The effect of lowering the legal drink-drive limit on the toxicological findings in

driver fatalities: A comparison of two jurisdictions

ABSTRACT: In December 2014, the legal blood alcohol limit for drivers in both Scotland

and New Zealand was reduced from 80 to 50 mg/100 mL. This paper reports a

retrospective study comparing changes in the toxicological findings in deceased drivers

and motorcyclists before and after the limit change in both jurisdictions. A year of fatal

motor vehicle crashes prior to and following the limit change is examined for both

countries. In Scotland, there was an increase in drug prevalence among fatally injured

drivers and motorcyclists, with the use of all drug groups increasing after the limit

change, with the exception of cannabinoids. In New Zealand, there was a reduction in

cases involving drugs only, but increases in the numbers of deceased drivers and

motorcyclists positive for alcohol only, and co-using alcohol and drugs.

KEYWORDS: Forensic science, drink-drive limit; fatality; drug driving; motor vehicle

crash; road traffic accident

Page 1 of 24

The role of alcohol as a major factor in motor vehicle crashes (MVCs) has been firmly

established (1). Driving is a complex task, requiring the co-operation of several different

cognitive and psychomotor functions at once (2) (psychomotor capability, physical space

awareness, reaction times, etc. (3)) and drivers may be significantly impaired by the use

of alcohol (4) and/or other psychoactive substances (5).

The Long Beach/Fort Lauderdale study demonstrated a relative crash risk due to alcohol

showing a dose–response relationship beginning at blood alcohol concentrations (BACs)

of 40–50 mg/100 mL and increasing exponentially at BACs of 100 mg/100 mL or greater

(1). Driving skills are significantly impaired at BACs in the 50–80 mg/100 mL range (6),

and even BACs as low as 20 mg/100 mL can affect driver performance by slowing

reaction times (7).

The first country to introduce a legal per se BAC limit was Norway, in 1936 (8). The limit

was set at 50 mg/100 mL, and has since been reduced to 20 mg/100 mL (9). Sweden

reduced their legal BAC limit from 80 to 50 mg/100 mL in 1957 (10,11), and this has

been followed by the same reduction in numerous other countries (12). Worldwide, legal

BAC limits now range from 20–150 mg/100 mL with some countries adopting a zero-

tolerance approach (13).

The effectiveness of BAC limit reduction laws has been investigated in recent

years by several researchers, using different evaluation methods and analysis

procedures (14). The most common method is to use official statistics, i.e., numbers of

arrests, collisions, and fatalities (15). Using these official statistics, and different study

designs, researchers have been able to draw conclusions about the effects of various

legislative interventions on alcohol-impaired driving (7), across a range of jurisdictions

(11). These include Australia (6,16,17), Austria (18), Brazil (19), Denmark (20), Hong

Kong (21), Japan (22), Norway (23), Sweden (24) and the USA (14,25), with the majority

of studies appearing to show reductions in alcohol-related collisions, injuries and/or

fatalities (15). However, the effectiveness measures and the analysis procedures have

Page 2 of 24

varied from study to study (11), with researchers often relying on proxy measures of

alcohol involvement (7) such as nighttime crashes compared to daytime crashes (11)

rather than on measured BACs. Previous evaluation studies where BACs have been

measured have also focussed solely on alcohol findings, without considering the use of

drugs by drivers. It is important to take into account drug findings, as drug use by drivers

is increasing (26), accompanied by a widespread belief that drug driving is acceptable

(27) and that crash risk is not significantly increased by drug use (28). Moreover, a

possible consequence of reinforcing anti-drink-drive messages with this kind of legislative

intervention, is a difference in the perceived or real risk of being apprehended by the

police when driving under the influence of drugs (29) compared to alcohol. This may

result in a change in the number of fatal MVCs where drivers are positive for drugs. For

instance, in a study undertaken in Norway before (1989–1990) and after (2001–2002) the

reduction in BAC limit from 50 to 20 mg/100 mL, the proportion of fatally injured drivers

positive for drugs and alcohol increased from 12.4% to 22.8%. During the same period,

the proportion of fatally injured drivers positive for alcohol alone dropped from 32.9% to

23.9% (30).

On 1 December 2014 the legal BAC limit was reduced in New Zealand from 80 to

50 mg/100 mL by means of the Land Transport Amendment Act (no 2) 2014. On 5

December 2014 the same reduction in BAC limit was introduced in Scotland via the Road

Traffic Act 1988 (Prescribed Limit) (Scotland) Regulations 2014. Both countries ran high-

profile media campaigns informing members of the public about the change (31,32). In

New Zealand, the new limit applies to drivers aged 20 and over; there is now a zero-

tolerance approach to drink driving for drivers under 20 years of age. Drug driving

legislation in Scotland is governed by section 4 of the Road Traffic Act 1988 and is an

impairment offence, with no statutory defence for the use of prescription medications by

an impaired driver. In New Zealand, section 11A of the Land Transport Act 1998 sets out

a zero-tolerance offence of driving while impaired with evidence of a qualifying drug in

the blood. The Act includes a defence for drivers who can prove that they were using a

Page 3 of 24

qualifying drug in accordance with a current prescription and following the instructions of

a healthcare professional.

