Forensic Science International, 45 (19901273- 280 Elsevier Scientific Publishers Ireland Ltd.
TRAFFIC ACCIDENTS INFLUENCE
AND
273
DRIVERS
LARS QUIST CHRISTENSEN, LISBETH MARIA NIELSEN Bispebjerg Hospital Anuesthesiologic Denmark, Copenhagen Denmarkr
department,
SUSPECTED
FOR
DRUG
NIELSEN and SOREN LOUMANN
Pain clinic and The Medicolegal Council of
(Received December 21st, 19891 (Accepted January lOth, 1990)
Summary All records from the Danish Medicolegal Council concerning drivers suspected for drug influence were examined for the 5 year period 1981- 1985. 461 records were included, 62 women and 399 men. In 250 cases drugs from more than one of ten groups had been taken thus making 786 combinations of drug/driving. The major drug group was benzodiaxepines, accounting for 65% of all drug intake. Opioids also contributed substantially, found in 38ob of the cases. A traffic accident had occurred in 180 (39ob) of the records. Drivers who had been taking antidepressives were involved in an accident in 67%, significantly above the mean. For benzodiazepines. the corresponding percentage was 43%. while for opioids it was only 23%, significantly below the mean. This striking difference has been demonstrated in most of the studies concerning drugs in traffic. It may support the hypothesis that opioids do not necessarily make driving dangerous, as do antidepressives, barbiturates and especially benxodiazepines. Key words: Traffic accidents: Drug influence.
Introduction
Different methods have been used to clarify the deleterious effects of drugs on driving behavior [l]. Laboratory tests mainly evaluating psychomotor performance, simulated car driving and real driving are the commonly used methods which offer controlled circumstances. Another approach is epidemiological studies, mostly based on two different principles: 1. 2.
Occurrence of traffic crashes in a grouo known to be taking a specific drug compared with a group without that drug intake. Evaluation of the drug intake in a group with abnormal traffic performance, e.g. crashes, compared with the drug intake among ‘non-crashers’.
The present
study is based on the first principle. We here present
all
Correspondence to: Lars Quist Christensen, Godthibsvej 1023.th 2000 Frederiksberg, Denmark. 0379-0738/9O/SO3.50
0 1990 Elsevier Scientific Publishers Ireland Ltd. Printed and Published in Ireland
274
cases evaluated by the Medicolegal Council of Denmark in the period 19811985 dealing with traffic offenders suspected of drug influence. Materials
All records in the file concerning drivers suspected of driving under the influence of drugs were examined for the 5-year period 1981-1985. 476 cases were found. The cases had been requested by the police authorities, when suspicion was raised that a driver was influenced by drugs because of traffic crashes W = 1841, otherwise abnormal traffic behavior or denunciation W = 2921. This method implies highly selected material. The Medicolegal Council receives a form from the Police authorities, principally including information about self-reported drug intake, a description of the incident and the result of a clinical examination accompanied by a blood and/or urine sample. Antidepressants, CNS stimulating agents, cannabinoids and opioids were often detected in urine whereas other drugs were found mainly in blood. Based on the results of the clinical examination and from the biological specimens the Medicolegal Council made an estimation whether the driver was supposed to have been influenced by drugs. In some instances sufficient data to make this estimation was not available. Methods
From the records the following information was collected: Year of incident, age, sex, drugs taken, blood alcohol concentration (BAC), accident/no accident and the estimate concerning the degree to which the driver was influenced by drugs. 100 different legal and illegal drugs were detected. These were classified into the following ten groups: barbiturates (BB) benzodiazepines (BZ), other sedatives (AS), cyclic antidepressives (CA), antiepileptics (EP), neuroleptics (major tranquilizers) (NL) opioids including heroin (OP), CNS stimulants (ST), cannabinoids (TC) and the remaining drugs (AN). Results are expressed as mean r S.D. Statistical analyses were performed with Chi-square test, Wilcoxon test, Fishers test and Students t-test for unpaired data. Results
Records (4761 were registered out of which 15 were excluded, 13 with BAC at or above 0.8 g/l (legal limit in Denmark) and 2 without specimen or information on drugs taken available. The sex distribution in the remaining 461 cases was 62 females and 399 males. Age was stated in 448 cases and the distribution for each sex is shown in Fig. 1. Females were 39 f 10 years of age and males 32 -+ 11 years (P < 0.0011.The mean age for all drivers was 33 f 11 years.
