Schaffer Online Library of Drug Policy Sign the Resolution
Contents | Feedback | Search
DRCNet Home
| Join DRCNet
DRCNet Library | Schaffer Library | Drugs and Driving 

 


Drinking and Driving in France : Recent Legal Changes and Main Trends During the Years 1988-1993

Biecheler-Fretel M.B, Fontaine H.

INRETS - DERA, 2 Avenue du General Malleret-Joinville, 94114 Arcueil Cedex, France

INTRODUCTION

An analysis of the historical "World-wide Decline in Drinking and Driving" has been presented at the Traffic Safety International Conference in the Hague, on September 1993 and the trends and patterns analysed country by country. The statistical results about drinking and driving in France over the period 1983-1991 have been summarised (Biecheler-Fretel and Jayet, 1994). What has happened since? In this paper we examine some aspects of the evolution in France on the more recent years : legal changes, roadside breath data and road accident data. We will insist on the way we establish the statistics about road alcohol related accidents. This knowledge is imperatively needed for comparing results from one country to another and over the time in a same country, as this was strongly underlined in the Hague by Herb Simpson, the ICADTS' president (TRB,1994).

RECENT LEGAL CHANGES (1990-1994)

The measures that occured since 1990 show that the battle against drinking and driving remains always in France a particular focus of the legislative activity. In 1994 France made a step in the direction of the proposal of the European Commission to harmonize the maximum blood alcohol content to 0,5 g/l (CEMT, 1988,1993).

ROADSIDE BREATH TEST DATA (1988-1993)

There are two statistical sources of information on drink driving : statistics as part of police enforcement or statistics from roadside surveys conducted for scientific purposes.

Road Enforcement Results

Random tests are carried out on drivers, even when no accident or offence has taken place, on the order of the public prosecutor (since 1978) or on initiative of police officers (since 1990).

Table 1 presents the results of tests carried out by the national gendarmerie (covering rural areas and urban areas under 10 000 inhabitants) and municipal police (covering urban areas over 10 000 inhabitants). The spectacular increase in the number of random tests (from 2,7 millions in 1990 to 6 millions in 1994) reflects the stronger and stronger mobilisation against drink-driving activity, on rural roads especially. The percentages of tests measuring over the legal limit are stable during the period 1988-1994 and particularly over the three last years : about 1,1% on rural roads (gendarmerie's control) and 0,8% in urban areas (police's control).

Table 1
Roadside Random Breath Tests (1988-1994)

Year Gendarmerie
mainly rural areas
Police
urban areas
Drivers tested (1000s) % Over the legal limit Drivers tested (1000s) % Over the legal limit
1988 1 254 1,11% 359 0,97%
1989 1 627 1,00% 577 0,90%
1990 2 071 1,20% 706 0,81%
1991 2 949 1,15% 758 0,71%
1992 3 765 1,14% 713 0,78%
1993 4 649 1,02% 768 0,77%
1994 5 256 1,07% 769 0,76%

Table 2 present the results of tests following motoring offences. Under the current legislation breath tests are compulsory in case of suspected violation : dangerous overtaking or territorial violations (priority, stop, red light, wrong way..) and also lack of insurance or driving license, refusal to comply with directives. Since 1990 illegal drinking appears to be largely more often related to other road violations on urban network (4%) than on rural roads (1%).

Table 2
Roadside Screening Breath Tests after Motoring Offences (1988-1993)

Year Gendarmerie
mainly rural areas
Police
urban areas
Drivers tested (1000s) % Over the legal limit Drivers tested (1000s) % Over the legal limit
1988 936 2% 117 3%
1989 1 011 2% 140 4%
1990 1 305 1% 162 3%
1991 1 931 1% 160 3%
1992 1 363 1% 155 4%
1993 1 275 1% 162 4%

Roadside Inrets Surveys

National surveys were carried out in France in 1970 and 1977 and 1979 in the context of epidemiological studies (Biecheler-Fretel et al, 1983). Then, starting in the 1980s regional surveys were carried out for scientific purposes (Biecheler-Fretel, Filou, 1993 ; Filou, 1994) The ratio "% 0.25 - 0.40 mg/l / % over 0.40 mg/l ", which compares the magnitude of pre-illegal drink driving versus illegal driving, is greater in 1991/92 (about 1.04) than in 1985 (about 0.95) That means clearly the growing importance of pre-illegal BrAC in the incidence of drink driving. A further examination of trends according to age confirms also the diminution of incidence of pre-illegal and illegal levels among young drivers (<25 years) and its increase among older drivers (> 55 years).

