EUR-Lex Access to European Union law

Back to EUR-Lex homepage

This document is an excerpt from the EUR-Lex website

Document 32001Y0214(01)

Commission recommendation of 17 January 2001 on the maximum permitted blood alcohol content (BAC) for drivers of motorised vehicles (Text with EEA relevance)

OJ C 48, 14.2.2001, p. 2–10 (ES, DA, DE, EL, EN, FR, IT, NL, PT, FI, SV)

In force


Commission recommendation of 17 January 2001 on the maximum permitted blood alcohol content (BAC) for drivers of motorised vehicles (Text with EEA relevance)

Official Journal C 048 , 14/02/2001 P. 0002 - 0010

Commission recommendation

of 17 January 2001

on the maximum permitted blood alcohol content (BAC) for drivers of motorised vehicles

(2001/C 48/02)

(Text with EEA relevance)



1.1. Background

1.1.1. Article 71(1)(c) of the Treaty on European Union requires that common transport policy in the EU should include measures to improve transport safety. In April 1997, the Commission adopted a programme of measures to improve road safety(1). This programme included a specific reference (pages 20 and 41) to the effectiveness of reducing inappropriate drinking and driving as an accident and casualty reduction measure. A follow up Communication was adopted on 17 March 2000(2) which reassessed the priorities in the EU for improving road safety. In this, inappropriate drinking and driving is once again identified as a key road safety priority.

1.1.2. Inappropriate drinking and driving is defined normally as driver impairment in terms of:

- a driver with a blood alcohol content (BAC) above a prescribed legal limit which is punishable as a criminal offence,

- or as a driver with a (BAC) above a prescribed limit which is punishable by some form of administrative sanction such as a fine,

- or as a driver with a (BAC) below a prescribed limit but where the driver is judged to be impaired.

1.1.3. Since the mid-1960's, a substantial body of research has analysed the driver impairment effect of alcohol consumption, and has conclusively demonstrated the direct correlation between BAC and the risk of accident involvement. Policies to reduce inappropriate drinking and driving, sometimes constructed around reduced BAC limits, have consistently been regarded in all motorised countries as a priority to reduce accidents and their consequent casualities(3)(4)(5). These policies are normally based upon a mixture of BAC limits, police enforcement, publicity, rehabilitation, and legal or administrative sanctions, with the mix varying according to Member State.

1.1.4. In 1988, the Commission brought forward a proposal for a Directive on harmonised BAC levels, the aim of which was to set a maximum BAC limit of 0,5 mg/ml throughout the Community(6)(7). This proposal was not adopted because of the objections of some Member States on the grounds of subsidiarity. In July 1999, the Transport Committee of the European Parliament decided not to reconfirm its first reading of COM(88) 707 final, and requested that a new proposal be drawn up to better reflect advances in the assessment of the measures required to reduce inappropriate drinking and driving.

1.1.5. The Commission considers that both national and European action to reduce inappropriate drinking and driving within the EU is best served by a more harmonised regime of legal maximum BAC limits. This will provide a clearer and more consistent message to drivers in the EU that drinking and driving is a dangerous activity wherever they are driving in the EU. However it accepts that there are long standing national political objections on the grounds of subsidiarity to such a measure, and therefore does not intend to renew the original proposal for a Directive in line with the commitment made in the Communication on road safety priorities. The Commission therefore proposes to bring forward a Commission Recommendation which sets out the Commission's view on a number of actions relating to:

- BAC limits

- random breath testing

- harmonisation of measurement standards for breath testing equipment

- information exchange concerning best practices

- support for research and development

- support for Europe-wide publicity campaigns

- coordinating the use of alcohol-related road accident data records

For the time being, the Commission does not intend to withdraw the original proposal for a Directive in case Member States wish to reactivate it as a consequence of the Communication.

1.2. The Magnitude of the drinking and driving problem

1.2.1. There have been many assessments of the road safety problem of inappropriate drinking and driving and what should be done to reduce it(8). In some Member States it has been estimated that between 1 % and 5 % of drivers in traffice have BAC levels above their respective maximum national BAC limits. Furthermore, such drivers are proportionately much more likely to be involved in road accidents, accounting for up to 20 % of fatal and serious injuries, and up to 25 % of driver fatalities. They are still a significant road safety problem despite large reductions in drinking and driving road accidents during the past twenty years.

1.2.2. The Commission estimates that at least 10000 road users still die annually in road accidents involving inappropriate drinking and driving, and that this is a reasonable benchmark to focus on the loss to society in the EU. The social value of this lost life using the EUR 1 million rule formulated in the EU Road Safety Programme amounts to EUR 10 billion per annum.