275
Frequency
pJ
cl
Females
Males
O
I
15
20
30
40
50
60
70 Years
of
I-
age
Fig. 1. The relative age distribution for male Vv = 38’7) and female W = 61) drivers to be influenced by drugs. P = 0.000012 was obtained by Students t-test.
suspected
Figure 2 shows the distribution of age for drivers involved in a crash W = 173) and for those not UV = 275). Mean age of the drivers with a crash was 36 -t 12 years, significantly higher than for the rest with 31 + 11 years (P < 0.001). The drug group by far the most frequent represented was BZ, taken by 298 (65%) of 461 drivers. Next was OP found in 173 (38%) cases. The two minor groups were CA and EP by 12 (2.6%) each. Numbers are seen in Table 1. In Fig. 3 the relative frequencies of drug groups among females and males are shown. Women had been taking drugs from the groups AN and and P = 0.057 respectively. For CA more often than expected, P = 0.0092 two other groups men were represented in surplus, OP (P = 0.037) and TC
Frequency 35% 1
Drivers
60
from
crash
70 Years
Fig. 2. The relative age distribution for drivers involved in an accident involved in an accident (N = 275). P = 0.000012 by Students l-test .
of age
W = 173). and those
not
276 TABLE
1
NUMBER OF TIMES OTHER DRIVERS.
DRUGS
WERE
AN
AS
BB
BZ
Accident No accident
14 24
15 22
34 33
Total
38
37
67
DETECTED
AMONG
ACCIDENT
DRIVERS
AND
CA
EP
NL
OP
ST
TC
Total
128 170
8 4
5 7
15 17
40 133
9 19
21 50
289 479
298
12
12
32
173
28
71
768
P = 0.0161. In 250 (54%) of the cases drugs from two or more groups were found, thus making the total number of ‘a drug group found’ 768 and total percentage in Fig. 3 above 100%. There was no significant difference between males and females in this respect. A traffic accident preceded 180 (39%) of the cases, without differences between sex. The relative frequency of crash for each drug group is shown in Fig. 4, the absolute numbers in Table 1. Of those cases with CA, 67% were preceded by an accident (P = 0.0381 and with BB 51% (P = 0.027). BZ was only marginally above mean with 43% (P = 0.058). Among the cases with OP only 23% (P < 0.00011 had crashed. The remaining groups did not differ significantly. Two or more drugs found was not correlated with a higher incidence of accident compared to one drug found. Among the drivers who could be evaluated 72% were actually found to be influenced by drugs. This percentage for each group is seen in Fig. 5. For BB it was 91% Cp = 0.00531 and for TC 85% Cp = 0.0581. Below mean were cases with CA having 55% (P = 0.0491 estimated as influenced. Frequency 8O%j
Opioids*
Antidepressives
Barbiturates
Benzodiazepines Other
Antiepileptics sedatives
Neuroleptics
Other
drugs*
Cannabinoids* Stimulating
drugs
Fig. 3. The frequency by which each drug group was detected among male and female drivers. An asterisk indicates P < 0.05 for the drug group. Overall Chi square was 27.40, P = 0.0022.
Opioids* Other drugs Antidepressives* Barbiturates* Cannabinoids Antiepileptics Benzodiazepines Stimulating drugs Other sedatives Neuroleptics Fig. 4. The proportion of the drivers that had been taking a specific drug, that were involved in an accident. An asterisk indicates P < 0.05 for the drug group. Overall Chi-square was 32.05, P = 0.00039. Frequency 100% 1
Mea
=7
Other drugs Antidepressives* Opioids Barbiturares* Antiepileptics Cannabinoids Benzodiazepines Stimulating drugs Other sedatives Neuroleptics Fig. 5. The percentage of drivers that had been taking a specific drug, that were estimated influenced by drugs. An asterisk indicates P < 0.05 for the drug group. Overall Chi-square was 21.70, P = 0.017.