In this paper, we compare changes in the toxicological findings in fatally injured drivers

and motorcyclists from Scotland and New Zealand before and after the introduction of

the new BAC limits. These two jurisdictions represent an interesting comparison for this

type of study, as both countries have similar populations (33,34), and cultures, but

different drug and alcohol use patterns (35,36). As the change in legislation happened in

both countries at the same time, this study should avoid confounding factors such as

broad social changes in alcohol use and driving behaviour, which may have occurred

during this period (15). To minimize potential biases resulting from variations in roadside

testing policies e.g., increased use of checkpoints or random breath testing, screening

only alcohol-negative drivers for drugs (37), administrative differences, etc., this study

focusses on fatally injured drivers and motorcyclists (25). Another reason that fatal MVCs

were examined rather than non-fatal MVCs, is because the declared impetus behind both

countries’ campaigns was to reduce road deaths (38,39).

To the authors’ knowledge, this is the first study to examine the overall toxicological

findings in driver fatalities before and after a BAC limit change, including drug findings.

The change in legislation at the same time represents a rare opportunity to compare

changes in the toxicological findings around the lowering of a BAC limit across two

jurisdictions. Although comparisons of alcohol- and drug-impaired driving among

countries can be complicated by differing methods in each country of measuring and

reporting (30), this study examines data resulting from two very similar MVC-

investigation strategies.

Methods

The Toxicology Laboratory of the Department of Forensic Medicine & Science (FMS) at

the University of Glasgow, receives forensic toxicology cases for fatal MVCs from all

regions of Scotland except the far North. The Toxicology Laboratory of the Institute of

Page 4 of 24

Environmental Science and Research (ESR) in Wellington, New Zealand carries out all

forensic toxicology analyses for fatal MVCs in New Zealand.

Fatal driver and motorcyclist MVC cases received by both laboratories between

December 2013 and December 2015 were selected for this study (covering the 12-month

periods before and after the BAC limits changed). Deceased cyclists, pedestrians and

passengers were not included. In Scotland, blood samples were initially analysed for

alcohol by headspace gas chromatography with flame ionisation detection (GC-FID) and

screened for a panel of eight drug groups by enzyme-linked immunosorbent assay

(ELISA). Any drug groups presumptively positive following ELISA were confirmed by solid-

phase extraction (CLEAN SCREEN® DAU cartridges) followed by gas chromatography-

mass spectrometry (GC-MS) or liquid chromatography-tandem mass spectrometry (LC-

MS/MS). All cases included in this study were also analysed for paracetamol by high-

pressure liquid chromatography (HPLC) and for basic drugs by GC-MS.

In New Zealand, solid-phase extraction (TOXI-TUBES®) followed by a liquid

chromatography-time-of-flight-mass spectrometry (LC-TOF-MS) screen was carried out,

which can detect 180 drugs of abuse and medications (40). This was combined with an

ELISA screen for cannabinoids, positive results from which were verified by LC-MS/MS.

More details on the analytical methods used are given in Table S1 of the Supplementary

Information.

Most samples analysed were post-mortem blood, with ante-mortem hospital samples

analysed in five cases from Scotland and nine cases from New Zealand. Analysis was

carried out in both countries as part of routine casework. Details of the post-mortem

interval in each case were not routinely collected. The laboratories also do not routinely

receive information on crash culpability as part of death investigations. It should also be

noted that not all fatal MVCs result in an autopsy; the decision to request an autopsy

varies between police/Coronial areas and is influenced by the individual case

circumstances as well as economic considerations (41).

Page 5 of 24

Similarly, only cases where full toxicological analyses were completed were included in

the scope; cases where the submitting authority requested only alcohol, or only certain

types of drugs (e.g., drugs of abuse only or prescription drugs only) were removed. In

Scotland, there were 106 driver and motorcyclist fatalities identified from the in-house

database. In 17 cases (16.0%), no or limited toxicological analysis was requested, leaving

a total of 89 cases within the scope of the study: 38 occurred before the BAC limit

change and 51 occurred afterwards.

In New Zealand, 326 driver and motorcyclist fatalities were identified. In 57 cases

(17.5%) limited toxicological analysis was requested, leaving a total of 275 cases within

the scope of the study: 129 occurred before the BAC limit change and 146 occurred

afterwards.

Results & Discussion

Geographical area

Fig. 1 shows the regions of Scotland and New Zealand where the MVCs reported in this

study took place over both years. There were small changes in the distribution of crashes

among the geographical regions between the two years (2013–14 and 2014–15) and not

all regions of Scotland were represented. In addition, some regions of Scotland reported

fatal MVCs in one year, but not in the other. However, the regions in both countries with

the highest number of crashes (Edinburgh and the Bay of Plenty) remained the same

throughout the study. Interestingly, neither area represents the most populous region of

either country. However, the New Zealand Transport Agency (NZTA) report that people

living in the Bay of Plenty were least likely to recognise the risk of drink-driving (42).

Summary

A summary of the cases identified in this study is shown in Table 1. According to the

table, in both countries, there was an apparent increase in the number of fatal MVCs

after the BAC limit change. However, official statistics (Fig. 2) demonstrate that while

there was an increase in overall road traffic fatalities in New Zealand from 2014 to 2015,

Page 6 of 24

a decrease was observed in Scotland. Therefore, it appears that in New Zealand, the

number of fatal MVCs was not reduced by a change in the legal BAC limit. It is unclear

what other factors may have caused the increase in road traffic fatalities in New Zealand,

but they may include increased numbers of drivers and distances being driven, changes

in the use of seatbelts, mobile phone use, speeding, etc.

