This article was downloaded by: [York University Libraries]On: 14 November 2014, At: 07:30Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,37-41 Mortimer Street, London W1T 3JH, UK

Click for updates

Traffic Injury PreventionPublication details, including instructions for authors and subscription information:http://www.tandfonline.com/loi/gcpi20

Prevalence of Alcohol and Drugs Among Car andVan Drivers Killed in Road Accidents in Norway: AnOverview From 2001 to 2010Asbjørg S. Christophersena & Hallvard Gjerdea

a Division of Forensic Sciences, Norwegian Institute of Public Health, Oslo, NorwayAccepted author version posted online: 02 Oct 2013.Published online: 27 May 2014.

To cite this article: Asbjørg S. Christophersen & Hallvard Gjerde (2014) Prevalence of Alcohol and Drugs Among Car and VanDrivers Killed in Road Accidents in Norway: An Overview From 2001 to 2010, Traffic Injury Prevention, 15:6, 523-531, DOI:10.1080/15389588.2013.848981

To link to this article: http://dx.doi.org/10.1080/15389588.2013.848981

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) containedin the publications on our platform. However, Taylor & Francis, our agents, and our licensors make norepresentations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of theContent. Any opinions and views expressed in this publication are the opinions and views of the authors, andare not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon andshould be independently verified with primary sources of information. Taylor and Francis shall not be liable forany losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoeveror howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use ofthe Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematicreproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in anyform to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Traffic Injury Prevention (2014) 15, 523–531Copyright C© Taylor & Francis Group, LLCISSN: 1538-9588 print / 1538-957X onlineDOI: 10.1080/15389588.2013.848981

Prevalence of Alcohol and Drugs Among Car and VanDrivers Killed in Road Accidents in Norway: An OverviewFrom 2001 to 2010

ASBJØRG S. CHRISTOPHERSEN and HALLVARD GJERDE

Division of Forensic Sciences, Norwegian Institute of Public Health, Oslo, Norway

Received 24 May 2013, Accepted 22 September 2013

Objectives: To examine the prevalence of alcohol and drugs in blood samples collected from car and van drivers killed in trafficaccidents in Norway during the time period from 2001 to 2010.

Methods: Blood samples (n = 676, 63% of all killed drivers) were analyzed for alcohol, psychoactive medications, and illicit drugs.The cutoff limits for positive results were set according to the new legislative limits under the Norwegian Road Traffic Act. The resultswere assessed in relation to sex and age, time of day and day of week, and single- versus multiple-vehicle and all investigated vehicleaccidents.

Results: Alcohol or one or more drugs was detected in samples from 40.2 percent of all investigated drivers, with 28.7 percentshowing blood concentrations of at least 5 times the legislative limits. For the investigated female drivers, the total prevalence was 24.0percent. Among the single-vehicle accidents, alcohol or drugs was found in 63.8 percent of the cases, with 49.1 percent showing bloodconcentrations of at least 5 times the legislative limits. Alcohol was detected in 25.3 and 49.1 percent of samples from all investigateddrivers and among drivers killed in single-vehicle accidents, respectively. Psychoactive medications were found in 14.4 and 17.7 percentand illicit drugs in 14.1 and 19.2 percent, respectively. The most commonly detected group of medications was benzodiazepines, andamphetamines and tetrahydrocannabinol were the most commonly detected illicit drugs. The prevalence of alcohol alone was highestamong drivers under the age of 25, and the combination of alcohol with other drugs was highest among drivers under the age of35. Drivers between the ages of 25 and 54 showed the highest prevalence of medications and/or illicit drugs without the presence ofalcohol. The highest prevalence of alcohol or drugs was found among drivers killed in single-vehicle accidents on weeknights (83.8%)and on weekend nights (89.3%).

Conclusions: The findings confirm that a large number of fatally injured drivers, in particular among drivers involved in single-vehicleaccidents, had concentrations of alcohol or drugs above the new legislative limits introduced in 2012. In many cases, concentrationsof at least 5 times the limits were found. The proportion of drivers killed who tested positive for alcohol or other drugs did not changeduring the study period; however, the total number of drivers killed per year decreased by about 20 percent. Some changes were alsoobserved with regard to the types of benzodiazepines and amphetamines detected during the 10-year period.

Keywords: fatal accidents, single-vehicle accidents, alcohol, drugs, driving under influence

Introduction

Many studies have documented an increased risk of involve-ment in motor vehicle accidents when driving under the in-fluence of alcohol, illicit drugs, or psychoactive medications(Bernhoft et al. 2012; Blomberg et al. 2009; Borkenstein et al.1974; Gjerde el al. 2011; Kelly et al. 2004; Movig et al. 2004).For many years, it has been well documented that the risk of

Managing Editor David Viano oversaw the review of this article.Address correspondence to Dr. Asbjørg S. Christophersen, Di-vision of Forensic Medicine and Drugs Abuse Research, Nor-wegian Institute of Public Health, P.O. Box 4404 Nydalen, 0403Oslo Norway. E-mail: Asbjorg.Christophersen@fhi.noColor versions of one or more of the figures in the article can befound online at www.tandfonline.com/gcpi.

being involved in an accident when driving under the influ-ence of alcohol increases as a function of the blood alcoholconcentration (BAC; Blomberg et al. 2009; Borkenstein et al.1974; Phillips and Brewer 2011).

In recent years, studies on the risks caused by psychoactivesubstances have focused more on illicit drugs and medications.In several controlled clinical experimental studies, using psy-chomotor tests, adverse effects on human performance thatmay reduce driving ability have been documented (Berghauset al. 2007; Ramaekers et al. 2006; Verster and Mets 2009).Other sources of complementary information are based uponcase-control studies, which are regarded as one of the bestways to approach the question of whether a psychoactive com-pound is related to an increased accident risk (Barbone et al.1998; Berghaus et al. 2007; Blomberg et al. 2009; Gjerde el al.2011; Kelly et al. 2004; Movig et al. 2004; Orriols et al. 2009,

Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

7:30

14

Nov

embe

r 20

14

524 Christophersen and Gjerde

2010; Ravera et al. 2011; Smink et al. 2005). Another type ofresearch that has contributed to the knowledge on substancesrecognized as hazardous to road traffic safety is epidemio-logical studies on drug-related accidents (A. Holmgren et al.2007; P. Holmgren et al. 2005; Jones et al. 2009; Mørland et al.2011).