278
Discussion
This publication includes all records on cases about suspected drug influence while driving, evaluated by the Danish Medicolegal Council in the period 1981- 1985. As a sample from all drivers influenced by drugs, it is by nature quite biased. A driver will have a record only after several incidents. The police must have become suspicious of a driver influenced by drugs other than alcohol. The driver must then have a clinical examination performed and a biological specimen taken. Only in that case the police may send in a form to the Medicolegal Council, who will make a record. In spite of this strong selection the material seems to be comparable to the background population in Denmark. The mean age of the drivers in this material does not differ from that of killed and injured drivers in Copenhagen in 1984. Also the subgroup of drivers involved in a traffic accident has a mean age and sex distribution similar to drivers causing accidents in Copenhagen in 1984 [2]. A recent Danish paper concerning the frequency of BZ in blood samples submitted for BAC determination at the Institute of Forensic Chemistry in Copenhagen [3] states that 6% of a sample from 26,000 yearly specimens had a BZ concentration >0.2 pmol/l. The age distribution in this material does not differ from that of the BZ group in the present paper. There was 1% with BZ and a BAC < 0.5 g/l. By extrapolation, this is calculated to 1200 drivers with BZ in blood and BAC less than 0.5 g/l, in a 5-year period. The Medicolegal Council had 298 records concerning BZ, i.e. only about 25% of the cases in which the police suspected drug influence, where BAC was below 0.5 g/l and BZ would have been detectable in blood. The authors find it surprising that the drivers involved in an accident were older (36 years) than the rest (31 years, see Fig. 21, but did not find papers confirming this observation. Some of the explanation might be selection by police, and increased (legal) drug intake with increasing age. The major drug group was BZ, taken by 70-80%. secondly OP and then TC. This appearance of drugs does not differ appreciably from the results of other newer investigations [4-g]. Where differences are encountered, it seems to be due mainly to different grouping and analyzing techniques (e.g. BZ vs. sedatives and blood vs. urine). Some authors have found surprisingly many ‘illicit drugs’ (subgroup of OP, ST and TC) [lo - 121. This may be due to cultural differences and differing attendance by the police. Kaempe & Dalgaard analyzed 283 liver specimens from persons killed by traffic 1968 - 1974 [13]. In 20% drugs were detected, mainly BB, but only in one case BZ. To evaluate the different drug groups’ contribution to accidents, the frequency of each group among accident drivers is compared to that of the other drivers, as seen in Fig. 4. Since BZ was the biggest group by far, the result of this group was close to the average of the total study group. Anyway, it was more frequent among crash drivers, though not statistically significant (P = 0.0581. The traffic hazards caused by BZ have been settled
279
in a controlled trial [14] and a number of epidemiological investigations [68,151. CA was by 67% mainly found after an accident. This is in agreement with the findings of other papers [4,8]. The drug group could therefore be assumed to be of qualitative importance, though only sporadically found, 2.6% of the cases in this material. As far as CA is used as a treatment for depressive disease, this probably significantly [l] biases the result. The frequency by which BB was found in this material decreased over time. In Denmark the group had limited indications since 1985, only available for anesthesia and for treatment of epilepsy. However, it was still of some quantitative importance, also found by another investigation [4]. Two studies did not find this coherence between BB and accidents. Ulrich, et al. examined urine from an accident group and blood from a control group. BB was detected with equal frequency in both. An interview inquiry of 14,824 breathalyzed motorists gave the same result [8]. Among accident drivers OP was found significantly less often than expected. This probably has several causes. The police might have knowledge about drug abusers in local area, and thus be more likely to suspect these drivers to be influenced by drugs. Secondly, in steady state the dosage of OP does not always imply a traffic danger, because of tolerance and increased attention to compensate a possible decrease in performance [1618]. In this matter Ulrich [7] and MacPherson [8] had deviating results, but as outlined above, with a weaker design of investigation. Comparing the evaluation of drug influence in the individual groups only two ma]or divergences were found. Drivers taking BB were found to be influenced very often in agreement with the high incidence of crashes. On the contrary, drivers taking CA were found to be influenced less often than they were involved in accidents. This might be explained by their depressive disease state. Conclusion
This material points out a likely traffic danger by drivers taking drugs, mainly BB, BZ and CA, in accordance with several other investigations. In contrast, it seems that drivers under influence of opioids do not contribute excessively to accidents, even though a few papers report different results. It should be stated that all studies except one were epidemiologic. No certain connection between cause and measured effect is therefore unveiled. References 1
S.L. Nielsen, L.Q. Christensen Ugeskr. Laeg., 151 (1989) 1822-
2
Statistical yearbook (1985) p. 150.
and L.M. 1825.
of Copenhagen
1985.
