The purpose of the legal limit on drivers’ blood alcohol concentration (BAC) is to reduce death and injury on the roads. After over 50 years of continual dissemination of public information, it should be widely known by now that the best advice is never to drive after drinking — and if the world were an ideal one in terms of road safety, almost every driver’s BAC would be zero or near zero. But there is more to life than road safety, and legislation is about what it is reasonable to require of people for the common good. Therefore up to the time of writing, against a background of advice to avoid driving altogether after drinking any alcohol, the legal limit in England and Wales on a driver’s BAC stands at 80mg/100ml (80mg of alcohol per 100ml of blood), as set in 1967. It was realised early in the history of motoring that too much alcohol made one unfit to drive, and this was recognised in law by creating the offence of driving while under the influence of drink; nevertheless, in the mid-1960s there was still active debate as to whether moderate drinking increased or (as seemed reasonable to some at that time) decreased the risk of collision, and hence the risk of death or injury on the roads. All this changed with the publication in 1964 of a large-scale field study at Grand Rapids in the USA in 1962—3: its findings quantified the relationship between BAC level and the risk of involvement in a collision, providing convincing evidence for greatly increased risk which depends on drivers’ alcohol levels. This brought the issue of a legal BAC limit onto the political agenda, and the question then became “What maximum level of BAC is appropriate?” The value chosen — 80mg/100ml — was probably determined by a combination of statistical considerations and wider ones. Much has changed since 1967. International practice, including the widespread adoption elsewhere in Europe of limits of 50mg/100ml or lower, has combined with changed public attitudes and increased statistical evidence and understanding to cast considerable doubt upon whether the BAC limit of 80mg/100ml, set in 1967, is still the most appropriate one today. Successive UK governments have retained this limit, but over the last 20 years the question of whether it should be lowered to 50mg/100ml has been debated with increasing vigour. The matter was considered as part of a far-reaching review in 2010, and the limit was lowered to 50mg/100ml in Scotland in 2014. The numbers of casualties in reported collisions which involve a drink-driving offence reveal a marked fall over the period from 1979 (when reporting largely assumed its present form) to 2013. With enforcement of the 80mg/100ml limit, accompanied by public information about the dangers of drink-driving, the annual numbers of those killed in such collisions, and of those seriously injured in them, had by 2013 fallen to 15% and 13% respectively of the corresponding numbers for 1979. By comparison, over the same period there were lesser falls — to 30% and 28% respectively — in the numbers killed and numbers seriously injured in all other kinds of reported collisions. Therefore by 2013, the last year for which complete data is available at the time of writing, the numbers killed or seriously injured in reported collisions involving a drink-driving offence stood at only about one half of what they would have been if drink-driving had continued to contribute to road casualties to the extent that it did in 1979. Numbers of casualties in reported collisions which involve a drink-driving offence are often mistakenly described as the numbers of casualties ‘due to drink-driving’, but doing this omits casualties that occur in collisions where a driver has a BAC which is appreciable — say at least 20mg/100ml — but is within the legal limit. Some of these casualties would not have occurred if the driver’s BAC had been lower, and are therefore also attributable to drink-driving, though without the driver having committed a drink-driving offence. Lowering the limit from 80 to 50mg/100ml can be expected to reduce casualties further through moderation of drinking among those currently driving in the following three categories: a. those with BACs below 50, and thus already within the new limit, but wishing to feel more confident of keeping within it; b. those with BACs between 50 and 80 and wishing to comply with the new limit, just as they were with the existing limit; or c. those with BACs somewhat above the existing limit, say up to 110mg/100ml, but who were intending to comply with the existing limit and would still intend to comply with the new limit. In principle, lowering the limit might also lead to a moderation in drinking among those driving with BACs well above the existing limit; however, public information campaigns and enforcement of the drink-driving law since 1967 may well have had sufficient cumulative effect on drivers for those who still drive well above the limit to prove largely impervious to any lowering of it. To make a rigorous estimate of the reduction in casualties that could be expected from lowering the limit from 80 to 50 mg/100 ml would require, for each BAC level: • knowledge of current numbers of casualties in collisions involving a driver at that BAC level; • for those currently driving at that BAC level, knowledge of the BAC levels at which they would drive under the new limit; and • knowledge of the amount by which their risk of involvement in a collision would change as a result. Relevant information to meet the first and third of these requirements is available from routine collision statistics, interpreted with the help of certain assumptions, and from the technical literature; with respect to the second requirement, however, it is necessary to rely wholly upon assumptions. A range of assumptions is therefore proposed concerning those with BACs in the ranges 0—20, 20—50, 50—80, 80—110 and over 110mg/100ml. Under these assumptions, further reductions in casualties in collisions at BACs under 20 and over 110mg/100ml are regarded as small enough to be neglected here. For each of the BAC ranges 20—50, 50—80 and 80—110mg/100ml, estimates are made of numbers killed and numbers seriously injured in collisions involving drivers with BACs in these ranges, and also of the proportions of these numbers that would be prevented by lowering the limit. The pattern of numbers of casualties in collisions which involve a drink-driving offence has been stable over the years 2010—13, and at levels that are lower than those seen in earlier years by a margin which is substantially greater than for casualties in collisions of other kinds. Calculations for this stable period indicate that in round figures, for every four deaths recorded as occurring in collisions involving a drink-driving offence, there is one more death occurring in a collision involving a drinking driver with a BAC below the legal limit that might not have happened if none of the drivers involved had been drinking at all. Further calculations lead to the estimate that lowering the BAC limit from 80 to 50mg/100ml at the beginning of 2010 would, over the four years 2010—13, each year have saved about 25 lives and saved about 95 people from being seriously injured. (Author/publisher)
Abstract