Varying results have been obtained for some of the sub-stances in earlier studies, and the relationship between blooddrug concentrations and increased accident risk has not alwaysbeen clearly demonstrated (Mørland 2000; Walsh et al. 2004).There are several reasons for these varying results, includ-ing the use of nonstandardized study protocols, with differentselections of the types of accident cases, different biologicalmatrices used for analyses of alcohol and drugs (whole blood,serum, urine, oral fluid etc.), and different analytical programsfor drug analysis and cutoff limits, in addition to nonrepre-sentative selections of controls in case-control studies.

Standardized guidelines for studies on alcohol- and drug-related road traffic accidents have previously been proposedby an expert committee (Walsh et al. 2008). One importantrecommendation was to use a broad analytical program, withcutoff limits of relevance to possible impairment. A programnot including the most relevant impairing substances couldmiss important information during risk calculations and eval-uations of which drugs could be the main cause of an accident.Several epidemiological studies have shown that combinationsof 2 or more drugs are frequently found in blood samples col-lected from injured or killed drivers (Drummer et al. 2004;Gjerde et al. 2011; A. Holmgren et al. 2007; P. Holmgren et al.2005; Mørland et al. 2011; Orriols et al. 2009). Multidrug useis also frequently detected in blood samples collected fromdrivers apprehended by the police, due to the suspicion of im-pairment (Christophersen and Mørland 2008; A. Holmgrenet al. 2007; Jones et al. 2009).

The prevalence of alcohol and drugs in regular traffic andin road traffic accidents varies from one country to another,as documented by the EU project Driving Under the Influ-ence of Drugs, Alcohol and Medicines (Houwing et al. 2011;Legrand et al. 2013, 2014), where 13 countries collaborated inperforming roadside surveys to study the prevalence of alco-hol and drugs among random drivers and drivers involved inroad traffic accidents. All participating countries performedthe same set of analyses, for the same set of drugs, using thesame cutoff limits (Pil et al. 2009). The results showed that theprevalence of alcohol and illicit drugs among drivers in normaltraffic and those involved in accidents was higher in the south-ern European countries, whereas the use of medications, suchas benzodiazepines (BZDs), was more common among driversin the Nordic countries (Houwing et al. 2011; Isalberti et al.2011; Legrand et al. 2013, 2014). Another study comparingalcohol and drugs among drivers killed on Nordic roads usingthe same protocol for the selection of cases, biological matrix,analytical program, and cutoff limits for alcohol and drugsshowed that BZDs were more commonly detected in samplesfrom drivers killed in Norway, whereas alcohol was more fre-quently found in samples from other countries (Mørland et al.2011). Studies from later years have also shown that some psy-choactive medications and illicit drugs appear to play a moreimportant role during evaluation of causes and accident risk.

An older study from Norway, which included drivers killedfrom 1989 to 1990, where the blood samples were analyzed fora comprehensive number of psychoactive medications and il-licit drugs, in addition to alcohol, showed that alcohol was thedominant psychoactive substance detected (28.8% of all acci-dents), with all other drugs being less common (total 16.4%;Gjerde et al. 1993).

In 1936, Norway was the first country to introduce a le-gal limit for driving under the influence of alcohol. The BAClimit at that time was set to 0.5 g/L but was further reduced to0.2 g/L in 2001. In 1959 the Norwegian Road Traffic Act wasextended to include a law on impairment due to drugs. Thislaw was changed in 2012, when legislative limits for 20 drugs,including both illicit drugs and medications with warning la-bels, were implemented (Vindenes et al. 2012).

The aim of this study was to investigate the prevalence ofalcohol and drug impairment among drivers of cars and vansinvolved in fatal accidents during the time period from 2001to 2010, according to the new legislative blood drug concen-trations limits established under the Norwegian Road TrafficAct. An additional aim was to compare the prevalence ofalcohol and other drugs among drivers killed in multiple-and single-vehicle accidents. Based upon results from ear-lier accident studies, drivers killed in single-vehicle accidentsare considered, in most cases, culpable because no other ve-hicles were involved, the latter in contrast to drivers killedin multiple-vehicle accidents, which also includes nonculpa-ble drivers, where the culpable drivers may or may not havebeen impaired by alcohol or drugs and in some cases survived(Drummer et al. 2004; Gjerde et al. 1993; Mørland et al. 2011;Stoduto et al. 1993).

Due to the comprehensive changes in the Norwegian RoadTraffic Act, applying legislative limits for 20 nonalcoholicdrugs, the results from this overview study during the last10 years may provide important background data for furtherevaluation of the effects due to the changes in the law and inplanning other preventative measures.

Materials and Methods

Samples were selected from those routinely submitted by thepolice for alcohol and drug analysis in cases of fatal traffic ac-cidents. Additional information was obtained from the Nor-wegian Road Traffic Accident Registry, operated by StatisticsNorway, which is based upon information submitted by thepolice. This registry contains the national identification num-ber of the drivers, in addition to information on the accident,such as the place, date and time of day, number of vehiclesinvolved, and other relevant information.

