| An evaluation of the Standardised Field Sobriety Tests for the detection of impairment associated with cannabis with and without alcohol (2006) | |||||||||||||||
|
|||||||||||||||
Abstract | |||||||||||||||
| Reports indicate that in Victoria, New South Wales and Western Australia, 23.5% of drivers in fatal accidents had consumed drugs other than alcohol, and that 29.1% of drivers had a Blood alcohol contentration (BAC) level of 0.05% or higher. Alcohol has been detected in combination with drugs in almost 10% of cases. Cannabis was most prevalent among drugs other than alcohol detected in specimens (13.5%) (Drummer et al., 2003). The combination of drugs as an influence on road traffic accidents is becoming a growing concern and research has been conducted to identify how these drugs impair performance. Krueger & Vollrath (2000) reported that recent consumption of cannabis improved lane positioning; however, when combined with alcohol, lane position deviated, and participants drove faster. The consumption of low dose and high dose cannabis alone has also been associated with an increase in vehicle lane weaving (straddling solid and barrier lines) (Papafotiou, 2004b). Furthermore, a trend towards greater braking latency after consumption of higher doses of cannabis has been reported (Ligouri et al., 1998). Generally, alcohol has been reported to increase hazardous simulated driving, and cannabis has been reported to slow a driver’s speed (Stein et al., 1983). The findings of several studies have directly suggested that the effect of Tetrahydrocannabinol (THC) consumption on driving performance may be greater for nonregular cannabis users than for regular cannabis users (Marks & MacAvoy, 1989; Wright and Terry, 2002; Papafotiou, 2004c). Wright and Terry (2002) also provide evidence to suggest that regular cannabis users may develop cross-tolerance to the effects of drugs and alcohol. In Victoria, Australia, Standardised Field Sobriety Tests (SFSTs) have been introduced as means of testing for impairment in drivers who have consumed drugs other than alcohol. The use of SFSTs, although designed for the detection of alcohol-intoxicated drivers (up to 0.08%), has been implemented in programs for the detection of drugs other than alcohol. To date, one study exists that has evaluated the sensitivity of the SFST battery to predict drug intoxication and driving impairment. This project assessed the relationship between each individual sobriety test (Horizontal Gaze Nystagmus, Walk and Turn, and One Leg Stand test) and individual scored signs of the SFSTs with the administration of cannabis (Papafotiou et al., 2004a). Papafotiou et al. (2004a) found that, unlike in the case of alcohol where the HGN test is reported to be the best test for impairment associated with the administration of alcohol, in the case of cannabis the test best related to impairment is the One Leg Stand test. This finding highlights the need for additional research into the relationship between performance on the SFST battery and drug intoxication (drugs other than alcohol). The present study had several aims: to examine the effects of cannabis and cannabis together with alcohol on driving performance; to examine the effects of cannabis and alcohol on Standardised Field Sobriety Tests (SFSTs) performance; to examine the efficiency of SFSTs to predict driving performance associated with the administration of cannabis and alcohol; to examine any differences between the effects of cannabis and alcohol on performance in regular cannabis users and non-regular cannabis users; and to examine any differences between SFSTs ratings by researchers (Swinburne University) and SFSTs ratings by police officers (Victoria Police) in order to identify the inter-rater reliability of SFSTs. The project consisted of two parts: cannabis (0% THC, 1.8% THC and 3% THC) with low dose alcohol (0.03% BAC); and cannabis (0% THC, 1.8% THC and 3% THC) with high dose alcohol (0.05% BAC). Each part was made up of six randomized, double-blind sessions and both utilised the same experimental design and procedure. The total sample comprised 80 individuals; 31 female and 49 male. Age varied between 21 and 35 years ( M = 26.45, SD = 5). Part one was comprised of 40 participants; 15 females and 25 males. Of these participants, 24 were regular cannabis users and 16 non-regular cannabis users. In part two, 40 participants included 16 females and 24 males. Of those participants, 24 were regular users, and 16 were non-regular cannabis users. In each experimental session, after the administration of cannabis and alcohol, participants were asked to perform a driving simulator task and the SFSTs. Blood samples were taken throughout each session in order to determine the level of drug in plasma associated with observed impairment. The results of the present study are consistent with past research in which the level of THC detected in blood is higher when cannabis (THC) is consumed with alcohol, than when cannabis is consumed only (Lukas & Oronzco, 2001). Also consistent with past research, regular cannabis users reported higher levels of THC in plasma than non-regular users (Papafotiou, et al., 2004c). Data from the driving simulator indicated that the mean number of times the errors of 'straddling the solid line', 