<<Figure 1 here>>

It should be noted that although in most jurisdictions evidence for a positive impact on

road safety, such as a reduction in fatal MVCs, is observed when legal BAC limits are

lowered, this has not always been the case (15). Fig. 2 shows numbers of overall road

traffic fatalities including pedestrians, cyclists and passengers for both countries,

whereas Table 1 reports driver fatalities only.

<<Figure 2 here>>

According to Table 1, the number of cases in all categories increased in New Zealand,

over the two-year period, except drugs-only MVCs, where a decrease from 56 to 48 cases

was observed. In Scotland, there were small changes in the numbers of MVCs across

each category, with the number of drugs-only MVCs doubling from 10 to 21 over the

period studied.

Age and gender

The ages and genders of the deceased drivers and motorcyclists in this study are shown

in Table 2. Consistent with previous studies, the majority of fatalities were male.

Vehicle type

Page 7 of 24

Fig. 3 shows the breakdown of vehicle types in this study. As can be seen from the figure,

car drivers made up the majority of fatalities in both countries, followed by motorcyclists

and other vehicle types. MVCs involving quad bikes and tractors were only included in

this study if they occurred on a public road. There were only small changes in the

percentages of drivers, motorcyclists and other vehicle drivers across the two-year

period in both countries.

Alcohol

A summary of the BAC results from this study is shown in Table 3. Eighteen of the 93

BACs measured in this study were <30 mg/100 mL (N = 16, New Zealand and N = 2,

Scotland), and therefore we cannot exclude the possibility that some of the alcohol

detected was a result of post-mortem bacterial production (43). In Scotland, a sample

was deemed positive for alcohol at BAC ≥10 mg/100 mL. In New Zealand, a sample was

deemed positive for alcohol at BAC ≥6 mg/100 mL. In some cases, BACs between the

analytical cut-offs for the two methods were observed: 8 of the BACs for New Zealand fell

between 6 and 10 mg/100 mL. These BACs are included in the dataset without

modification. No subtractions from the BACs for analytical uncertainty (a practice known

as ‘guard banding’ (44)) were made in either jurisdiction.

<<Figure 3 here>>

Two-tailed t-tests conducted in MS Excel demonstrated that the differences between the

mean BACs before and after the limit change for both countries were not significant.

Table 3 shows that the percentage of total fatal MVC cases where the driver or

motorcyclist was over the relevant BAC limit increased in both countries following the

change, with a more pronounced increase noted in New Zealand. This is consistent with

the previous observation that the lowering of a BAC limit to 50 mg/100 mL increases the

proportion of drivers over the limit (30). In addition, in a study following trends in fatal

MVCs around the reduction in BAC limit in Denmark from 80 to 50 mg/100 mL, the

proportion of fatalities where the driver had a BAC ≥50 mg/100 mL was higher in the

year after the limit change than in the preceding five years (20). Although, a recent study

Page 8 of 24

showed a decrease in BACs ≥50 mg/100 mL in fatally injured drivers following a

reduction in BAC limit in Hong Kong (21).

Fig. 4 shows area charts of the BAC data obtained in this study from (a) New Zealand and

(b) Scotland. The charts represent the cumulative percentages of the BAC data,

demonstrating how the BAC distribution in fatally injured drivers and motorcyclists has

altered after the change in BAC limit for each country.

<<Figure 4 here>>

From Fig. 4 (left) it can be seen that in New Zealand, the proportion of fatally injured

drivers and motorcyclists with high BACs appears to have increased. Before the BAC limit

change ~50% of fatally injured drivers and motorcyclists had a BAC <100 mg/100 mL,

following the change, only ~30% of BACs were <100 mg/100 mL. It is possible that a 50

mg/100 mL limit does not have a particular impact on those people who tend, once they

pass a certain initial BAC, to drink very heavily (6). Due to the small numbers of MVC

cases involving alcohol in Scotland it is difficult to draw any further meaningful

conclusions.

The difference in the profile of BAC results between the two jurisdictions may be due to

differing social norms in the two countries about the acceptable amount of drinking

before driving (23). Reports in the UK media suggest that attitudes towards drink-driving

in Scotland are overwhelmingly negative (45). In New Zealand, the NZTA report that peer

pressure and social drinking remain strong influences on drivers (42). Although high-

profile campaigns about the change in BAC limit ran in both countries, researchers have

noted that there is little evidence that mass communication campaigns significantly alter

people’s drinking and driving behaviour (7). Another factor responsible for the

differences in BAC distribution between the two countries may be differences in the

prevalence of alcohol use disorders; previous work has suggested a link between drinking

and driving with a BAC of 150 mg/100 mL and a clinical diagnosis of alcoholism (46).

Page 9 of 24

Table 3 shows that ~50% (range, 44–52%) of alcohol-positive drivers had a BAC ≥150

mg/100 mL in both countries before and after the BAC limit change. However, this does

not appear to correlate with alcohol use disorder prevalence estimates for the two

jurisdictions, as measured by the World Health Organisation, which were 1.52% (female)

and 6.42% (male) in the UK (data for Scotland alone not available), and 2.20% (female)

and 3.50% (male) in New Zealand (47).