Nielsen,
Analgesics,
Copenhagen
Benzodiazepines
Statistical
Office,
and
traffic.
Copenhagen,
64
280 3 4 5
10
11 12
13 14
15 16 17 18
A. Steentoft, K. Worm and K. Christensen, The incidence of benxodiazepines in Danish road users. Ugeskr. Laeg., 150 (1988) 2383-2386. D. Wilson, Experience with drugs and driving in Queensland, Australia. Med Sci Law, 25 (1985) 2 - 10. A.P. Mason and A.J. McBay, Ethanol, Marijuana and other drug use in 600 drivers killed in single-vehicle crashes in North Carolina, 1978- 1981. J. Forensic Sk., 29 (1984) 987- 1026. R. Honkanen, L. Ertama, M. Linniola et al., Role of drugs in traffic accidents. BT. Med J., 281 (1980) 1309- 1312. L. Uirich, 0. Rudin, A. Amsler and P. Zink, Htiufigkeit von medikamenten im straBenverkehr. Z. Rechtsmed, 93 (1984) 95-110. R.D. MacPherson, J. Per1 and G.A. Stamer, Self-reported drug usage and crash-incidence in breathalyzed drivers. Accid Anal Prev., 16 (19841139-148. K. Wehr and R.D. Maier, Illicit drugs among arrested drivers in the’border region of Aix La Chapelle. In L. Goldberg feds.), Alcohol, Drugs and traffic Safety, Vol. 1, Proceedings, 8th International Conference on Alcohol, Drugs and Traffic Safety, 1980, Wiksell, Stockholm, 1981, pp. 303-314. P. Holmgren, E. Loch and J. Schubert, Drugs in motorists traveling Swedish roads: On-theroad-detection of intoxicated drivers and screening for drugs in these offenders. Forensic Sci Znt.. 27 (1985157-65. A.F. Williams, M.A. Peat, B.S. Crouch, J.K. Wells and B.S. Finkle. Drugs in fatally injured young male drivers. Publ. Health Rep., 100 (19851 19-25. R. Warren, H. Simpson, J. Hilchie et al., Drugs detected in fatally injured drivers in theprovince of Ontario. In L. Goldberg feds.), Alcohol, Drags and traffic Safety, Vol. 1, Pro ceedings, 8th International Conference on Alcohol, Drugs and Traffic Safety, 1980, Wiksell, Stockholm, 1981, pp. 203- 217. B. Kaempe and J.B. Dalgaard, Drugs, alcohol and carbon monoxide in victims of fatal traffic accidents. Ugeskr. Laeg., 141 (19791 1037-1040. J.J. de Gier, B.J. t’Hart, F.A. Nelemans and H. Bergman, Psychomotor performance and real driving performance of outpatient receiving diazepam. Psychopharmacology, 73 (1981) 340-344. J.W.F. Haffner. 0. Bo and K.M. Lunde, Alcohol and drug consumption as causual factors in road traffic accidents in Norway. J. Traf. Med, 2 (1974152-56. D.V. Babst. S. Newman, N. Gordon and A. Warner, Driving records of methadone maintenance patients in New York state. J. Drug Issues, 3 (19731285- 292. R.D. Blomberg, D.F. Preusser, Narcotic use and driving behavior. Accia AnaL Prev., 6 (1974) 23-32. P.R. Dal Monte, G. Formica, G.L. Milandri and M. Ruggieri, Road accidents in young drugaddicts. Comparison with a test group. In L. Goldberg feds.), AlcohoZ Drags and traffic Safety, Vol. 1, Proceedings 8th International Conference on Alcohol, Drugs and Traffic Safety, 1980, Wiksell, Stockholm, 1981, pp. 160- 164.