2.1. The benefit of more harmonised BAC limits within the EU

2.1.1. The main benefit of more uniform legal maximum BAC limits within the EU is to provide a clearer and more consistent message to drivers of passenger and freight vehicles that, above a certain limit, drinking and driving is a dangerous activity. From a driver perspective more uniform limits will also provide a standard reference for national enforcement, and Europe-wide publicity campaigns. Where drivers are driving within the EU, they should be more aware of a more uniform limit above which, if they are caught, they will know they have commited a drink driving offence.

2.2. The determination of an appropriate maximum BAC limit for the EU

2.2.1. The European Commission is of the view that the complete detachment of drinking from driving would significantly reduce road accidents, and particularly fatalities in road accidents. Available statistics indicate that at least 10000 lives could be saved annually if drink driving was eliminated. However, the extent to which this ideal state could be achieved depends critically upon the level of enforcement (both roadside testing and technological driver restriction) and driver compliance with the law. Logically the legal BAC limit (and breath equivalent) required to support this goal should be zero, which implies in practice a BAC measurement of between 0,1 mg/ml and 0,2 mg/ml, depending on the tolerance level that is allowed for.

2.2.2. From a road safety perspective, the complete elimination of inappropriate drinking and driving is the optimum goal. But it has to be accepted that without public acceptance and credible enforcement, setting a BAC limit that is too low will not be effective. The Commission view is that when determining a legal maximum BAC limit, or regime of limits, in the EU, the following issues need to be taken into account:

- the level above which the risk of accident involvement, and especially fatal accident involvement, starts to increase significantly,

- the higher risk of accident involvement of some road user groups, such as inexperienced drivers and riders of motorised two wheel vehicles,

- the need to contribute to driver training and education schemes, especially for younger drivers, and,

- the potentially greater levels of death, injury, damage, disruption, congestion and delay following an accident which larger vehicles can cause.

2.2.3. The Commission view is that, in general, consideration should be made to the fact that public attitudes towards BAC limits seem to indicate support for lower limits in the EU(9). On the basis of accumulative research evidence and analysis, the maximum BAC limit of 0,5 mg/ml, first proposed on 5 December 1988, should be the recommended maximum legal limit within the EU.


3.1. Legal basis

3.1.1. Most Member States have already invoked a BAC limit of 0,5 mg/ml in their national legislation either as a legal limit, above which a driver is deemed to have committed a criminal offence, or as a BAC limit above which an administrative fine or additional sanction is incurred. The Commission recommends that all Member States should adopt this maximum limit, and move towards even lower limits in the interest of increasing road safety. However, the Commission believes the choice of BAC limit, together with its consequential enforcement and penalties, is a matter for national authorities.

3.2. Reductions in BAC limits by Member States

3.2.1. The following Member States have already lowered their BAC limit since they introduced their first legal limit:


3.2.2. However, the following Member States have not reduced their BAC limits since they were first introduced into their national legislation:


3.2.3. In addition, Spain has adopted a lower BAC limit of 0,3 mg/ml for inexperienced drivers, for drivers of heavy goods vehicles and buses, and drivers of vehicles carrying dangerous goods. Austria has adopted a lower limit of 0,1 mg/ml for the same broad categories, and also motorcyclists under 18 years of age. The Netherlands are considering whether to adopt a lower BAC limit of 0,2 mg/ml for inexperienced drivers.


4.1. Available evidence

4.1.1. Research and statistical analyses indicate that there are certain categories of drivers who constitute the core of the drinking and driving problem. They are typically:

- relatively young and inexperienced male drivers,

- drivers who habitually drive with very high BAC levels, and,

- occasional drinkers with low BAC levels (sometimes even below the maximum allowed limit) who are relatively less tolerant to alcohol.

4.1.2. It is difficult to assess the effect of a reduction in any BAC limit because the specific effect is often, but not always, obscured by simultaneous changes in enforcement, penalties and publicity. However, there is ample evidence from Australia, Japan, the United States of America and some EU Member States that, generally, reductions in a BAC limit supported by effective enforcement, penalties and publicity, can significantly reduce drinking and driving, even among high risk offenders. However, the role of rehabilitation is particularly significant for tackling high risk offenders. There is also some evidence to show that reductions in BAC levels alone, without any supporting enforcement and publicity, can also significantly reduce drinking and driving, but this probably implies an unrealistic degree of compliance with the law. A summary of some of the evidence is set out in Annex 1. On balance, the available evidence indicates that a BAC reduction to 0,5 mg/ml for those Member States still at 0,8 mg/ml could not only reduce casualties among young male drinkers, but also among the hard core of problem drivers who habitually drive after drinking excessively, as long as the reduction is also accompanied by effective targeted and random enforcement.

