Effects of MDMA (ecstasy), and multiple drugs use on (simulated) driving performance and traffic safety
Department of Psychology, University of Groningen, Psychopharmacology (Berl). 2004 Jan 9
Results
Three subjects decided to withdraw from the experiment half-way through. In total, 20 participants in the experimental group, 15 male and five female, completed all drug and non-drug conditions; their average age was 27 years (SD 4.5) and average mileage 17,000 km/year (SD 14,000). On average, they had used MDMA 26 times (SD 27, range 2–100) for a total of 40 tablets (SD 45, range 2–150). Sixty percent had driven a car at least once while under the influence of MDMA, on average 8 times (SD 14, range 0–50). Most participants from the experimental group had ample experience with other drugs, and combining MDMA with these drugs. Results on self-experienced psychological and physiological questionnaires were comparable with the study of Davison and Parrott (1997). Physiological effects such as high heart rate, unsteadiness and headache were also occasionally reported, as were increased heart rate, hyperthermia, dilated pupils, a dry mouth, increased sweating, and tingling skin. Subjectively experienced effects were increased perception of sound, touch, and colour, especially after multi-drug use. Common effects such as being more talkative after MDMA were also confirmed. In the control group, 13 non-drug-users completed the non-drug test rides; average age 24 years (SD 5.5), average mileage 14,000 km/year (SD 4000).
MDMA dosage and other drugs used.
The average consumed dosage MDMA was 59 mg (SD 22, range 25–98 mg MDMA). This is a relatively low dose; the normal average dose is 120 mg (Logan and Couper 2001). After the first MDMA ride, the majority
(70% ) took more MDMA before the multi-drug ride. All participants additionally took other drugs in that time span, marihuana (80% ) and alcohol (90% ) most commonly. In Table 1 all substances used by the participants are listed. In the experimental group breath analysis showed a blood alcohol concentration of 0.00‰ for all participants in the non-drug and MDMA condition. In the multi-drug condition, the average BAC was 0.39‰ (SD 0.39, range 0.00–1.09). In the control group, with the exception of one subject in whom breath analysis indicated 0.11‰, the measured BAC was always zero.
Performance measures
The standard deviation of lateral position (swerving) increased significantly [F(1,18 )=5.5, P=0.031] with 0.03 m in the multi-drug condition compared to the non-drug condition, both in the built-up area and on the motorway. From non-drug to MDMA to multi-drug, an increase in average speed was found in the built-up area of, respectively, 2.5 and 7.1 km/h. Standard deviation of the driving speed during the test rides also significantly increased from non-drug to MDMA to multi-drug [F(1,19)=4.8, P=0.040 and F( 1,19 )=8.83, P=0.008]. Control and experimental groups did not differ in the non-drug conditions [F(1,31 )=1.95, NS]. The gap acceptance test in the multi-drug condition showed significantly smaller accepted gaps than in the non-drug condition [F(1,18 )=4.75, P=0.043]. Although the average time-headway to the car in front was smaller in the multi-drug condition compared with the non-drug condition, this effect was not significant. Differences between control group and experimental group were not significant either, whereas average time headway was smaller in the control group, as was variance in headway. On the busy part of the motorway, cars ahead stopped to a standstill twice. In both conditions, movement time and reaction time were determined. No effects on the averages of parameters were found: 0.76 s and 1.64 s, respectively. However, standard deviation in RT between participants increased from 0.35 (non-drug condition) to 0.52 (MDMA condition) to 0.88 (multi-drug condition). On average, the standard deviation in RT for the control group was 0.39. The ultimate indicator of driving safely is the absence of crashes. Although crashes are relatively rare in real traffic, in the simulator on the motorway section where lead cars suddenly braked, sometimes accidents happened. None of the control group participants had ever had a crash. During two of the 20 non-drug rides with the experimental group crashes occurred; while under the influence of MDMA, the simulator car collided with another car (100% increase) 4 times. Under the influence of multiple drugs, participants crashed 5 times
Physiological measures
The pattern of average heart rate in beats per minute is remarkably similar in all conditions for different sections (city driving, gap acceptance, car-following, the trafficlight scenario, driving on the motorway and a rest, see Fig. 1). After drug use, heart rate was significantly increased, after multiple drugs on average up to 18 beats per minute. Heart rate was also higher at the points where a decision had to be taken, in particular at the two gap acceptance tasks. Data for heart rate variability in the 0.10 Hz frequency band (Fig. 2) were normalised (Van Roon 1998 ), reflecting mental effort (Mulder 1992). This was clearly the case at the first gap acceptance test, where energy was suppressed (indicative of increased mental effort). The second gap acceptance test showed less consistent results; however, during rest, energy increased again. The reduced variability in the multi-drug condition may suggest either increased effort or a ceiling effect, heart rate in this condition being extremely high.