Blood samples from drivers who are killed are collectedshortly after the accident when the driver is still alive, shortlyafter death, or during the legal autopsy. All samples collectedshortly after an accident and before a possible autopsy aresent to the Norwegian Institute of Public Health (NIPH)for toxicological analyses of alcohol and psychoactive drugs.All autopsy samples, except for those collected in 2 coun-ties in the middle of Norway, are also sent to the NIPH foranalyses. The type of equipment used for sample collection

Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

7:30

14

Nov

embe

r 20

14

Road Accidents in Norway 525

Table 1. Drugs analyzed, legislative limits used as cutoff values, and prevalence (numbers in brackets) of alcohol, illicit drugs, andmedications in blood samples collected from drivers killed in traffic accidents

Above legislative limitsa 5× above legislative limits

Legislative All accidents Single-vehicle All Single-vehicleCompounds Comments limits (ng/mL) (n = 676) accidents (n = 271) accidents accidents

Alcohol Changed from 0.05 g/L in 2001 0.02 g/L 25.3 (171) 49.1 (133) 20.6 (139) 40.6 (110)Alprazolam Anxiolytic 3 0.9 (6) 1.1 (3) 0.6 (4) 0.7 (2)Amphetamine ADHD treatment. Illicit 41 8.0 (54) 9.6 (26) 4.4 (30) 4.8 (13)Clonazepam Anticonvulsant 1.3 1.9 (13) 3.0 (8) 0.7 (5) 1.1 (3)Cocaine Stimulant. Illicit 24 0.3 (2) 0.4 (1) 0.0 (0) 0.0 (0)Diazepam Anxiolytic, sedative 57 5.5 (37) 7.4 (20) 1.0 (7) 1.8 (5)Flunitrazepam Hypnotic 1.6 1.8 (12) 1.5 (4) 0.4 (3) 0.4 (1)MDMA Stimulant. Illicit 48 1.0 (7) 1.8 (5) 0.3 (2) 0.7 (2)Methadone Opioid replacement therapy 25 0.9 (6) 0.4 (1) 0.0 (0) 0.0 (0)Methamphetamine Stimulant. Illicit 45 3.8 (26) 4.8 (13) 1.9 (13) 2.2 (6)Morphine Strong pain killer 9 0.6 (4) 0.7 (2) 0.3 (2) 0.4 (1)Nitrazepam Hypnotic 17 0.7 (5) 0.7 (2) 0.0 (0) 0.0 (0)Oxazepam Anxiolytic, sedative 172 2.7 (18) 3.0 (8) 0.1 (1) 0.4 (1)THC Cannabis, marihuana. Illicit 1.3 7.2 (49) 10.0 (27) 1.3 (9) 1.5 (4)Zolpidem Hypnotic 31 0.4 (3) 0.7 (2) 0.1 (1) 0.0 (0)Zopiclone Hypnotic 12 2.5 (17) 1.5 (4) 0.6 (4) 0.4 (1)Medications 14.4 (98) 17.7 (48) 3.8 (26) 4.8 (13)Illicit drugs 14.1 (95) 19.2 (52) 7.4 (50) 8.5 (23)One or more substances

including alcohol40.2 (272) 63.8 (173) 28.7 (194) 49.1 (133)

aA higher cutoff than the legal limit was used for clonazepam during the study period.

and the storage conditions before analyses have been previ-ously described (Gjerde et al. 2011). The analytical results arerecorded in the forensic toxicological database at NIPH. In afew cases, the police initially request limited testing, usuallyfor analysis of alcohol only; these samples are thawed andanalyzed for the remaining necessary drugs (Table 1).

The cases selected in this study were car and van driverswho died within 24 h after involvement in a road traffic ac-cident during the time period from 2001–2010, where bloodsamples had been sent to NIPH for toxicological analyses.Recorded data included age, gender, date and time of the ac-cident, time of sampling (if available), date and time of death,type of vehicle, single-vehicle accident (yes/no), and policedistrict. Drugs detected as a result of reported treatment, orlikely treatment, after the accident were omitted from furtherevaluation; this included low concentrations of morphine anddiazepam, if detected without the presence of the metaboliteN-desmethyldiazepam.

Permission for the study, including the coupling of the datawith the Road Traffic Accident Registry, was given by theRegional Committee for Medical Health Research Ethics, theMinistry of Justice, and the Director of Public Prosecutions.

The results from the accident cases were divided in 3groups—all accidents, multiple-vehicle accidents, and single-vehicle accidents—and then further divided by sex and age,day and night, and weekdays or weekends.

Analysis of Alcohol and Drugs

The analytical program and method references have been pre-viously described (Gjerde et al. 2011). The samples and anal-yses were handled according to the normal routines for foren-sic toxicological analysis. The laboratory is accredited by the

Norwegian Accreditation (Norwegian Accreditation) accord-ing to ISO17025 (ISO Central Secretariat 2005) for samplehanding, method evaluation, analyses, and interpretation ofresults. The laboratory also participates regularly in differentinternational proficiency testing programs.

The compounds analyzed for, including blood drug con-centrations and cutoff limits, are presented in Table 1. Thecutoff limits were set according to the legislative limits underthe Norwegian Road Traffic Acts (Vindenes et al. 2012), im-plemented in February 2012, except for that of clonazepam,for which 5 ng/mL was used as the cutoff limit during most ofthe study period. The results from positive screening analyseswere not reported if sufficient blood volumes for confirmationanalyses were not available.

Statistical Analyses

Pearson’s 2-sided chi-square test for categorical data wasapplied.

Results

The number of car and van drivers who died on Norwegianroads decreased from about 135 per year in 2001–2003 toapproximately 110 drivers per year in 2008–2010 (StatisticsNorway 2011). The number of biological samples sent for tox-icological analyses decreased similarly during the same timeperiod, from about 80 to 60 per year. During the study pe-riod, from 2001 to 2010, 1077 care and van drivers were killedin traffic accidents on Norwegian Roads, of whom 81 per-cent (n = 872) were male and 19 percent (n = 205) werefemale. Autopsy samples from 2 counties were investigated at

Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

7:30

14

Nov

embe

r 20

14

526 Christophersen and Gjerde

Trondheim University Hospital and therefore represent someof the missing cases (approximately 5% of all driverskilled).

