Intended for healthcare professionals


At what blood alcohol concentration should drink-driving be illegal?

BMJ 1994; 308 doi: (Published 23 April 1994) Cite this as: BMJ 1994;308:1055
  1. A Guppy

    Legislators decide the maximum legal blood alcohol concentration for drivers,1 but their decisions should be informed by the best available scientific evidence, and here researchers have a role.

    Data from case-control studies of accident involvement and experimental studies of alcohol's effects on driving skills were considered before Britain introduced its blood alcohol limit in 1967.2 Results from the Grand Rapids case-control study3 and the Medical Research Council's research on simulated driving4 were interpreted as supporting the limit of 17.4 mmol/l. Different analyses of the data, however, could have supported a lower legal limit of 10.9-13.0 mmol/l.5, 6

    Reviewing data from several case-control studies, the National Research Council of the United States concluded that focusing on the risk of crash responsibility (as opposed to crash involvement) or alternatively controlling for the driver's drinking experience supported legal limits below 17.4 mmol/l.6 It decided that “for large groups of drivers the responsibility for crashes increases when the driver judged culpable for the crash has a blood alcohol concentration greater than 8.7 mmol/l.” When the case control data were controlled for experience with alcohol the risk of crash involvement began increasing at the lowest blood alcohol concentrations measured.

    Behavioural studies suggest that driving related skills are significantly impaired at blood alcohol concentrations below 10.9 mmol/l and that little evidence exists for a threshold below which driving related skills are unimpaired.1,6,7 In another recent review of behavioural studies, Kruger concluded that driving skills were significantly impaired at blood alcohol concentrations between 6.5 and 17.4 mmol/l, depending on the task's demands and complexity.8 Kruger believed, however, that “a general endangerment probably only takes effect at blood alcohol concentrations at or above 15.2 to 17.4 mmol/l.” This probably reflects the fact that experienced drinkers with good psychomotor skills show relatively little impairment, even at blood alcohol concentrations of nearly 21.7 mmol/l, on simple driving tasks.9

    Furthermore, Kruger suggested that “additional arguments” would be needed to justify a limit lower than 15.2 mmol/l. Probably the strongest additional arguments come from the falls in drink-driving behaviour and related casualties that lowering the legal limit may achieve.

    A recent report examined the effect of reducing the limit from 17.4 mmol/l to 10.9 mmol/l in the Australian Capital Territory.10 Among drivers sampled at random road blocks significantly fewer had blood alcohol concentrations over 32.6 mmol/l and between 10.9 and 17.4 mmol/l than before the law was changed.

    In addition, the proportion of drivers with blood alcohol concentrations above 17.4 mmol/l involved in accidents fell by one third, and these effects seemed sustained for at least 12 months after the change. The overall number of accidents fell by 6%. This study suggested that behavioural change is possible without increasing the demands made on the police - one of the factors that would militate against change.

    Whether a limit of 10.9 mmol/l would substantially affect the more serious offenders is difficult to judge. A lower legal limit might not deter those current drink-drivers who believe that their skills are unimpaired by their current levels of drinking unless they thought that there was a high risk of getting caught.11, 12 Many serious offenders (those with multiple convictions or very high blood alcohol concentrations) are dependent on alcohol, and changing the limit is unlikely to affect their behaviour.12 A lower limit operating alongside random breath testing might, however, (???) drink-drivers.13

    The combined results of these studies show clear support emerges for changing the legal drink-driving limit from 17.4 mmol/l to 10.9 mmol/l. While Britain and other countries debate this issue, clearly the higher limits such as those found in some states of the United States and in the Republic of Ireland should be lowered. The researchers have done their job; it is now in the hands of the elected representatives.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    12. 12.
    13. 13.
    View Abstract