TRAFFIC ACCIDENTS INFLUENCE
AND
273
DRIVERS
LARS QUIST CHRISTENSEN, LISBETH MARIA NIELSEN Bispebjerg Hospital Anuesthesiologic Denmark, Copenhagen Denmarkr
department,
SUSPECTED
FOR
DRUG
NIELSEN and SOREN LOUMANN
Pain clinic and The Medicolegal Council of
(Received December 21st, 19891 (Accepted January lOth, 1990)
Summary All records from the Danish Medicolegal Council concerning drivers suspected for drug influence were examined for the 5 year period 1981- 1985. 461 records were included, 62 women and 399 men. In 250 cases drugs from more than one of ten groups had been taken thus making 786 combinations of drug/driving. The major drug group was benzodiaxepines, accounting for 65% of all drug intake. Opioids also contributed substantially, found in 38ob of the cases. A traffic accident had occurred in 180 (39ob) of the records. Drivers who had been taking antidepressives were involved in an accident in 67%, significantly above the mean. For benzodiazepines. the corresponding percentage was 43%. while for opioids it was only 23%, significantly below the mean. This striking difference has been demonstrated in most of the studies concerning drugs in traffic. It may support the hypothesis that opioids do not necessarily make driving dangerous, as do antidepressives, barbiturates and especially benxodiazepines. Key words: Traffic accidents: Drug influence.
Introduction
Different methods have been used to clarify the deleterious effects of drugs on driving behavior [l]. Laboratory tests mainly evaluating psychomotor performance, simulated car driving and real driving are the commonly used methods which offer controlled circumstances. Another approach is epidemiological studies, mostly based on two different principles: 1. 2.
Occurrence of traffic crashes in a grouo known to be taking a specific drug compared with a group without that drug intake. Evaluation of the drug intake in a group with abnormal traffic performance, e.g. crashes, compared with the drug intake among ‘non-crashers’.
The present
study is based on the first principle. We here present
all
Correspondence to: Lars Quist Christensen, Godthibsvej 1023.th 2000 Frederiksberg, Denmark. 0379-0738/9O/SO3.50
0 1990 Elsevier Scientific Publishers Ireland Ltd. Printed and Published in Ireland
274
cases evaluated by the Medicolegal Council of Denmark in the period 19811985 dealing with traffic offenders suspected of drug influence. Materials
All records in the file concerning drivers suspected of driving under the influence of drugs were examined for the 5-year period 1981-1985. 476 cases were found. The cases had been requested by the police authorities, when suspicion was raised that a driver was influenced by drugs because of traffic crashes W = 1841, otherwise abnormal traffic behavior or denunciation W = 2921. This method implies highly selected material. The Medicolegal Council receives a form from the Police authorities, principally including information about self-reported drug intake, a description of the incident and the result of a clinical examination accompanied by a blood and/or urine sample. Antidepressants, CNS stimulating agents, cannabinoids and opioids were often detected in urine whereas other drugs were found mainly in blood. Based on the results of the clinical examination and from the biological specimens the Medicolegal Council made an estimation whether the driver was supposed to have been influenced by drugs. In some instances sufficient data to make this estimation was not available. Methods
From the records the following information was collected: Year of incident, age, sex, drugs taken, blood alcohol concentration (BAC), accident/no accident and the estimate concerning the degree to which the driver was influenced by drugs. 100 different legal and illegal drugs were detected. These were classified into the following ten groups: barbiturates (BB) benzodiazepines (BZ), other sedatives (AS), cyclic antidepressives (CA), antiepileptics (EP), neuroleptics (major tranquilizers) (NL) opioids including heroin (OP), CNS stimulants (ST), cannabinoids (TC) and the remaining drugs (AN). Results are expressed as mean r S.D. Statistical analyses were performed with Chi-square test, Wilcoxon test, Fishers test and Students t-test for unpaired data. Results
Records (4761 were registered out of which 15 were excluded, 13 with BAC at or above 0.8 g/l (legal limit in Denmark) and 2 without specimen or information on drugs taken available. The sex distribution in the remaining 461 cases was 62 females and 399 males. Age was stated in 448 cases and the distribution for each sex is shown in Fig. 1. Females were 39 f 10 years of age and males 32 -+ 11 years (P < 0.0011.The mean age for all drivers was 33 f 11 years.