Blood samples from 676 (63%) of the drivers killed weresent to the NIPH for toxicological analyses. Of these, 83 per-cent (n = 560) were from males and 17 percent (n = 116)were from females; 271 of the samples (40%) were from driverskilled in single-vehicle accidents and 405 (60%) from multiple-vehicle accidents. There were some differences between the fa-tally injured drivers who were tested for alcohol or drugs andthose who were not tested. Females constituted 16.9 percentof the drivers tested and 21.7 percent of the drivers who werenot tested (P = .05). Drivers fatally injured in crashes dur-ing weekend nights constituted 14.1 percent of the driverstested and 7.5 percent of the drivers who were not tested(P < .01). There were no significant differences for other timeintervals of the week. Drivers aged 25–34 years constituted21.9 percent of those tested and 16.7 percent of those nottested (P = .04), whereas drivers above 64 years of age con-stituted 14.6 percent of the drivers tested and 23.7 percent ofthose not tested (P < .01). There were no significant differencesfor other age groups. Drivers killed in single-vehicle crashesconstituted 40.1 percent of the drivers tested and 33.7 percentof those who were not tested (P = .04). This means that thesample of fatally injured drivers tested most likely overesti-mates the proportion of drivers positive for alcohol and drugsbecause the samples not tested were much less likely to have al-cohol or drugs (more females; fewer on weekend nights; feweraged 25–34; more drivers over age 64; fewer in single-vehiclecrashes).

In about 90 of the cases, analysis for one or more of thedrugs presented in Table 1 was not performed; in 37 of thesecases, alcohol was the only drug analyzed for. This limitedtesting was either due to a very high BAC level detected ordue to insufficient volumes of blood samples to perform thecomplete analytical program.

Fig. 1. Distribution of BAC levels in samples testing positive foralcohol and drugs and samples positive for alcohol only (BAC ≥0.2 g/L).

Alcohol and Drugs

Table 1 shows the prevalence of alcohol and drugs among allinvestigated drivers killed and those killed in single-vehicleaccidents. Blood concentrations of 5 times or more above thenew legislative limits are also presented for the single drugsand drug groups in both accident categories.

All Fatal Accidents

In total, 40.2 percent (n = 272) of the investigated driverskilled had one or more substances detected at concentrationsabove the legislative limits. For 28.7 percent (n = 194) of thedrivers, the blood concentrations were 5 times or more abovethe legislative limits for at least one compound detected.

Alcohol was the most frequently detected compound.BACs above the legal limit of 0.2 g/L were found in 25.3percent of the cases and above 1.0 g/L in 20.6 percent ofthe cases. The distribution of BACs among drivers who hadused alcohol only or in combination with drugs is presentedin Figure 1. The prevalence of illicit drugs above the legisla-tive limits was 14.1 percent, and 7.4 percent of the driverskilled had blood concentrations of at least 5 times the limitfor one or more compounds. The most frequently detected il-licit drugs were amphetamine/methamphetamine (9.0%) andtetrahydrocannabinol (THC) (7.2%); drug concentrations ofat least 5 times the legislative limits were frequently detectedfor these substances (Table 1).

Altogether, 14.5 percent of the drivers killed had used psy-choactive medications before the accident, resulting in bloodconcentrations above the legislative limits. For 3.8 percent ofthe cases, the dosages used had resulted in blood concentra-tions of at least 5 times the limit. BZDs represented the mostfrequently detected group of medications (Table 1).

Among drivers killed who tested positive for medicationsor illicit drugs above the legislative limits (but no alcohol;n = 101), 43.6 percent had used 2 or more drugs before theaccident. Among drivers killed with only one drug detected(illicit or medicinal), 50.5 percent had used drugs at doses rep-resenting blood concentrations at least 5 times the legislativelimit.

Multiple- and Single-Vehicle Accidents

Figures 2, 3, and 4 present the prevalence of alcohol alone,drugs alone, and combinations of alcohol and drugs amongall drivers killed and those killed in multiple- and single-vehicleaccidents. The total prevalence of alcohol or drugs was high-est among drivers killed in single-vehicle accidents (63.8%),much higher than those killed in multiple-vehicle accidents(24.8%).

More details on the single drugs that were detected abovethe legislative limits and on those detected at 5 times the limitsamong drivers killed in single-vehicle accidents are presentedin Table 1. The same blood drug pattern was found amongsingle-vehicle drivers compared to all investigated drivers, ex-cept that the prevalence of most single drugs or drug groupswas higher (Table 1).

Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

7:30

14

Nov

embe

r 20

14

Road Accidents in Norway 527

Fig. 2. Prevalence of alcohol alone, combinations of alcohol withdrugs, and drugs only. (A) All drivers killed; (B) drivers killed inmultiple-vehicle accidents; and (C) drivers killed in single-vehicleaccidents.

Fig. 3. Prevalence of alcohol alone, combinations of alcohol withdrugs, and drugs only.

Fig. 4. Prevalence of alcohol alone, combinations of alcohol withdrugs, and drugs only, among drivers killed at different time pe-riods of the week. (A) All drivers killed; (B) drivers killed inmultiple-vehicle accidents; and (C) drivers killed in single-vehicleaccidents.

A total of 49.1 percent of the drivers killed in single-vehicleaccidents had used one or more drugs that resulted in bloodconcentrations at least 5 times the limit (Table 1). Illicit drugswere detected among 19.2 percent of the drivers killed insingle-vehicle accidents, and 8.5 percent had concentrationsof more than 5 times the legislative limit. Medications weredetected among 17.7 percent and 4.8 percent had blood con-centrations of more than 5 times the limit. The most commonlydetected illicit drugs and medications were the same as thosedetected among all investigated drivers killed.

Age Distribution

In total, 26 percent of the drivers killed in Norway from 2001to 2010 were younger than age 25. Drivers aged 25–34 repre-sented 20 percent of the drivers killed, with those aged 35–54representing 26 percent and those above the age of 54 repre-senting 28 percent.