4.2. The need to support legal maximum BAC limits with effective enforcement

4.2.1. Only four Member States (United Kingdom, Ireland, Italy and Luxembourg) still retain a maximum legal BAC limit of 0,8 mg/ml. In the case of the United Kingdom this legal limit has remained unchanged since it was first set in 1967. The very significant reductions in inappropriate drinking and driving in the United Kingdom in the last two decades, where fatal accidents have been reduced by 70 %, have been achieved against a background of unchanged BAC legislation but effective enforcement and publicity campaigns which have modified drinking and driving behaviour. The Commission believes that even in Member States such as the United Kingdom, which have been very successful in reducing fatal drink-drive accidents, further reductions are possible by moving towards a lower maximum legal limit of 0,5 mg/ml and extending other measures. Furthermore, significant reductions are also possible in those Member States who have already adopted aa legal maximum limit of 0,5 mg/ml, of lower, by expanding enforcement activity and adopting more rigorous penalties for drivers convicted of inappropriate drinking and driving. In particular, the Commission endorses the view that extensive and regular random breath testing is a very cost effective deterrent to combat drinking and driving.


5.1. Inexperienced drivers

5.1.1. It is clear that inexperienced (mainly young) drivers are disproportionately involved in drink-drive accidents. Since the risk of involvement in a drink-drive accident by inexperienced drivers is relatively high compared to more experienced drivers, the Commission believes that a lower BAC limit of 0,2 mg/ml is appropriate while drivers are learning to drive, or have only been in possession of a full driving licence for less than two years, and recommends that Member States adopt this lower limit for this vulnerable group of drivers. Some States may wish to impose a longer period of possession of the full driving licence before permitting BAC limits greater than 0,2 mg/ml.

5.2. Riders of two-wheel motor vehicles

5.2.1. Riders of two-wheel motor vehicles have a risk of being killed or seriously injured in a road accident which statistically is many times higher than for a car driver. Riders, especially young inexperienced riders, who drink and drive could be particularly at risk from alcohol impairment. Although available statistics at a European level are far from conclusive, some national fatality statistics indicate that riders of two-wheel motor vehicles do not mix drinking and driving to the same extent as car drivers, and that they already adopt a more responsible attitude to their greater risk of accident involvement. However, because in some States very young riders (as young as 14) are allowed to drive low-powered two-wheel motor vehicles, and the age profile of riders is relatively low in the EU, a lower limit would seem appropriate to reinforce good riding habits instilled during training. The Commission believes that the relatively high risk of road accident involvements by this road user group should be reflected by a lower maximum BAC limit of 0,2 mg/ml for riders, and recommends that Member States adopt this lower limit.

5.3. Drivers of goods vehicles (up to 3,5 tonnes GVW) and buses (more than eight passenger seats)

5.3.1. Available statistics indicate that the risk of a large bus being involved in a fatal or serious injury road accident is greater than for a car, per vehicle kilometre, reflecting the much greater passenger occupancy of buses and their operational characteristics. For a large goods vehicle the risk is about the same as for a car, per vehicle kilometre. These relative risks also apply for the total casualty risks in accidents involving these vehicles. However, because buses carry many more passengers than cars their associated occupant casualty risk per occupant kilometre is very much lower. Nevertheless EU statistics show that fatal accidents involving large goods vehicles and buses account for about 18 % of all fatal accidents in the EU.

5.3.2. Although road accident statistics are not yet comprehensive enough to make a detailed evaluation of relative risk across the EU, there are factors other than risk which also need to be considered in relation to determining appropriate BAC limits for drivers of large vehicles. Once involved in a road accident, large vehicles have the potential to cause severe property damage, disruption, delay, and traffic congestion especially in tunnels, on bridges, on main arterial roads, or in densely populated urban areas. The Commission believes that, considering the growth in Community freight and bus passenger traffic during the past decade, a uniform maximum BAC limit of 0,2 mg/ml is appropriate for all drivers of large goods vehicles and buses and recommends that Member States should adopt such a lower limit. Indeed, the Commission notes that some operators already adopt a no alcohol policy for their professional drivers.