Drugs

Drugs were categorised into ‘families’ as previously described (48). Table 4 shows the

identity of the drugs in each family in this study. It should be noted that these results do

not account for polydrug use (49). Due to differences in the analyses undertaken by the

two countries, not all blood drug concentrations were available. In cases where

emergency medical treatment was indicated in the case records, findings of morphine,

fentanyl, ketamine, midazolam or lidocaine were excluded.

Fig. 5 shows a bar chart of the drug types detected in this study before and after the BAC

limit change for both countries. A case was considered ‘positive’ for a drug type if it

contained one or more member(s) of that group, including metabolites. It should be

noted that many of the drivers and motorcyclists from both countries were positive for

more than one drug group. Fig. 5 shows that cannabinoids were the most common drug

finding in fatally injured drivers in New Zealand, whereas in Scotland, cannabinoids were

overtaken by opioids after the BAC limit change. In New Zealand, the prevalence of

cannabis use by fatally injured drivers increased over the two-year period, although it

appears to have remained constant in Scotland. Note that some of the cases from

Scotland were positive only for carboxy-THC, an inactive metabolite of cannabis that can

remain in the blood for some time after cannabis is used. Conversely, the prevalence of

opioid use (mainly codeine) by fatally injured drivers has markedly increased in Scotland,

whilst falling in New Zealand. In a recent survey of 2434 New Zealanders, the proportion

of respondents driving after taking drugs was highest for opiates, followed by synthetic

cannabis, and cannabis (50). A study of living drugged drivers in the UK reported

Page 10 of 24

cannabis as the most common finding (49), however opioids may be more likely to be

involved in fatalities than cannabis due to their pharmacological effects (51).

Benzodiazepine prevalence increased in both jurisdictions, with diazepam being the most

common benzodiazepine encountered in both datasets (all benzodiazepine-positive MVCs

in Scotland contained diazepam). Clonazepam was found in five cases in New Zealand

but was not seen in Scotland. As Table 4 demonstrates, this study does not include

newer benzodiazepines such as diclazepam and etizolam.

There was only a small change in the prevalence of stimulant use by fatally injured

drivers and motorcyclists in Scotland, and the prevalence in New Zealand remained

steady over the two-year period.

<<Figure 5 here>>

Other prescription drug use was recorded as being higher in fatally injured drivers in New

Zealand compared to Scotland. This likely reflects differences in the screening

approaches used by the two laboratories, including the instrumentation and analytical

cut-offs applied (see Table S1 of the Supplementary Information). Prescription drug use is

a particularly complex issue when assessing its involvement in impaired driving (52),

with some prescription drugs having no effect or resulting in driving improvement, and

others causing significant impairment (53). Those prescription drugs known to cause the

greatest driving impairment (opioids and benzodiazepines) were placed in other drug

families. Zopiclone was only detected in New Zealand, and was considered part of the

other prescription drug group, although unlike some of the drugs in that group, it is

known to cause driving impairment (3). Many of the prescription drugs detected were

antidepressants; whilst some antidepressants are known to have a sedating effect (e.g.,

amitriptyline), it has also been shown that depressed individuals receiving long-term

antidepressant treatment exhibit inferior driving skills compared to healthy controls (54).

The interpretation of prescription drugs found in fatally injured drivers is further

complicated as it is often unknown whether substances were taken on prescription or

Page 11 of 24

recreationally (e.g., gabapentin) (52). Information on prescriptions was not available in

every case, and drivers may also hold a valid prescription for a drug, but take more than

directed.

The prevalence of over-the-counter drugs increased in Scotland over the two-year period.

All MVC cases containing over-the-counter drugs in Scotland before the BAC limit change,

and seven of the eight afterwards were positive for paracetamol (78%). In New Zealand,

15 of the 18 MVC cases that were positive for over-the-counter drugs before (83%) and

14 of the 15 MVC cases afterwards (93%) contained paracetamol. In New Zealand and

Scotland, five and three cases respectively, that were positive for paracetamol were also

positive for codeine, suggesting the use of codeine–paracetamol preparations, which are

not known to cause impairment (55). Whilst over-the-counter drugs such as paracetamol

and ibuprofen are not associated with driving impairment or increased crash risk, over-

the-counter antihistamines such as diphenhydramine can significantly impair driving

ability (56).

Differences in the drug findings between the two countries could be due to differences in

social attitudes to driving under the influence of drugs. In surveys of attitudes to drug

driving in New Zealand (57) and Scotland (58), respondents’ from both jurisdictions

perceived that their chances of being caught drug driving were lower than for drink

driving. Moreover, in Scotland, most people surveyed did not believe that their driving

was adversely affected by drug use (58).

Other drugs of abuse such as novel psychoactive substances (e.g., synthetic

cannabinoids), hallucinogens (lysergic acid diethylamide, NBOMe compounds, etc.),

inhalants and gamma-hydroxybutyrate (GHB) were not part of routine toxicological

testing in either country in this study.

Limitations

There are a number of important limitations to this study. Firstly, it was not possible to

account for other concomitant policies, variations in enforcement activities or public

awareness, changes in other laws or trends in alcohol/drug consumption and motor

Page 12 of 24

vehicle utilization that might affect the occurrence of, and toxicological findings in, traffic

fatalities during the study period (11,19). Therefore, differences regarding the effect of

the new legislation between the two regions studied may be explained by divergent

police enforcement or other supporting measures, such as media coverage and

preventive actions in each locality (19) or other factors we were not able to measure.