There were significant differences in the findings of alcohol,illicit drugs, and medications for the different age groups (P ≤

Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

7:30

14

Nov

embe

r 20

14

528 Christophersen and Gjerde

.001 for alcohol, illicit drugs, and medications). No significantdifferences were found between the age groups up to the age of54 on the total prevalence of cases testing positive for alcoholand/or other drugs (Figure 2); however, alcohol alone wasmore frequently detected among drivers killed under the ageof 25 for all investigated cases, and drivers under the age of 25and those between the ages 35 and 54 killed in single-vehicleaccidents showed the highest prevalence of alcohol withoutthe presence of other drugs. Combinations of alcohol anddrugs were most often found among drivers up to the ageof 35 (Figure 2). Cases with medications and/or illicit drugsdetected, without the presence of alcohol, were more oftenfound among drivers aged 25–54.

Gender

The prevalence of samples testing positive for one or moresubstances was higher among male drivers than female drivers(Figure 3). Only 24 percent of the investigated female drivershad used alcohol or drugs before the accident compared to44 percent of the male drivers (P < .001). The prevalence ofalcohol only and alcohol in combination with other substancesamong female drivers was 6.0 and 1.7 percent, respectively,compared to 20.9 and 8.0 percent, respectively, among maledrivers (P < .02). Significant differences were also found forillicit drugs (P < .01) but not for medications (P = .60).

Time Periods

Most of the accidents occurred during the daytime (between4:00 a.m. and 9:59 p.m.) on both weekdays (54%) and duringthe weekend (26%); 7 percent occurred on weeknights and 12percent on weekend nights. According to data from the Nor-wegian Roads Administration (Tor H. Eliassen, August 2009,email, personal communication), the proportions of motor ve-hicles on the roads during the time periods investigated were65 and 25 percent during the daytime on weekdays and week-ends, respectively, and 6 and 5 percent at night on weekdaysand weekends, respectively. Thus, the relative accident rate wasmuch higher on weekend nights than at any other time duringthe week.

There were significant differences in the prevalence of al-cohol (P < .001) and illicit drugs (P < .01) between the dif-ferent time periods, but no difference was observed for medi-cations (P = .44). The highest prevalence of alcohol or otherdrugs was found among drivers killed in accidents at night(Figure 4). For single-vehicle accidents on weeknights andweekend nights, 83.8 and 89.3 percent, respectively, of thedrivers killed had blood concentrations of alcohol, drugs, or acombination of them above the legislative limits, which prob-ably contributed to involvement in the accident.

Changes in Alcohol and Drug Prevalence Duringthe Study Period

No significant changes in the total prevalence of alcohol ordrugs was observed during the study period. However, changeswere observed for some of the individual drugs, particu-

Fig. 5. Alcohol and drug findings in blood samples collected fromdrivers killed from 2001 to 2010 (selected parameters).

larly for methamphetamine and amphetamine (P = .05 andP < .01, respectively, when comparing the first 5 years and thelast 5 years). The number of samples per year with metham-phetamine detected doubled during the study period, andthe number of cases with amphetamine decreased. The de-tection of flunitrazepam disappeared almost entirely duringthe study period, whereas alprazolam, and for a period oftime also clonazepam, increased (Figure 5). This pattern wasalso found in blood samples collected from drivers appre-hended by the police due to the suspicion of driving un-der the influence of drugs (Bogstrand et al. 2011; NIPH2011), in addition to being recorded among the drug seizures,as registered by the National Criminal Investigation Service(2011).

The most prevalent medications detected were diazepam(average 5.5%), oxazepam (average 2.7%), and zopiclone(average 2.5%). There was significant variation between thedifferent years, probably due to a random variation (data notshown in Figure 5). The apparent drop in the prevalence ofmedications from 2003 to 2004 may also be due to randomvariation.

Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

7:30

14

Nov

embe

r 20

14

Road Accidents in Norway 529

Discussion

The main strength of this study was that 95 percent of allinvestigated blood samples from car and van drivers killed inNorway during a 10-year period were analyzed for drugs, inaddition to alcohol and that for 87 percent of the samples, thesame comprehensive analytical program for detecting drugswas used. All analytical results were further evaluated usingcutoff limits equivalent to the legislative impairing blood drugconcentration limits set for each individual substance. Theuse of a standard protocol for the analytical program maygive important background data for further investigations onalcohol, psychoactive medications, and illicit drugs in relationto accidents and for evaluating the effects of the road safetyprogram, with particular focus on the effect related to thelegislative limits set for drugs.

Blood samples were received from only 63 percent of alldrivers killed during this time period. The distributions forage, gender, and time of the week were somewhat different forthe drivers investigated for alcohol or drug impairment com-pared to those not investigated; there was thus some selectionbias. The reasons for missing cases could probably be an eco-nomic priority made by the police or that the distance to theautopsy location was too far, in addition to the fact that thepolice could have already concluded the question of intake ofalcohol or drugs before the accident as being unlikely. Sam-ples from fatal traffic accident cases investigated at anotherlaboratory in Norway represented less than 5 percent of thetotal number of cases, thereby allowing a conclusion that thegeographical bias was small. A higher frequency of investi-gated cases would increase the validity of the results. If theuncertainties associated with the information on alcohol anddrugs in the missing cases are included, the total prevalenceof detecting impairing substances could be between 26 and 42percent and 44 and 65 percent for all possible accidents andsingle-vehicle accidents, respectively. However, we believe thatthe actual prevalence values, including those cases that werenot analyzed, were closer to the higher figures, consideringthat it is very unlikely that none of the noninvestigated driverskilled had used alcohol or any type of drugs before the acci-dent. A separate study comparing alcohol- and drug-relatedtraffic fatalities in Norway, Sweden, and Finland from 2001to 2002 showed no significant differences between the coun-tries for the total prevalence of alcohol and drugs, except afew variations between the individual drug groups (Mørlandet al. 2011); in this study more than 90 percent of the motorvehicle drivers killed every year in Sweden and Finland wereinvestigated, which was also reported earlier from Sweden (P.Holmgren et al. 2005; Jones et al. 2009).