275
Frequency
pJ
cl
Females
Males
O
I
15
20
30
40
50
60
70 Years
of
I-
age
Fig. 1. The relative age distribution for male Vv = 38’7) and female W = 61) drivers to be influenced by drugs. P = 0.000012 was obtained by Students t-test.
suspected
Figure 2 shows the distribution of age for drivers involved in a crash W = 173) and for those not UV = 275). Mean age of the drivers with a crash was 36 -t 12 years, significantly higher than for the rest with 31 + 11 years (P < 0.001). The drug group by far the most frequent represented was BZ, taken by 298 (65%) of 461 drivers. Next was OP found in 173 (38%) cases. The two minor groups were CA and EP by 12 (2.6%) each. Numbers are seen in Table 1. In Fig. 3 the relative frequencies of drug groups among females and males are shown. Women had been taking drugs from the groups AN and and P = 0.057 respectively. For CA more often than expected, P = 0.0092 two other groups men were represented in surplus, OP (P = 0.037) and TC
Frequency 35% 1
Drivers
60
from
crash
70 Years
Fig. 2. The relative age distribution for drivers involved in an accident involved in an accident (N = 275). P = 0.000012 by Students l-test .
of age
W = 173). and those
not
276 TABLE
1
NUMBER OF TIMES OTHER DRIVERS.
DRUGS
WERE
AN
AS
BB
BZ
Accident No accident
14 24
15 22
34 33
Total
38
37
67
DETECTED
AMONG
ACCIDENT
DRIVERS
AND
CA
EP
NL
OP
ST
TC
Total
128 170
8 4
5 7
15 17
40 133
9 19
21 50
289 479
298
12
12
32
173
28
71
768
P = 0.0161. In 250 (54%) of the cases drugs from two or more groups were found, thus making the total number of ‘a drug group found’ 768 and total percentage in Fig. 3 above 100%. There was no significant difference between males and females in this respect. A traffic accident preceded 180 (39%) of the cases, without differences between sex. The relative frequency of crash for each drug group is shown in Fig. 4, the absolute numbers in Table 1. Of those cases with CA, 67% were preceded by an accident (P = 0.0381 and with BB 51% (P = 0.027). BZ was only marginally above mean with 43% (P = 0.058). Among the cases with OP only 23% (P < 0.00011 had crashed. The remaining groups did not differ significantly. Two or more drugs found was not correlated with a higher incidence of accident compared to one drug found. Among the drivers who could be evaluated 72% were actually found to be influenced by drugs. This percentage for each group is seen in Fig. 5. For BB it was 91% Cp = 0.00531 and for TC 85% Cp = 0.0581. Below mean were cases with CA having 55% (P = 0.0491 estimated as influenced. Frequency 8O%j
Opioids*
Antidepressives
Barbiturates
Benzodiazepines Other
Antiepileptics sedatives
Neuroleptics
Other
drugs*
Cannabinoids* Stimulating
drugs
Fig. 3. The frequency by which each drug group was detected among male and female drivers. An asterisk indicates P < 0.05 for the drug group. Overall Chi square was 27.40, P = 0.0022.
Opioids* Other drugs Antidepressives* Barbiturates* Cannabinoids Antiepileptics Benzodiazepines Stimulating drugs Other sedatives Neuroleptics Fig. 4. The proportion of the drivers that had been taking a specific drug, that were involved in an accident. An asterisk indicates P < 0.05 for the drug group. Overall Chi-square was 32.05, P = 0.00039. Frequency 100% 1
Mea
=7
Other drugs Antidepressives* Opioids Barbiturares* Antiepileptics Cannabinoids Benzodiazepines Stimulating drugs Other sedatives Neuroleptics Fig. 5. The percentage of drivers that had been taking a specific drug, that were estimated influenced by drugs. An asterisk indicates P < 0.05 for the drug group. Overall Chi-square was 21.70, P = 0.017.