There are a number of factors that affect drink-driving behaviour, including increased

perceived risk of prosecution; changes in behaviour such as not drinking at all before

driving, using a taxi, and an increase in designated sober drivers (6). Although we have

attempted to minimise the effects of differing roadside testing policies by examining

fatally injured drivers rather than impaired drivers, it is possible that such policies

affected driver behaviour during the period studied.

This study does not take into account culpability for MVCs or distinguish between single-

and multiple-vehicle crashes; the presence of drugs and/or alcohol in fatally injured

drivers therefore is not the same as a causal factor (59).

As mentioned above, assessing the impact of prescription and over-the-counter drugs on

driving is complex. Risk profiles vary across drug classes and can vary within classes and

even by drug, depending on dosage, duration of therapy, and underlying driver

characteristics (60). The use of such drugs may even improve driving, as there is strong

evidence that unrelieved pain may decrease psychomotor and cognitive performance

(55). In addition, many of the deceased drivers in this study were positive for multiple

substances, creating difficulty in determining whether or not certain drugs contributed to

any impairment at the time of the MVC (52). We were also unable to assess the

contribution of drug–drug interactions in polydrug use cases (61), and there is the

possibility that some drugs might have degraded in post-mortem blood between

sampling and testing (9).

In terms of the alcohol findings, this study demonstrates that BACs at and around the

legal BAC limits for New Zealand and Scotland do not represent the majority of fatal

Page 13 of 24

MVCs. The mean BACs measured (Table 3) were not within the 50–80 mg/100 mL range,

either before or after the BAC limit change, and the numbers of BACs within the range

50–80 mg/100 mL were very small (<3) in both countries.

The extent to which our findings can be generalised to other countries may be limited by

the small numbers of fatal MVCs recorded. There was also no control group unaffected by

the reduced BAC limits, consequently the observed changes in toxicological findings may

not necessarily be a result of the legal amendments (23). The dataset for Scotland is not

complete, as it excludes the far Northern regions, and the small number of fatal MVCs in

Scotland means that the trends observed around the change in BAC limit should be

treated with caution. Finally, datasets from both countries did not include all fatal driver

and motorcyclist MVCs during the period examined, due to some instructing authorities

requesting only limited toxicological analyses or no autopsy being carried out.

Conclusion

This study has compared changes in the full toxicological findings in fatally injured

drivers and motorcyclists in two jurisdictions, in the years before and after a change in

legal blood alcohol limit for driving. In Scotland, there was an increase in cases positive

for drugs, with the use of all drug groups increasing after the BAC limit change, with the

exception of cannabinoids. In New Zealand, there was a reduction in cases involving

drugs only, but increases in the numbers of deceased drivers and motorcyclists positive

for alcohol only, and co-using alcohol and drugs (cannabinoids, benzodiazepines, and

other prescription drugs). In terms of the BAC findings, there was no statistically

significant difference in mean BACs over the two-year period for either country, and BACs

at and around the legal limits did not represent the majority of fatal MVCs occurring.

However, some differences in BAC distribution were observed: in New Zealand there was

an increase in the proportion of fatally injured drivers and motorcyclists with high BACs

after the limit change. Due to the small numbers of fatalities in Scotland, detailed

conclusions could not be drawn.

Page 14 of 24

Future work could examine data from both jurisdictions in later years following the BAC

limit change, as studies have suggested that some effects wear off with time (16,19). The

study could also be extended to include other fatally injured road users (e.g., passengers

and pedestrians) or examine the impact of new drug-driving legislation involving drug

limits to be introduced in Scotland in 2019.

References

1. Blomberg RD, Peck RC, Moskowitz H, Burns M, Fiorentino D. The Long Beach/Fort Lauderdale

relative risk study. J Saf Res 2009;40(4):285–92.

2. Raes E, Verstraete AG. Drug-impaired driving. In: Jamieson A, Moenssens A, editors. Wiley

Encyclopedia of Forensic Science. Chichester: John Wiley & Sons, 2009;877–83.

3. Gunja N. In the Zzz zone: the effects of z-drugs on human performance and driving. J Med

Toxicol 2013;9(2):163–71.

4. Lapham SC. The limits of tolerance: convicted alcohol-impaired drivers share experiences

driving under the influence. Perm J 2010;14(2):26–30.

5. Burch HJ, Clarke EJ, Hubbard AM, Scott-Ham M. Concentrations of drugs determined in blood

samples collected from suspected drugged drivers in England and Wales. J Forensic Legal Med

2013;20(4):278–89.

6. Brooks C, Zaal D. Effects of a reduced alcohol limit for driving. Canberra: Federal Office of Road

Safety; 1993.

7. DeJong W, Hingson R. Strategies to reduce driving under the influence of alcohol. Ann Rev

Public Health 1998;19(1):359–78.

8. Andenaes J. Drinking-and-driving laws in Scandinavia. Scand Stud Law 1984;28:11–23.

9. Valen A, Bogstrand ST, Vindenes V, Gjerde H. Toxicological findings in suspected drug-impaired

drivers in Norway — Trends during 1990–2015. Forensic Sci Int 2017;280:15–24.

10. Laurence Ross H. The Scandinavian myth: the effectiveness of drinking-and-driving legislation

in Sweden and Norway. J Leg Stud 1975;4(2):285–310.