The age group with the lowest investigation ratio was thoseover 64 years of age (51%). Earlier studies have shown that theuse of alcohol and/or psychoactive medications has also con-tributed to accidents among this group of drivers (Mørlandet al. 2011).

The annual number of traffic fatalities and drivers killed inNorway during the last 10 years shows a decrease of about20 percent. During this time period, the proportion of driverskilled who also tested positive for alcohol or drugs did notchange. Thus, this study shows that the influence of alcohol

and drugs still has a significant impact on more than 40 percentof all investigated drivers killed (Table 1). The latter problemwas in particular demonstrated for single-vehicle accidents,where approximately 2 out of 3 drivers killed had used oneor more substances before the accident, resulting in impairingblood drug concentration levels (Table 1). For this type of acci-dent the main cause and the responsibility for the crash couldbe directly linked to the individual driver. This same tendencyhas also been documented in earlier studies (Drummer et al.2004; Gjerde et al. 2013; P. Holmgren et al. 2005; Jones et al.2009; Mørland et al. 2011).

The results from the investigated drivers killed may evenrepresent minimum numbers, partly because not all psychoac-tive substances were included in all sample analyses, becausehigher cutoff levels were used in some cases for clonazepam,and, finally, because a few of the positive results from primaryscreening analyses could not be confirmed with an alterna-tive method due to insufficient sample volumes and the resultstherefore were reported as negative.

Despite the Road Traffic Act’s low legal BAC limit of0.2 g/L, alcohol is still the dominating substance detectedamong investigated drivers killed, often detected at very highBAC levels and in combination with drugs (Figure 1). Theresults are similar to those found in a study from 1989–1990concerning the prevalence of alcohol. However, the preva-lence of drugs has increased when compared to the resultsfrom 1989–1990 (Gjerde et al. 1993).

During the last few decades, the use of other drugs has alsobecome a significant negative factor influencing road trafficsafety in many countries, and the trend appears to be in-creasing compared to earlier studies (Drummer et al. 2004;P. Holmgren et al. 2005; Jones et al. 2009).

Due to possible changes in the drug use pattern, with fre-quent multidrug detections, it is important to provide an an-alytical program where a large number of psychoactive drugscan be detected. A comprehensive analytical program is par-ticularly important if the results are to be used for case-controlstudies (Gjerde et al. 2011) or compared with similar studiesfrom other countries with a different drug use pattern on theassumption that the other countries are using the same studyprotocol (European Monitoring Centre for Drugs and DrugAddiction 2008; P. Holmgren et al. 2005; Jones et al. 2009;Mørland et al. 2011).

Impairment

The legislative limits for most of the illicit drugs and med-ications have been chosen to correspond with the degree ofimpairment comparable to a BAC of 0.2 g/L. The possibil-ity of a developed tolerance to a drug detected should not beconsidered; likewise, tolerance is not considered for alcohol.

The high BACs, in addition to concentrations of medica-tions and illicit drugs, show that a considerable number ofthe drivers killed had used high doses of one or more of thesubstances that are prohibited in combination with driving.The latter was in particular demonstrated for alcohol, be-cause BACs of more than 5 times the legal limit were detectedin more than 80 percent of these cases (Table 1).

Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

7:30

14

Nov

embe

r 20

14

530 Christophersen and Gjerde

For medications and illicit drugs, the prevalence of caseswith blood drug concentrations of at least 5 times the legallimit was lower than for that of alcohol. Many of the driverswho had high blood concentrations of different medications,which all have warning labels, obviously had used dosages thatwere not in accordance with therapeutic recommendation. Insome of these cases, the medications may have been boughtillegally. We have earlier documented that the drivers appre-hended by the police for driving under the influence of drugsand who had used BZDs represented a younger age group thanpatients using BZDs in accordance with the prescription, whowere mostly over the age of 50 (Christophersen and Mørland2008). Correlations have also been documented between theamounts of seized BZDs, for the individual compounds, andthe prevalence of the same BZDs detected in blood samplesamong apprehended drivers (Bogstrand et al. 2011), implyingthat the BZDs detected in many of the fatal accident casesamong the younger drivers may represent illegal acquisitions.

Age and Gender

A disproportionately large number of the drivers killed wereunder the age of 25 (26%), though only 8 percent of drivers innormal Norwegian traffic are younger than 25 years (Vagane2005). Surprisingly, there was no difference in prevalence ofalcohol and other drugs (in total) among the drivers killedbetween the different age groups up to the age of 55 (Fig-ure 2). However, some differences were found among the typesof drugs detected, when comparing the prevalence of alcohol,medications, and illicit drugs, between the different age groups,particularly for single-vehicle accidents (Figure 2). Larger dif-ferences were recorded comparing female and male drivers.Female drivers represented only 19 percent of all drivers killed,though the prevalence of females in normal Norwegian trafficis approximately 29 percent (Gjerde et al. 2013). Alcohol ordrugs were found in blood samples from 24 percent of femaledrivers and 43 percent of male drivers. Thus, it appears thatfemale drivers take greater care to follow the law, especiallywith regard to the use of alcohol before driving.

Critical Accident Time Periods

The highest prevalence of alcohol and drugs was found amongthose who died on weeknights and on weekend nights (Fig-ure 4), in particular for single-vehicle accidents. This findingshould be taken into consideration by the police when plan-ning for traffic controls and by other authorities working onpreventive actions.

Risk Evaluation

The present work represents a descriptive epidemiologicalstudy, lacking information to perform a culpability study.However, risk evaluation-based case-control studies have pre-viously been performed, where some of the presented datahave been used (Gjerde et al. 2011, 2013).