278
Discussion
This publication includes all records on cases about suspected drug influence while driving, evaluated by the Danish Medicolegal Council in the period 1981- 1985. As a sample from all drivers influenced by drugs, it is by nature quite biased. A driver will have a record only after several incidents. The police must have become suspicious of a driver influenced by drugs other than alcohol. The driver must then have a clinical examination performed and a biological specimen taken. Only in that case the police may send in a form to the Medicolegal Council, who will make a record. In spite of this strong selection the material seems to be comparable to the background population in Denmark. The mean age of the drivers in this material does not differ from that of killed and injured drivers in Copenhagen in 1984. Also the subgroup of drivers involved in a traffic accident has a mean age and sex distribution similar to drivers causing accidents in Copenhagen in 1984 [2]. A recent Danish paper concerning the frequency of BZ in blood samples submitted for BAC determination at the Institute of Forensic Chemistry in Copenhagen [3] states that 6% of a sample from 26,000 yearly specimens had a BZ concentration >0.2 pmol/l. The age distribution in this material does not differ from that of the BZ group in the present paper. There was 1% with BZ and a BAC < 0.5 g/l. By extrapolation, this is calculated to 1200 drivers with BZ in blood and BAC less than 0.5 g/l, in a 5-year period. The Medicolegal Council had 298 records concerning BZ, i.e. only about 25% of the cases in which the police suspected drug influence, where BAC was below 0.5 g/l and BZ would have been detectable in blood. The authors find it surprising that the drivers involved in an accident were older (36 years) than the rest (31 years, see Fig. 21, but did not find papers confirming this observation. Some of the explanation might be selection by police, and increased (legal) drug intake with increasing age. The major drug group was BZ, taken by 70-80%. secondly OP and then TC. This appearance of drugs does not differ appreciably from the results of other newer investigations [4-g]. Where differences are encountered, it seems to be due mainly to different grouping and analyzing techniques (e.g. BZ vs. sedatives and blood vs. urine). Some authors have found surprisingly many ‘illicit drugs’ (subgroup of OP, ST and TC) [lo - 121. This may be due to cultural differences and differing attendance by the police. Kaempe & Dalgaard analyzed 283 liver specimens from persons killed by traffic 1968 - 1974 [13]. In 20% drugs were detected, mainly BB, but only in one case BZ. To evaluate the different drug groups’ contribution to accidents, the frequency of each group among accident drivers is compared to that of the other drivers, as seen in Fig. 4. Since BZ was the biggest group by far, the result of this group was close to the average of the total study group. Anyway, it was more frequent among crash drivers, though not statistically significant (P = 0.0581. The traffic hazards caused by BZ have been settled
279
in a controlled trial [14] and a number of epidemiological investigations [68,151. CA was by 67% mainly found after an accident. This is in agreement with the findings of other papers [4,8]. The drug group could therefore be assumed to be of qualitative importance, though only sporadically found, 2.6% of the cases in this material. As far as CA is used as a treatment for depressive disease, this probably significantly [l] biases the result. The frequency by which BB was found in this material decreased over time. In Denmark the group had limited indications since 1985, only available for anesthesia and for treatment of epilepsy. However, it was still of some quantitative importance, also found by another investigation [4]. Two studies did not find this coherence between BB and accidents. Ulrich, et al. examined urine from an accident group and blood from a control group. BB was detected with equal frequency in both. An interview inquiry of 14,824 breathalyzed motorists gave the same result [8]. Among accident drivers OP was found significantly less often than expected. This probably has several causes. The police might have knowledge about drug abusers in local area, and thus be more likely to suspect these drivers to be influenced by drugs. Secondly, in steady state the dosage of OP does not always imply a traffic danger, because of tolerance and increased attention to compensate a possible decrease in performance [1618]. In this matter Ulrich [7] and MacPherson [8] had deviating results, but as outlined above, with a weaker design of investigation. Comparing the evaluation of drug influence in the individual groups only two ma]or divergences were found. Drivers taking BB were found to be influenced very often in agreement with the high incidence of crashes. On the contrary, drivers taking CA were found to be influenced less often than they were involved in accidents. This might be explained by their depressive disease state. Conclusion
This material points out a likely traffic danger by drivers taking drugs, mainly BB, BZ and CA, in accordance with several other investigations. In contrast, it seems that drivers under influence of opioids do not contribute excessively to accidents, even though a few papers report different results. It should be stated that all studies except one were epidemiologic. No certain connection between cause and measured effect is therefore unveiled. References 1
S.L. Nielsen, L.Q. Christensen Ugeskr. Laeg., 151 (1989) 1822-
2
Statistical yearbook (1985) p. 150.
and L.M. 1825.
of Copenhagen
1985.