5.4. Drivers of vehicles carrying dangerous goods

5.4.1. There is no evidence to suggest that vehicles carrying dangerous goods are any more likely to be involved in a road accident than vehicles not carrying dangerous goods, or cars. However, accidents involving vehicles carrying dangerous goods have the potential to be very serious. The Commission believes that a lower maximum limit of 0,2 mg/ml is appropriate for drivers of vehicles carrying dangerous goods and recommends that such a limit shoud be adopted. Most vehicles carrying dangerous goods will be greater than 3,5 tonnes GVW.


6.1. The available evidence

6.1.1. Estimation of the magnitude of the fatality reduction effect of reductions in BAC limits, supported by national enforcement and publicity, is an extremely complex task. Available estimates from Europe, the United States of America, Japan and Australia of the degree of accident and casualty reduction from lowering BAC limits, combined with parallel measures to reduce inappropriate drinking and driving, vary widely, but they indicate that such action is very effective. The effects can be cumulative, but will be heavily influenced by changes in driver attitudes and behaviour. The available evidence (summarised in Annex 1) suggests that a concerted effort to reduce inappropriate drinking and driving in the EU, based around a regime of uniform EU-wide BAC limits supported by national enforcement and publicity, could realistically reduce fatalities in road accidents involving inappropriate drinking and driving by about 10 %. This is not a scientifically based calculation but represents a reasonable minimum target based upon international evidence.

6.1.2. It should be recognised that the relationship between the maximum legal BAC limit and the incidence of inappropriate drinking and driving is not clear because many other factors influence drinking and driving behaviour. The proportion of drivers killed who were found to be over their relevant BAC limit varies widely between Member States, although the Commission estimates it to be approximately 25 % for the EU as a whole. The proportion of killed drivers found to be over the national BAC limit is reported to be higher in North America where the legal BAC limit varies between 1,0 and 0,8 mg/ml, and much lower in Japan where the limit has been fixed at 0,5 mg/ml since 1970. On the other hand the United Kingdom has significantly reduced inappropriate drinking and driving over the last twenty years by a mixture of enforcement and penalties linked to regular publicity campaigns, all coordinated around a limit of 0,8 mg/ml which has not been reduced since it was first set in 1967.

6.1.3. Different approaches have been used to reduce inappropriate drinking and driving in response to national circumstances with different degrees of success. The Commission takes the view that the appropriate mix of the many measures available to reduce inappropriate drinking and driving should be decided by Member States, but that in the context of a common transport policy these are best arranged around a more uniform regime of BAC limits.

6.2. The Distribution of fatality reduction effects

6.2.1. A 10 % reduction in all EU fatalities in accidents involving inappropriate drinking and driving would save some 1000 lives annually. This represents 2,5 % of the total fatality toll in road accidents in the EU. Available evidence indicates that about two thirds of the lives saved annually would be drivers over the relevant BAC limit and the remaining third would be sober drivers and vehicle occupants and other road users. However, the proportions of fatalities saved will vary between Member States. Greater reductions are possible but depend mostly upon the level of enforcement, especially in those Member States where current levels of enforcement are relatively low. In particular, the use of extensive breath-testing strategies, in which drivers will perceive that they have a significant chance of being tested, are a cost-effective deterrent despite the concomitant requirement to increase funding for this activity.

6.2.2. The recommended general limit of 0,5 mg/ml, with 0,2 mg/ml for certain categories of driver, will affect some Member States more than others because 10 Member States already arrange their national drink-drive policies around a BAC maximum limit of 0,5 mg/ml. Sweden adopted a lower BAC maximum limit of 0,2 mg/ml in 1990 and two Member States already invoke lower limits for special categories of driver. Consequently accident and casualty reduction effects arising from reduced maximum BAC limits will vary between Member States, but they will be mostly determined by the effectiveness of national enforcement and driver compliance.


7.1. The "economic problem" in remote rural areas

7.1.1. For those Member States who still have a maximum BAC limit of 0,8 mg/ml (particularly the United Kingdom and Ireland), it is argued that a reduction to 0,5 mg/ml could, under certain assumptions concerning drinking behaviour and driving practices, lead to a reduction in alcohol sales and economic loss. This could arise for retailers (pubs) in remote rural areas to which the only means of transport is by private car, and where drivers reduce either their consumption of alcohol, or the frequency of pub visits, or both.

7.1.2. The Commission view is that the social value of accident reduction from lower drink-drive limits throughout the Community, backed up by effective national enforcement, training, and publicity, is more than likely to outweigh possible economic loss in some regions. There is no reason why driving habits, as well as drinking habits, cannot change to adapt to the need to separate driving from drinking above BAC levels proven to be detrimental to driving ability.