11. Fell JC, Voas RB. Reducing the illegal blood alcohol limits for driving: effects on traffic safety.

In: Verster JC, Pandi-Perumal SR, Ramaekers JG, de Gier JJ, editors. Drugs, driving and traffic

safety. Basel: Birkhäuser, 2009;415–37.

Page 15 of 24

12. Dunbar JA, Penttila A, Pikkarainen J. Drinking and driving: choosing the legal limits. Br Med J

1987;295(6611):1458–60.

13. http://apps.who.int/gho/data/view.main.54600 (accessed July 25, 2017).

14. Kaplan S, Prato CG. Impact of BAC limit reduction on different population segments: a Poisson

fixed effect analysis. Accid Anal Prev 2007;39(6):1146–54.

15. Mann RE, Macdonald S, Stoduto G, Bondy S, Jonah B, Shaikh A. The effects of introducing or

lowering legal per se blood alcohol limits for driving: an international review. Accid Anal Prev

2001;33(5):569–83.

16. McLean AJ, Kloeden CN, McColl RA, Laslett R. Reduction in the legal blood alcohol limit from

0.08 to 0.05: Effects on drink driving and alcohol-related crashes in Adelaide. In: McLean AJ,

Kloeden CN, editors. Thirteenth International Conference on Alcohol, Drugs and Traffic Safety;

1995 Aug 13-18; Adelaide. Adelaide: NHMRC Road Accident Research Unit.

17. Homel R. Drink-driving law enforcement and the legal blood alcohol limit in New South Wales.

Accid Anal Prev 1994;26(2):147–55.

18. Bartl G, Esberger R. Effects of lowering the legal BAC limit in Austria. In: Laurell H, Schlyter F,

editors. Fifteenth International Conference on Alcohol, Drugs and Traffic Safety; 2000 May 21-

26; Stockholm. Ann Arbor, MI: ICADTS.

19. Andreuccetti G, Carvalho HB, Cherpitel CJ, Ye Y, Ponce JC, Kahn T, et al. Reducing the legal

blood alcohol concentration limit for driving in developing countries: a time for change? Results

and implications derived from a time–series analysis (2001–10) conducted in Brazil. Addiction

2011;106(12):2124–31.

20. Bernhoft IM, Behrensdorff I. Effect of lowering the alcohol limit in Denmark. Accid Anal Prev

2003;35(4):515–25.

21. Cheng W-C, Dao K-L. The occurrence of alcohol/drugs by toxicological examination of selected

drivers in Hong Kong. Forensic Sci Int 2017;275:242–53.

22. Nagata T, Setoguchi S, Hemenway D, Perry MJ. Effectiveness of a law to reduce alcohol-

impaired driving in Japan. Inj Prev 2008;14(1):19–23.

23. Assum T. Reduction of the blood alcohol concentration limit in Norway—effects on knowledge,

behavior and accidents. Accid Anal Prev 2010;42(6):1523–30.

24. Norström T, Laurell H. Effects of lowering the legal BAC-limit in Sweden. In: Mercier-Guyon C,

editor. Fourteenth International Conference on Alcohol, Drugs and Traffic Safety; 1997 Sep 21-

26; Annecy. Annecy: CERMT.

Page 16 of 24

25. Hingson R, Heeren T, Winter M. Lowering state legal blood alcohol limits to 0.08%: the effect

on fatal motor vehicle crashes. Am J Public Health 1996;86(9):1297–9.

26. Kahn CA. Commentary: drugs and driving. Ann Emerg Med 2015;66:670–2.

27. Neale J, McKeganey N, Hay G, Oliver J. Recreational drug use and driving: A qualitative study.

Edinburgh; 2000.

28. Neale J. Drug driving in Scotland: prevalence and correlates amongst drug users entering

treatment. Int J Drug Policy 2004;15(1):27–35.

29. Gjerde H, Normann PT, Pettersen BS, Assum T, Aldrin M, Johansen U, et al. Prevalence of

alcohol and drugs among Norwegian motor vehicle drivers: a roadside survey. Accid Anal Prev

2008;40(5):1765–72.

30. Sweedler BM, Stewart K. Worldwide trends in alcohol and drug impaired driving. In: Verster JC,

Pandi-Perumal SR, Ramaekers JG, de Gier JJ, editors. Drugs, driving and traffic safety. Basel:

Birkhäuser Basel, 2009;23–41.

31. https://www.nzta.govt.nz/safety/driving-safely/alcohol-and-drugs/drink-driving-advertising/

limits/ (accessed December 29, 2016).

32. http://dontriskit.info/drink-driving/tv-campaign/ (accessed January 2, 2017).

33. http://www.gov.scot/Topics/People/Equality/Equalities/PopulationMigration (accessed August

24, 2016).

34. http://www.stats.govt.nz/tools_and_services/population_clock.aspx (accessed December 29,

2016).

35. http://www.health.govt.nz/nz-health-statistics/health-statistics-and-data-sets/drug-use-data-

and-stats (accessed November 10, 2016).

36. http://www.isdscotland.org/Health-Topics/Drugs-and-Alcohol-Misuse/Publications/ (accessed

January 14, 2017).