Nielsen,
Analgesics,
Copenhagen
Benzodiazepines
Statistical
Office,
and
traffic.
Copenhagen,
64
280 3 4 5
10
11 12
13 14
15 16 17 18
A. Steentoft, K. Worm and K. Christensen, The incidence of benxodiazepines in Danish road users. Ugeskr. Laeg., 150 (1988) 2383-2386. D. Wilson, Experience with drugs and driving in Queensland, Australia. Med Sci Law, 25 (1985) 2 - 10. A.P. Mason and A.J. McBay, Ethanol, Marijuana and other drug use in 600 drivers killed in single-vehicle crashes in North Carolina, 1978- 1981. J. Forensic Sk., 29 (1984) 987- 1026. R. Honkanen, L. Ertama, M. Linniola et al., Role of drugs in traffic accidents. BT. Med J., 281 (1980) 1309- 1312. L. Uirich, 0. Rudin, A. Amsler and P. Zink, Htiufigkeit von medikamenten im straBenverkehr. Z. Rechtsmed, 93 (1984) 95-110. R.D. MacPherson, J. Per1 and G.A. Stamer, Self-reported drug usage and crash-incidence in breathalyzed drivers. Accid Anal Prev., 16 (19841139-148. K. Wehr and R.D. Maier, Illicit drugs among arrested drivers in the’border region of Aix La Chapelle. In L. Goldberg feds.), Alcohol, Drugs and traffic Safety, Vol. 1, Proceedings, 8th International Conference on Alcohol, Drugs and Traffic Safety, 1980, Wiksell, Stockholm, 1981, pp. 303-314. P. Holmgren, E. Loch and J. Schubert, Drugs in motorists traveling Swedish roads: On-theroad-detection of intoxicated drivers and screening for drugs in these offenders. Forensic Sci Znt.. 27 (1985157-65. A.F. Williams, M.A. Peat, B.S. Crouch, J.K. Wells and B.S. Finkle. Drugs in fatally injured young male drivers. Publ. Health Rep., 100 (19851 19-25. R. Warren, H. Simpson, J. Hilchie et al., Drugs detected in fatally injured drivers in theprovince of Ontario. In L. Goldberg feds.), Alcohol, Drags and traffic Safety, Vol. 1, Pro ceedings, 8th International Conference on Alcohol, Drugs and Traffic Safety, 1980, Wiksell, Stockholm, 1981, pp. 203- 217. B. Kaempe and J.B. Dalgaard, Drugs, alcohol and carbon monoxide in victims of fatal traffic accidents. Ugeskr. Laeg., 141 (19791 1037-1040. J.J. de Gier, B.J. t’Hart, F.A. Nelemans and H. Bergman, Psychomotor performance and real driving performance of outpatient receiving diazepam. Psychopharmacology, 73 (1981) 340-344. J.W.F. Haffner. 0. Bo and K.M. Lunde, Alcohol and drug consumption as causual factors in road traffic accidents in Norway. J. Traf. Med, 2 (1974152-56. D.V. Babst. S. Newman, N. Gordon and A. Warner, Driving records of methadone maintenance patients in New York state. J. Drug Issues, 3 (19731285- 292. R.D. Blomberg, D.F. Preusser, Narcotic use and driving behavior. Accia AnaL Prev., 6 (1974) 23-32. P.R. Dal Monte, G. Formica, G.L. Milandri and M. Ruggieri, Road accidents in young drugaddicts. Comparison with a test group. In L. Goldberg feds.), AlcohoZ Drags and traffic Safety, Vol. 1, Proceedings 8th International Conference on Alcohol, Drugs and Traffic Safety, 1980, Wiksell, Stockholm, 1981, pp. 160- 164.