Self-report measures
Self-reported effort increased on a scale from 0 to 150 from 40.2 in the non-drug condition to 47.6 in the MDMA condition to 50.7 in the multi-drug condition [drug versus non-drug is significant; F(1,29)=5.02, P=0.037]. The control group indicated an average of 49.1 [F(1,31 )=1.83, NS]. On the driving quality scale, ranging from +100 (extremely well) to –100 (extremely badly), the control group on average rated normal driving (0.2). The experimental group in the non-drug condition indicated driving as better than normal (+16.5), under the influence of MDMA about normal (+3.8 ), and under the influence of multiple drugs lower than normal (–4.6). The difference between non-drug state and the two drugs conditions is significant [F(1,19)=5.13, P=0.035].
Urine analyses
Urine analyses were performed to confirm the selfreported drug use (Table 1) and detect non-reported psychoactive substances. In general, screening results of urine samples in the multi-drug condition matched the reported drug intake rather well: the presence of LSD, psilocybin, cocaine and amphetamine in addition to MDMA was confirmed. In the multiple-drug condition, three participants provided a urine sample positive for MDEA in addition to MDMA. In three other participants, in both drug conditions, urine analysis revealed the presence of amphetamine or MDEA instead of MDMA. Two participants had a positive MDMA result in urine in the non-drug condition. One had reported the use of MDMA on the day before, the other did not report the use of MDMA in the week before the test ride. One subject was positive for amphetamine. Cannabis was reported in 30% of participants in both the drug and non-drug conditions. Urine analysis revealed cannabis use in 60% of cases, but since marihuana can be detected in urine for several weeks in regular users, this was not proof of recent use. Two participants who admitted intake of cocaine before the multiple-drug test ride, were also positive for cocaine before the MDMA test ride; one even showed a positive result in the non-drug condition. For the participants in the control group, all screens were negative.
Discussion
The main conclusion must be that from a health point of view, using MDMA is not recommended, for many reasons. Most striking of all was the number of crashes in the multiple drugs session (25% of all rides) and in the MDMA session (20% ), at night and early in the morning. However, the experimental drug group had crashes in the non-drug control rides as well (10% ). The control group had no crashes. This finding strongly suggests that the group of MDMA users differs from the control group in more ways than just drug use. In a birth cohort study of 907 young New Zealanders (Horwood and Fergusson 2000; Fergusson and Horwood 2001), the relationships between cannabis use and alcohol use were established with traffic accident rates. Most of the elevated risks among subgroups of the population under investigation were found not to be due to cannabis use per se, and only partly to drink driving in itself. As a subgroup, the cannabis users in the study of Fergusson and Horwood (2001 ) were prone to engage in drink driving and other risky/illegal driving practices, associated with attitudes encouraging driving violations. They suggest that these driving behaviours are frequently part of more general tendencies towards risky and unsafe driving, among which is ill-judged decision making on the road (Horwood and Fergusson 2000). The present study strongly supports their findings, from a different methodological
angle. The present study has some limitations. Firstly, the study was carried out in a driving simulator, for ethical reasons a necessary restriction, but for validity reasons constraining the ecological power. However, we are convinced that this necessary “choice” has been successfully defended in the past (De Waard et al. 1999b). Secondly, a quasi-experimental design could not be avoided, which brought along diminished experimental control, albeit with a gain in realism. The relevance of the findings in terms of the type of effects is a natural given for science, but also has practical consequences for policy makers. Any investment in regulating drug (and drink) driving among young people should be accompanied by similar investments in reducing unsafe driving in all other aspects.
Obviously different drugs have different effects, so how do these penalties correspond to that 8)