7.1.3. It should be noted that the Commission does not recommend a total ban on alcohol for drivers. In addition, the use of public transport, car-sharing, nominating a non-drinking driver, moderating drinking behaviour or switching from alcoholic to non-alcoholic drinks, are all available options to maintain social drinking. Inevitably there is a conflict of interest here, but it should be remembered that every road death has an economic cost of EUR 1 million and that failure to reduce inappropriate drinking and driving has a real economic impact too. The purpose of the recommendation is to improve road safety and not to support the alcohol market in some regions.

7.2. The need for additional resources to enforce lower BAC limits

7.2.1. Any introduction of lower maximum BAC limits will, depending on changes in drinking and driving behaviour, increase the burden upon police enforcement officers for a given level of enforcement resources. The national enforcement effort is critical to force down the number of accidents due to inappropriate drinking and driving. The recent report from the European Working Group on Alcohol, Drugs and Medicines highlighted the importance of enforcement. Currently the level of enforcement throughout the Community varies significantly. The adoption of lower drink-drive BAC limits will inevitably mean that more resources will be required for effective enforcement and deterrence, especially in Member States where enforcement levels are relatively low. Without a meaningful increase in police resources for enforcement activity, lower drink drive limits could be counter productive because limited police resources could be taken up with processing more drivers detected with relatively low BAC levels and reduce the detection rate of more serious offenders more likely to cause serious accidents.

7.2.2. Sustained enforcement is essential to generate an effective deterrent effect, but enforcement programmes such as random breath testing are expensive and need resourcing. However, it should be noted that random breath testing is very cost-effective and is clearly worth investing in. Even so, it is not unreasonable to expect that the cost of increased enforcement necessary to bring about a reduction of 1000 fatalities a year in the EU, within a regime of more harmonised BAC limits, will be less than the conservative EUR 1 billion a year valuation of the expected fatalities saved. In addition, it should be recognised that increased liaison between national police forces offers the scope for improving the effectiveness of enforcement within existing enforcement resources.


8.1. At the national level

8.1.1. The report of the European Working Group on Alcohol, Drugs and Medicines highlights the shortcomings of current statistics concerning alcohol-related accidents. Systematic alcohol-testing of drivers following a road accident, and the recording of the results for statistical analysis, is essential to be able to monitor the effectiveness of measures to reduce inappropriate drinking and driving. In some Member States, even though police enforcement is expanding, the results of alcohol tests are not fully recorded in national road accident databases. Complete aclohol test (breath and blood) results would provide the means to compare the incidence of inappropriate drinking and driving between different groups of drivers.

8.2. At EU level

8.2.1. If all such test results were also provided to the European Community Road Accident Database (CARE)(10) the sample size would be significantly increased for analysis of BAC distributions (and breath equivalents) of different road user groups at a European level. This would provide a unique European data source for informing and monitoring the effects of drinking and driving policies. There is substantial analytical added value to be derived from the pooling of breath and blood alcohol test level results. In addition, within the community action programme(11) on injury prevention in the field of public health, the epidemiological monitoring of injuries by means of a community system for collection of data and information on injuries is being promoted (Euphin: European Union Public Health Information Network). Core injury indicators based upon mutually accepted standards and providing nationally representative information including indicators from other relevant policy areas will be provided (e.g. safety at work, transport and civil protection).


9.1. The need for mandatory EU standards

9.1.1. The European Working Group on Alcohol, Drugs and Medicines reported broad support from Member States for the setting of mandatory EU measurement standards for evidential breath testing equipment the results of which can be used as evidence in a court of law.

9.1.2. The ability to supplement blood test results with those taken from breath testing machines is an important means of expanding police enforcement. Member States with a maximum BAC limit of 0,5 mg/ml use a breath equivalent measurement (BRAC) of either 0,22 or 0,25 microgrammes of alcohol per millilitre of breath. But the various breath-testing devices currently on the market operate to varying degrees of accuracy. This means that Member States have to allow for measurement error in both their BAC and BRAC equivalent levels for tested drivers. The allowance for measurement error varies between Member States.

9.1.3. In the context of harmonised maximum BAC limits in the EU the recommended BAC limits of 0,5 mg/ml and 0,2 mg/ml do not include any allowance for measurement error. Setting harmonised measurement standards for testing equipement is a simpler way to reinforce harmonised BAC and BRAC limits based upon the technical capability of approved testing equipment. The allowable statistical limits for measurement error will be more closely harmonised by the development of standards for testing equipment.