ROAD ACCIDENT DATA (1988-1993)

Since 1970 alcohol tests are compulsory in cases of accidents involving injuries, but assessment of the alcohol level is a complex process (Got, 1989). The blood or breath alcohol test carried out theoretically give an information about the alcohol level ; but in a number of cases no such test is carried out. Whether or not a driver is tested depends on the state of the driver (uninjured, injured or killed) and consequently on the accident severity (fatal or injury). The quality of information is also linked to technical aspects of breath testing practice (breath screening test or/and blood sample, delay between the accident and the blood sample, difficult linkage of the driver's alcohol level with his accident report). The official statistics from national police reports underestimate the phenomena. Proper measurement requires more detailed data that allow to know the rate of unknown data as well as the adjustments required to estimate the incidence of alcohol in road accidents.

The INRETS report file 1/50 has been used to fulfill this aim. Since 1987, INRETS gathers at random one out of fifty national police reports each year, in order to get a file representative of accidents that occur in France. Some 3000 police reports related to injury accidents are thus selected every year (Fontaine et al, 1994). Each accident report give information on screening and evidential proof (with a blood sample a or breath test). Both breath tests given by the police officers at the time of the accident and tests of the blood alcohol levels are incomplete.

Screening Breath Test Data

Drivers not able to submit to a breathalyser or refusing it are not breath tested. Table 3 shows the percentages of drivers and riders breath tested at the time of the accident (injury or fatal) from 1988 to 1993. Over the period screening breath tests are practised in about 80% of the cases for injury accidents(79% in 1993) and around 45% for fatal accidents (42% in 1993).

Table 3
Screening Breath Test Data (1988-1993)

Year Injury accidents Fatal accidents
Tested Test Impossible (or refused) Not tested (or no information) Tested Test Impossible (or refused) Not tested (or no information)
1988 81% 6% 13% 49% 46% 4%
1989 80% 7% 13% 47% 45% 8%
1990 81% 7% 12% 47% 45% 8%
1991 80% 7% 13% 46% 43% 11%
1992 80% 8% 12% 48% 43% 9%
1993 79% 7% 14% 42% 48% 11%

Alcohol Levels Data

The results about alcohol levels (over or under the legal limit) are provided as following (Fontaine and Gourlet, 1994) :

It must be added that not all the results are available (connection between the alcohol level and the report). Table 4 shows both for injury (including fatal) and only fatal accidents, from 1988 to 1993, the percentages of drivers and riders tested and whose the alcohol level (under or over the legal limit) is available in the police report (known result). The knowledge on alcohol levels is improving for fatal accidents (from 66% known in 1988 to 76% in 1993). Considering the last year available, 1993, in 14% of the cases for injury accidents (respectively 24% of the cases for fatal accidents) the alcohol level of the driver remains unknown. These rates seem rather low in comparison with other European countries (Biecheler-Fretel, 1995). They nevertheless indicate that hypothesis and adjustments are required to produce overall estimates. Table 5 gives the percentages of the drivers over the legal limit : it shows the discrepancy between the two percentages where the denominator is either the drivers tested with a known result or the drivers involved. The error proves to be greater for fatal accidents.

Table 4
Percentage of Drivers Tested (blood or breath) with Known Results

Year Injury accidents Fatal accidents
1988 84% 66%
1989 84% 66%
1990 84% 70%
1991 83% 73%
1992 85% 78%
1993 86% 76%

Table 5
Percentages of Results Over the Legal Limit

Year Injury accidents
total positive / total known
(total positive/total involved)
Fatal accidents
total positive/ total known
(total positive/ total involved)
1988 5% (4%) 17% (11%)
1989 5% (4%) 15% (10%)
1990 6% (5%) 15% (11%)
1991 7% (6%) 15% (11%)
1992 8% (7%) 22% (17%)
1993 7% (6%) 18% (14%)
positive = over 0.4 mg/l (in the breath) or 0.8 g/l (in the blood)

Number of Alcohol Related Accidents (1988-1993)

An alcohol related accident is defined as an accident where at least one of the driver or rider involved has a alcohol level over the legal limit.