37. Horyniak D, Dietze P, Lenton S, Alati R, Bruno R, Matthews A, et al. Trends in reports of driving

following illicit drug consumption among regular drug users in Australia, 2007–2013: has

random roadside drug testing had a deterrent effect? Accid Anal Prev 2017;104:146–55.

38. Scottish Government. Reducing the drink driving limit in Scotland: a Scottish Government

consultation. Edinburgh; 2012.

39. Office of the Minister of Transport. Safer Journeys – lowering the legal alcohol limits for adult

drivers. Wellington: New Zealand Government; 2013.

Page 17 of 24

40. Hamnett HJ, Kappatos D, Zubkov O, Poulsen H. Toxicology case studies from New Zealand

involving LC-TOF-MS. In: Twenty Second International Symposium on the Forensic Sciences of

the Australian and New Zealand Forensic Science Society; 2014 Aug 31 - Sep 4; Adelaide.

41. Breen JM, Naess PA, Gjerde H, Gaarder C, Stray-Pedersen A. The significance of preexisting

medical conditions, alcohol/drug use and suicidal behavior for drivers in fatal motor vehicle

crashes: a retrospective autopsy study. Forensic Sci Med Pathol doi: 10.1007/s12024-017-9934-

x. Epub 2017 Nov 28.

42. http://www.transport.govt.nz/research/roadsafetysurveys/publicattitudestoroadsafety-survey/

2015-public-attitudes-to-road-safety-survey-results-2/ (accessed January 2, 2017).

43. Kugelberg FC, Jones AW. Interpreting results of ethanol analysis in postmortem specimens: A

review of the literature. Forensic Sci Int 2007;165(1):10–29.

44. Risk Solutions. Evaluation of the new drug driving legislation, one year after its introduction.

Warrington: Risk Solutions; 2017.

45. http://www.bbc.co.uk/news/uk-scotland-34999421 (accessed January 2, 2017).

46. Vingilis E. Drinking drivers and alcoholics. Are they from the same population? In: Smart RG,

Glaser FB, Israel Y, Kalant H, Popham RE, Schmidt W, editors. Research advances in alcohol and

drug problems. Boston, MA: Springer, 1983;299–342.

47. http://www.who.int/substance_abuse/publications/atlas_report/profiles/en/ - N (accessed July

30, 2017).

48. Drummer OH, Yap S. The involvement of prescribed drugs in road trauma. Forensic Sci Int

2016;265:17–21.

49. Rooney B, Gouveia GJ, Isles N, Lawrence L, Brodie T, Grahovac Z, et al. Drugged drivers blood

concentrations in England and Wales prior to the introduction of per se limits. J Anal Toxicol

2017;41(2):140–5.

50. Starkey NJ, Charlton SG, Malthotra N, Ameratunga S. Prevalence of psychotropic drug use prior

to driving. J Transp Health 2017;4:108–17.

51. Galski G, Williams JB, Ehle HT. Effects of opioids on driving ability. J Pain Symptom Manage

2000;19(3):200–8.

52. Robertson RD, Mainegra Hing M, Pashley CR, Brown SW, Vanlaar WGM. Prevalence and trends

of drugged driving in Canada. Accid Anal Prev 2017;99:236–41.

53. Wolff K, Brimblecombe R, Forfar JC, Forrest AR, Gilvarry E, Johnston A, et al. Driving under the

influence of drugs: report from the expert panel on drug driving. London: Department for

Transport; 2013.

Page 18 of 24

54. Wingen M, Ramaekers JG, Schmitt JAJ. Driving impairment in depressed patients receiving

long-term antidepressant treatment. Psychopharmacology 2006;188(1):84–91.

55. European Monitoring Centre for Drugs and Drug Addiction. Drug use, impaired driving and

traffic accidents. Luxembourg: EMCDDA; 2014.

56. Verster JC, Volkerts ER. Antihistamines and driving ability: evidence from on-the-road driving

studies during normal traffic. Ann Allergy Asthma Immunol 2004;92(3):294–304.

57. Malhotra N, Starkey NJ, Charlton SG. Driving under the influence of drugs: Perceptions and

attitudes of New Zealand drivers. Accid Anal Prev 2017;106:44–52.

58. Myant K, Hope S, McIntosh J, O'Brien T, McKeganey N, Stradling S. Illicit drugs and driving.

Edinburgh: Transport Research Planning Group; 2006.

59. Longo MC, Hunter CE, Lokan RJ, White JM, White MA. The prevalence of alcohol, cannabinoids,

benzodiazepines and stimulants amongst injured drivers and their role in driver culpability: Part

II: the relationship between drug prevalence and drug concentration, and driver culpability.

Accid Anal Prev 2000;32(5):623–32.

60. Kelley-Baker T, Waehrer G, Pollini RA. Prevalence of self-reported prescription drug use in a

national sample of U.S. drivers. J Stud Alcohol Drugs 2017;78(1):30–8.

61. Steuer AE, Eisenbeiss L, Kraemer T. Blood alcohol analysis alone versus comprehensive

toxicological analysis – Systematic investigation of missed co-ingested other drugs in suspected

alcohol-impaired drivers. Forensic Sci Int 2016;267:52–9.

62. http://www.pharmpress.com/product/MC_BNF/british-national-formulary (accessed August 1,

2017).

Page 19 of 24

TABLE 1—Summary of the cases (fatally injured drivers and motorcyclists) examined in this

study.

