see in court cases.

  • Brent:
    The graph is not of reaction times, but of calculated relative crash risk. It is saying that a single unit of alcohol will make little difference in your statistic likelihood of crashing, despite slowing your reaction times a little (i.e. you can compensate for it). It is also saying that fully half of the reduction in risk between 0.08 and 0.05 is achieved by lowering the level to 0.07% (i.e. you consume about half a unit of alcohol less). The conclusion should therefore be that a very slight drop in our alcohol consumption should make a difference in our accident statistics. As it is also very likely that the vast majority of people who consume alcohol and drive have levels 0.07% or less, it follows that the actual reduction in accidents attained by the proposed level reduction will be much, much smaller than is being intimated.

  • Anna:
    I agree with you about the drinking culture of New Zealanders. However, I doubt very much that lowering the limit will make any difference to it. I think you will find that the vast majority of people with a positive breath alcohol will have one below 250mcg (the equivalent of 0.05%BAC). Few will risk driving with a BAC higher than this because it is truly impossible to tell whether you are under or over the 0.08%.

    I think you will find two types of people in the 0.05-0.08% range. Those that did not eat while drinking or drank too fast, pushing their BAC temporarily high, and those who don’t give a rip about the limit but have not (yet) consumed enough to get over it. The first group will be criminalised for a stupid oversight. The second will be caught at some other checkpoint.

    It will be interesting to see if my hypotheses pan out in the research that Joyce has commissioned.

  • I realise that my experience is biased towards those people over or close to the drink drive limit but many thousands of people drive in excess of the breath alcohol level all the time. In my experience (casework and anecdotal), after a couple of drinks most people misjudge how affected they are by alcohol. I have given many presentations about alcohol and its effects in terms of forensic alcohol calculations and the majority of audience members admit to not being sure how much they should consume prior to driving but it doesn’t stop them drinking more than they should, largely beacuse they know their judgement was impaired at the time they made their decision even if they didn’t recognise that until after the event (usually the next day).
    My personal preference would be to reduce the BAC so that it is into the Subclinical zone (as opposed to the Euphoria stage – see for reasons as set out in that post and the associated one.
    I also would like to see the blood:breath ratio adjusted to be in line with what the scientific literature suggests, being 2300:1 (= 350 in breath) rather than 2100:1 (which is NZ 400 in breath).
    I agree in general with your breakdown of how people drink – what I have found is that by explaining to people about the effects of food, their height, weight and gender and the type of alcohol they are consuming they understand how to manage their drinking a bit better. Obviously this approach will work in a positive way for some people whilst others will abuse it – that’s human beings for you.