9.1.4. Some Member States also include a generous tolerance in the measurement of the BAC level to encourage apprehended drivers to opt for a breath test rather than a more expensive and time consuming blood test. In Sweden, even though a relatively low BAC level of 0,2 mg/ml has been operational since 1990, the tolerance level allowable for test results can allow drivers to drink up to 0,34 mg/ml in some circumstances. The purpose of this increased tolerance level is to increase the level of enforcement, since many more evidential breath tests than blood tests can be carried out for a given level of police resources. Such tolerance levels inevitably need to be determined by Member States who have to balance the deterrent effect of the legal BAC limit against the deterrent effect of a slightly higher but allowable BAC level at higher levels of enforcement.

9.1.5. A draft proposal for a Measuring Instruments Directive is currently being developed by the Commission which, if adopted, will harmonise the requirement for the measurement accuracy of evidential breath-testing equipment in the EU. Guidelines for international standards are set by the Organisation Internationale Métrologique Légale (OIML) and member countries base their legislation on these standards.


10.1. The problem of illegal drugs

10.1.1. Recent research into the effects of marijuana on driving performance(12) concluded that the driver impairment effects of using cannabis alone, which produced doses of tetrahydrocannabinol (THC) of 100 to 200 ug/kg, were not very different from the impairment effects attributable to other drugs, or BAC levels above 0,8 mg/ml. However, it was found that the same doses of THC when combined with alcohol sufficient to produce a BAC of 0,4 mg/ml, produced severe driver impairment effects.

10.1.2. This evidence of high risk for drivers and riders, and the increasing use of cannabis and detection of THC (as well as other illegal drugs) in drivers involved in accidents(13), is of growing concern for road safety in the EU. However, there are many illegal drugs, and it is unclear how they all interact with alcohol to impair driving ability. Furthermore, the EU action plan on drugs 2000 to 2004, endorsed by the European Council of Feira, asks the Commission and Member States to undertake research into the effects of driving under the influence of illicit drugs and pharmaceuticals. The possession and use of all illegal drugs is, by definition, an offence punishable within the context of national legislation. Since possessing or using an illegal drug is a punishable offence anyway, the question of whether driving whilst under the influence of illegal drugs should be treated as a more serious offence than drink driving needs to be considered by Member States, and consequently this aspect is not addressed in this recommendation.

10.2. The problem of legal drugs

10.2.1. The use of legal drugs (medicines), by themselves or in combination with alcohol, and their driver impairment effect, is also a very important issue in the context of increased accident risk. But there are a huge number of legal drugs, and their driver impairment effects are by no means as comprehensively researched as the effects of alcohol have been. For example, Benzodiazepines are a widely used group of medicine which, in combination with alcohol, may be at least as driver-impairing as low-dose combinations of alcohol and cannabis.

10.2.2. The whole issue of the effects of legal drugs on drivers is, in itself, a very complex topic, of which the specific aspect of alcohol combined with legal drugs is just one part. However, Member States may wish to consider how to deal with this issue in their national legislation. At a European level, there will be obvious benefits to be derived from an exchange of information concerning research results and best practice.

(1) Promoting road safety in the EU: the programme for 1997 to 2001 (COM(97) 131 final).

(2) Priorities in EU road safety: progress report and ranking of actions (COM(2000) 125).

(3) A strategic road safety plan for the European Union: ETSC February 1997.

(4) Report of the Working Group on Alcohol, Drugs and Medicines for the High Level Group. October 1999.

(5) NHTSA reporter Vol. 10, No 2: Spring 1999, No 3: Summer 1999.

(6) COM(88) 707 final.

(7) COM(89) 640 final.

(8) Reducing traffic injuries resulting from alcohol impairment: ETSC: January 1995. Report of the Working Group on Alcohol, Drugs and Medicines for the High Level Group. October 1999.

(9) Sartre 95 and 98 (Social attitudes to road traffic risk in Europe).

(10) Council Decision 93/704/EC.

(11) Decision 372/99/EC of the European Parliament and the Council of 8 February 1999 adopting a programme of Community action on injury prevention in the framework for action in the field of public health (1999 to 2003).

(12) Marijuana, alcohol and actual driving performance NHTSA July 1999.

(13) Literature review on the relation between drug use, impaired driving and traffic accidents: Section 2.4.2: European Monitoring Centre for Drugs and Drug Addiction: February 1999.