Category N (%) New Zealand

N (%) Scotland

Before the change in BAC limit

129 cases 38 cases

Negative 43 (33) 22 (58)Alcohol only* 14 (11) 2 (5)Alcohol and drug(s) 16 (12) 4 (11)Drug(s) only 56 (43) 10 (26)After the change in BAC limit

146 cases 51 cases

Negative 50 (34) 21 (41)Alcohol only 21 (14) 3 (6)Alcohol and drug(s) 27 (18) 6 (12)Drug(s) only 48 (33) 21 (41)

*In Scotland, a sample was deemed positive for alcohol at BAC ≥10 mg/100 mL. In New Zealand, a sample was

deemed positive for alcohol at BAC ≥6 mg/100 mL.

Page 20 of 24

TABLE 2—Summary of the ages and genders of the deceased drivers in this study.N New

Zealand Before

New Zealand After

Scotland Before

Scotland After

Male 99 119 35 43Female 30 27 3 8Mean age* 45 41 43 42Median age 42 37 44 42Age range† 16–94 17–86 19–71 17–86

*Differences in mean ages between the two years not significant (p = 0.08 for NZ and p = 0.68 for

Scotland). †Legal ages for driving are 15 in New Zealand and 17 in Scotland.

Page 21 of 24

TABLE 3—Summary of the BAC data obtained in this study.New Zealand Before

New Zealand After

Scotland Before

Scotland After

N alcohol-positive cases 30 48 6 9 Mean BAC (mg/100 mL)*

126 ± 90†138 ± 82 153 ± 87 117 ± 60

Median BAC (mg/100 mL) 129 158 159 148 Range (mg/100 mL) 6–304 6–317 10–256 18–184N with BAC ≥150 mg/100mL 15 25 3 4 % Alcohol-positive cases with BAC ≥150 mg/100mL 50

5250 44

N over BAC limit at the time of MVC 21 37 5 7 % Alcohol-positive cases over BAC limit 77% 77% 83% 78% % Total MVC cases over BAC limit 18% 25% 13% 14%N alcohol and drugs cases 16 27 4 6 Mean BAC (mg/100 mL)‡

131 ± 96126 ± 80 135 ± 90 125 ± 53

Median BAC (mg/100 mL) 129 130 145 148 Range (mg/100 mL) 6–304 6–243 10–241 38–177

*Differences in mean BACs between the two years not significant (p = 0.57 for NZ and p = 0.39 for Scotland).

†Standard deviation. ‡Differences in mean BACs between the two years not significant (p = 0.85 for NZ and p =

0.86 for Scotland).

Page 22 of 24

TABLE 4—Drugs detected in this study.Family Both countries New Zealand only Scotland onlyBenzodiazepines

DiazepamTemazepam*

Clonazepam DesmethyldiazepamOxazepam

Cannabinoids† Δ9-Tetrahydrocannabinol (THC)

— 11-Nor-9-carboxy-Δ9-tetrahydrocannabinol (carboxy-THC)‡

Opioids CodeineMorphineDihydrocodeineTramadolMethadone

OxycodoneDextropropoxyphene§

Heroin((

Other prescription drugs

AmitriptylineCitalopramGabapentin

AmiodaroneAmlodipineCarbamazepineCyclizineDiltiazemDoxepinFluoxetineImipramineMetoprololOrphenadrinePromethazineQuetiapineQuinineSelegilineSertralineVenlafaxineWarfarinZiprasidone§

Zopiclone

Atenolol MirtazapineNortriptyline¶

TrazodoneValproic acid

Over-the-counter drugs

Paracetamol DextromethorphanChlorpheniramineLoratadineCetirizineIbuprofen

Diphenhydramine

Stimulants Ephedrine/pseudoephedrine

Benzylpiperazine (BZP)MethamphetamineMethylphenidateTrifluoromethylphenyl-piperazine (TFMPP)

AmphetamineBenzoylecgonineCocaethyleneCocaineEcgonine methyl ester 3,4-Methylenedioxyamphetamine (MDA)3,4-Methylenedioxymethamphetamine (MDMA)

*Temazepam was detected only as a metabolite of diazepam in Scotland, but as a drug in its own right in New

Zealand. †Synthetic cannabinoid compounds were not included in this study. ‡This metabolite is not targeted in

NZ. §Not prescribed in the UK (62). ((Confirmed by the presence of 6-monoacetylmorphine in urine in one case

only. ¶Detected only as a metabolite of amitriptyline.

Page 23 of 24

FIG. 1—Areas in Scotland (left) and New Zealand (right) where the fatal MVC cases included in this

study occurred over the two years. Shading indicates the number of MVCs. The highest number of

MVCs occurred in Edinburgh (N = 20) and Bay of Plenty (N = 46). The Highlands and Islands of

Scotland are not included in the dataset.

FIG. 2—Trends in road traffic fatalities in New Zealand and Scotland 2005–2016.

FIG. 3—Breakdown of the vehicle types involved in the cases examined. Other vehicle types

included large goods vehicles, vans, minibuses, quad bikes, tractors and scooters.

FIG. 4—Area charts showing the cumulative frequency of BACs in this study in (left) New Zealand

and (right) Scotland. Graphs drawn using R.

FIG. 5—Breakdown of the drug types detected in blood samples from deceased drivers and

motorcyclists in fatal MVCs before and after the BAC limit changes. For more information see Table

4.

Page 24 of 24