Using the above statistical information on alcohol levels of the drivers involved in accidents it has been possible to provide estimates of the number of alcohol related accidents in France as shown on Table 6. The numbers of injury and fatal accident has been steadily dropping since 1988 but the percentages of alcohol related accidents have not changed a great deal : around 16% of injury accidents and 36% of fatal accidents are alcohol related.

Table 6
Percentage of Alcohol Related Accidents (1988-1993)

Year Number of injury accidents
(and % of alcohol related)
Number of fatal accidents
(and % of alcohol related)
1988 175 887 (12%) 9343 (36%)
1989 170 590 (11%) 9302 (31%)
1990 162 573 (13%) 9128 (30%)
1991 148 890 (17%) 8509 (31%)
1992 143 362 (17%) 8114 (39%)
1993 137 500 (16%) 8005 (36%)

CONCLUSION

Since the beginning of the 1990s the trends related to drinking and driving prove to be in France in continuity with those observed during the 1980s. The spectacular increase in the number of random tests, rising up to 6 millions in 1994, reflects the stronger and stronger mobilisation against drink-driving activity, on rural roads especially. This maintain stable the incidence of alcohol use among drivers as shown by the results of enforcement : about 1,1% on rural roads and 0,8% in urban areas. The growing importance of pre-illegal BrAC (between 0.50 and 0.8) among drinking drivers is confirmed by INRETS roadside surveys.

In 1993, in 14% of the cases for injury accidents (respectively 24% of the cases for fatal accidents) the alcohol level of the driver remains unknown. The knowledge on alcohol levels is improving for fatal accidents. The numbers of injury and fatal accident have been steadily dropping since 1988 but the percentages of alcohol related accidents have not changed a great deal : around 16% of injury accidents and 36% of fatal accidents are alcohol related. Thus, as already stated in other European countries (Great Britain, Germany, Netherlands...) the general decrease in accident trends may be attributed to a general safety effect ; there is no specific overall effect of drinking and driving enforcement. It is too early to assess the effect of lowering the legal limit to 0.7, but, as we know that at a general level the most influencing factor if a driver abstains from alcohol or not is the limit in the country, probably a significant further reduction in drink driving in France can now be obtained by the significant lowering of the legal limit to 0.5 g/l.

REFERENCES

Biecheler-Fretel, M.B et al (1983) Alcohol , Driving and Safety. Summary and Analysis of Research done by the French National Road Safety Agency (ONSER), English translation of the ONSER report n° 65.

Biecheler-Fretel, M.B. and Filou, C. (1993). Drinking and Driving in France during the years 1983-91 : changes in legal framework and analysis of road users' behaviour. In "Alcohol, Drugs and Traffic Safety-T92" (Band 2) edited by Utzelmann , Berghaus and Kroj , Verlag TÛV Rheinland, Cologne 1993 (p 1017-1035).

Biecheler-Fretel, M.B, Jayet M.Ch (1994). Alcohol and driving in France : why we have seen changes and what is next? In The nature and reasons for the worldwide decline in drinking and driving, TRB Circular n° 422, TRB Washington (p 62-64).

Biecheler-Fretel, M.B (1995). The incidence of drink-driving in EU countries. In Reducing Alcohol-Related Crashes, the Role of The European Union.; ETSC Conference, 29th march 1995 (proceedings to be published).

CEMT (1988). Proposal for a council directive relating to the maximum permitted blood alcohol concentration for vehicles drivers COM(88) 707 final.

CEMT (1993). Resolution n°93/5 on Drink as a Factor in Road Accidents.

Filou, C. (1994). Drinking and Driving in France during the years 1980-93 : changes in legal framework and analysis of road users' behaviour. ICAA/CIPAT International Conference, Prague, June 5-10 ,1994.

Fontaine, H., Gourlet, Y. (1994). Sécurité des véhicules et de leurs conducteurs. Rapport INRETS n°175.

Fontaine, H., Gourlet, Y. (1994). Mobilité et accidents. Rapport de Convention INRETS/DSCR, INRETS -DERA Arcueil.

Got, C. (1989). Alcool et accidents mortels de la circulation. In actes du Séminaire International de Hambourg , juin 1988 : la sécurité routière, avant tout une question de responsabilité, CEMT, 1989.

TRB (1994). The Nature of and the Reasons for The Worldwide Decline in Drinking and Driving. Circular 422. Washington : Transport Research Board.