In addition to the use of alcohol and drugs, another factoraffecting the risk of traffic accident involvement is a risk-taking

lifestyle, including an unsafe driving style. An earlier study hasdocumented that some groups of drunk drivers killed in roadaccidents showed risky lifestyle behaviors, such as recklessdriving combined with alcohol abuse (Laapotti and Keskinen2008). Some studies have also documented high numbers ofrepeat arrests among drivers apprehended by the police due tothe suspicion of driving under the influence of alcohol or drugs(Christophersen et al. 2002; Jones 2005). A high prevalence ofprevious arrests due to alcohol- or drug-impaired driving wasalso found among drivers killed in traffic accidents, wherealcohol or drugs were present in the drivers’ blood samplescollected at the time of the accidents (Christophersen et al.2007). These findings support the impression that many killeddrivers represented a group with pronounced risk-taking be-haviors.

Penalties

Norway has relatively severe sentences for driving under theinfluence of alcohol and drugs. The sentencing is based uponthe BAC and drug concentrations. However, if a high concen-tration of a psychoactive medication is detected in the bloodsample from a suspected impaired driver and the driver hasa valid prescription, the result of the clinical examination forimpairment, performed at the time of blood sampling, is alsotaken into consideration when assessing for possible impair-ment. This is also the case when high concentrations of psy-choactive substances without legislative limits are detectedand in cases of multiple drug use. The penalty may be a fee,a prison sentence, or both, in addition to withdrawal of thedriving license for up to 2 years.

The high prevalence of alcohol and other drugs amongdrivers killed, together with the documentation of previousfrequent re-arrest rates among drivers who died in alcohol-or drug-related crashes, shows that other ways of sentenc-ing should be offered to this groups of drivers to reduce thenumber of future incidences of driving under influence. Suchmeasures could be the use of alcohol ignition interlock devicesand drug intervention programs.

Acknowledgments

We thank the staff at the Division of Forensic Medicine andDrug Abuse Research for analyzing all of the samples andBartho van der Linden and Bjørn Skuterud for managementof the database.

References

Barbone F, McMahon AD, Davey PG, et al. Association of road-trafficaccidents with benzodiazepines. Lancet. 1998;352:1331–1336.

Berghaus G, Ramaekers JG, Drummer OH. Demands on scientific stud-ies in different fields of forensic medicines and forensic science: trafficmedicine impaired driver: alcohol, drugs, diseases. Forensic Sci Int.2007;165:233–237.

Bernhoft IM, Hels T, Lyckegaard A, Houwing S, Verstraete AG. Preva-lence and risk of injury in Europe by driving with alcohol, illicit drugsand medicines. Procedia Soc Behav Sci. 2012;48:2907–2916.

Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

7:30

14

Nov

embe

r 20

14

Road Accidents in Norway 531

Blomberg RD, Peck RC, Moskowitz H, Burns M, Fiorentino D. TheLong Beach/Fort Launderdale relative risk study. J Safety Res.2009;40:285–292.

Bogstrand ST, Middelkoop G, Christophersen AS. Trends in am-phetamines and benzodiazepine use arrested for drug impaired drivingin Norway during 2000–2009. Norsk Epidemiol. 2011;21:61–66.

Borkenstein RF, Crowther RF, Shumare RP, Ziel WB, Zylman R. Therole of the drinking driver in traffic accidents (the Grand Rapid Study)Blutalkohol. 1974;11(suppl 1):7–13.

Christophersen AS, Hammer T, Mørland J. Fatal accident drivers withearlier arrests due to drugged driving. Paper presented at: JointMeeting of the International Association of Forensic Toxicologists(TIAFT), and International Council on Alcohol, Drugs and Traf-fic Safety (ICADTS) and featuring the 8th Ignition Interlock Sym-posium (IIS) T2007; August 26–30, 2007; Seattle, WA. Available at:http://www.icadts2007.org/print/abstract.pdf. Accessed March 10,2014.

Christophersen AS, Mørland J. Frequent detection of benzodiazepinesin drugged drivers in Norway. Traffic Inj Prev. 2008;9:98–104.

Christophersen AS, Skurtveit S, Grung M, Mørland M. Rearrest ratesamong Norwegian drugged drivers compared with drunken drivers.Drug Alcohol Depend. 2002;66:85–92.

Drummer OH, Gerostamoulos J, Batziris H, et al. The involvementof drugs in drivers of motor vehicle killed in Australian road traf-fic crashes. Accid Anal Prev. 2004;36:239–248.

European Monitoring Centre for Drugs and Drug Addiction. Drug Use,Impaired Driving and Traffic Accidents. Lisbon, Portugal: EuropeanMonitoring Centre for Drugs and Drug Addiction; 2008. Availableat: http://www.emcdda.europa.eu/attachements.cfm/att 65871 ENInsight8.pdf. Accessed February 13, 2013.

Gjerde H, Beylich K-M, Mørland J. Incidence of alcohol and drugs in fa-tally injured car drivers in Norway. Accid Anal Prev. 1993;25:479–483.

Gjerde H, Christophersen AS, Normann PT, Assum T, Øiestad EL,Mørland J. Norwegian road side survey of alcohol, and drugs use bydrivers (2008–2009). Traffic Inj Prev. 2013;14:443–452.

Gjerde H, Normann PT, Christophersen AS, Samuelsen SO, MørlandJ. Alcohol, psychoactive drugs and fatal road traffic accidents in Nor-way: a case control study. Accid Anal Prev. 2011;43:1197–1203.

Holmgren A, Holmgren P, Kugelberg FC, Jones AW, Ahlner J. Predom-inance of illicit drugs and poly-drug use among drugs impaired driversin Sweden. Traffic Inj Prev. 2007;8:361–367.

Holmgren P, Holmgren A, Ahlner J. Alcohol and drugs in drivers in fa-tally injured in traffic accidents in Sweden during the years 2000–2002.Forensic Sci Int. 2005;151:11–17.

Houwing S, Hagenzieker M, Mathijssen R, et al. Prevalence of Alcoholand Other Psychoactive Substance in Drivers in General Traffic. PartI: General Results. Leidschendam, The Netherlands: SWOV Institutefor Road Safety Research; 2011. DRUID Deliverable D 2.2.3.

Isalberti C, van der Linden T, Legrand SA, et al. Prevalence of Alcoholand Other Psychoactive Substances in Injured and Killed Drivers. Ghent,Belgium: Ghent University; 2011. DRUID Deliverable 2.2.5.

ISO Central Secretariat. ISO/IEC 17025:2005 General requirementsfor the competence of testing and calibration laboratories. Geneva,Switzerland: International Organization for Standardization; 2005.

Jones AW. Driving under the influence of drugs in Sweden with zeroconcentration limits in blood for controlled substances. Traffic InjPrev. 2005;6:317–322.

Jones AW, Kugelberg FC, Holmgren A, Ahlner J. Five years update onthe occurrence of alcohol and other drugs in blood samples fromdrivers killed in road-traffic crashes in Sweden. Forensic Sci Int.2009;186:56–62.

Kelly E, Darke S, Ross J. A review of drugs use and driving: epidemiology,impairment, risk perceptions. Drug Alcohol Rev. 2004;23:1394–1398.

Laapotti S, Keskinen E. Fatal drink-driving accident of young adult andmiddle-aged males—a risky driving style or risky lifestyle? Traffic InjPrev. 2008;9:195–200.

Legrand SA, Gjerde H, Isalberti C, et al. Prevalence of alcohol, il-licit drugs and psychoactive medicines in killed drivers in four Eu-

ropean countries. Int J Inj Contr Saf Promot. 2014 (in press). DOI:10.1080/17457300.2012.748809

Legrand SA, Isalberti C, der Linden TV, et al. Alcohol and drugs inseriously injured drivers in six European countries. Drug Test Anal.2013;5:156–165.

Mørland J. Driving under the influence of non-alcoholic drugs. ForensicSci Rev. 2000;12:80–105.

Mørland J, Steentoft A, Wiese Simonsen K, et al. Drugs related to motorvehicle crashes in northern European countries: a study of fatallyinjured drivers, Accid Anal Prev. 2011;43:1920–1926.

Movig KL, Mathijssen MP, Nagel PH, et al. Psychoactive substancesuse and risk of motor vehicle accidents. Accid Anal Prev. 2004;36:631–636.

National Criminal Investigation Service. Narkotika-og Dopingstatis-tikk 2010 [Drug and doping statistics 2010]. Oslo, Norway: Na-tional Criminal Investigation Service; 2011. Available at: https://www.politi.no/vedlegg/lokale vedlegg/kripos/Vedlegg 1106.pdf. Acces-sed March 10, 2014.

Norwegian Accreditation. Available at: http://www.akkreditert.no/en/.Accessed March 10, 2014.

Norwegian Institute of Public Health. Drug Statistics 2010. Oslo,Norway: Norwegian Institute of Public Health; 2011. Available at:http://www.fhi.no/artikler/?id = 91671. Accessed March 7, 2013.

Norwegian Road Traffic Act. Available at: http://lovdata.no/dokument/NL/lov/1965-06-18-4. Accessed March 10, 2014.

Orriols L, Delorme B, Gadegbeku B, et al. Prescription medicines andthe risk of road traffic crashes: a French registry-based study. PLoSMed. 2010;7:e1000366.

Orriols L, Salmi LR, Philip P, et al. The impact of medicinal drugs ontraffic safety: a systematic review of epidemiological studies. Pharma-coepidemiol Drug Saf. 2009;18:647–658.

Phillips DP, Brewer KM. The relationship between serious injuryand blood alcohol concentration (BAC) in fatal motor vehicle ac-cidents: BAC = 0.01% is associated with significantly more dan-gerous accidents that BAC = 0.00%. Addiction. 2011;106:1614–1622.

Pil K, Raes E, Verstraete A. The toxicological challenges in the Euro-pean research project DRUID. Forensic Sci Int Suppl Ser. 2009;1:29–32.

Ramaekers JG, Moeller MR, van Ruitenbeek P, Theunissen EL, Schnei-der E, Kauert G. Cognition and motor control as a function of �9-THC concentration in serum and oral fluid: limits of impairment.Drug Alcohol Depend. 2006;85:114–122.

Ravera S, van RN, de Gier JJ, de Jong-van den Berg LT. Road trafficaccidents and psychotropic medication use in The Netherlands: a case-control study. Br J Clin Pharmacol. 2011;72:505–513.

Smink BE, Ruiter B, Lusthof KJ, de Gier JJ, Uges DR, Egberts AC.Drug use and the severity of a traffic accident. Accid Anal Prev.2005;37:427–433.

Statistics Norway. Persons Killed or Injured, by Group of Road-User andSex. 2001–2010. Oslo, Norway: Statistics Norway; 2011. Available at:http://www.ssb.no/vtuaar en/arkiv/tab-2011-06-10-03-en.html. Ac-cessed February 12, 2013.

Stoduto G, Vingilis V, Kapur B, Wen-Jenn S, McLellan BA, Liban CB.Alcohol and drugs use among motor vehicle collision victims admittedto a regional trauma unit: demographic, injury, and crash character-istics. Accid Anal Prev. 1993;25:411–420.

Vagane L. Den Norske Reiselivsvaneundersøkelsen [The Norwegian TravelSurvey]. Oslo, Norway: Institute of Transport Economics; 2005.

Verster JC, Mets MAJ. Psychoactive Medication and Traffic Safety. IntJ Environ Public Health. 2009;6:1041–1054.

Vindenes V, Jordbru D, Knapskog AB, et al. Impairment based legislativelimits for driving under the influence of non-alcohol drugs in Norway.Forensic Sci Int. 2012;219:1–11.

Walsh JM, de Gier JJ, Christophersen AS, Verstraete AG. Drugs anddriving. Traffic Inj Prev. 2004;5:241–253.

Walsh JM, Verstraete AG, Huestis MA, Mørland J. Guidelines for re-search on drugged driving. Addiction. 2008;103:1258–1268.

Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

7:30

14

Nov

embe

r 20

14