'straddling the barrier line', 'insufficient stopping clear space' and 'slower initial speed on freeway' occurred increased with the consumption of THC (with or without alcohol). This suggests that individuals who have consumed cannabis are more likely to drive with two or more wheels of the vehicle moving over lines marked out for traffic moving in the same direction, or lines marked out for traffic moving in the opposite direction. In addition, individuals who have smoked cannabis are more likely to have insufficient clear space between their own vehicle and the vehicle in front of them if required to stop (driving too close to the vehicle in front). Although the consumption of THC was associated with slower initial speed when entering the freeway, this does not appear to be related to safe driving. The observation of an increase in straddling barrier and solid lines, and insufficient stopping clear space, indicates that driving slowly is associated with deficiencies in cognitive processes necessary to safely drive a motor vehicle. In the present study, the driving variables 'straddling the barrier line' and 'insufficient stopping clear space' were impaired by THC consumption. These errors occurred significantly more often in conditions that included the administration of alcohol. In addition, the error 'straddling the solid line' occurred significantly more often in the 0.05% BAC condition than the 0.03% BAC condition. With respect to impairment associated with the administration of alcohol (irrespective of THC consumption), several driving variables were impaired. 'Dangerous skidding', 'unsafe stopping distance between vehicles', 'violation of the speed limit', 'collisions', 'straddling the barrier line', 'wandering', and 'insufficient clear space' occurred significantly more frequently in alcohol conditions than placebo conditions. The results also indicated that the errors 'straddling the barrier line' and 'unnecessary stopping' occurred significantly more often in the 0.05% BAC condition than the 0.03% BAC condition. There were also significant interactions observed between the THC condition and the alcohol condition. In other words, the results showed that THC consumed with alcohol was significantly more impairing than consuming either drug alone. Specifically, while there were significantly more skidding errors in the alcohol condition compared to the placebo condition, the greatest number of errors occurred when alcohol was consumed with THC. In addition, while there were more inappropriate signaling errors in the alcohol condition compared to placebo condition, the greatest number of errors occurred when alcohol was combined with THC. Generally, regular cannabis users displayed more driving errors than non-regular cannabis users. Regular cannabis users drove fast, displayed dangerous skidding, stopped unnecessarily and were involved in collisions more often than non-regular users. A difference in SFSTs performance was observed between the THC only condition and the THCwith alcohol condition. During the Horizontal Gaze Nystagmus (HGN) test, the percentage of individuals exhibiting each sign of the HGN test increased with the administration of low THC and high THC when compared to placebo. When THC was administered together with alcohol, the percentage of individuals exhibiting each sign increased. The data reported that the sign Head Movements/Jerks (HMJ) was exhibited in a high percentage of individuals compared to traditionally scored signs of the HGN test. Furthermore, the use of the SFSTs (three test battery), when scoring HMJ in the HGN test, classified the highest percentage of participants as impaired than any other individual SFST test sign (high THC with alcohol condition). There was no significant difference between the presence of signs observed during the Walk and Turn (WAT) test during the THC only condition or the THC with alcohol condition (with the exception of Steps off the Line (SOL)). Analysis of the One Leg Stand (OLS) test suggests there was a significant relationship between all scored signs of the OLS test and the THC only condition and THC with alcohol condition. Specifically, more errors were observed during the OLS test the higher the dose of THC administered. Overall SFST performance was significantly related to the administration of THC only and THC with alcohol. When THC was administered with alcohol, the percentage of individuals classified as impaired using the SFSTs more than doubled than when no alcohol was administered. These results highlight the sensitivity of SFSTs to test for the presence of alcohol. Overall SFST test scores including HMJ increased the percentage of individuals classified as impaired. At best, the use of SFSTs resulted in the correct classification of up to 73.9% of participants as either impaired or not impaired on driving (this rate was reported after the administration of low dose THC). In this case, the best test of impairment was the OLS test. Driving ability was impaired when the level of THC in plasma was between 3.1 and 11.2 ng/ml. Regular users performed worse on the driving task than non-regular users. When this occurred, the level of THC in plasma in regular users was between 3.6 ng/ml and 13 ng/ml (higher than the level of THC detected in nonregular users). Results from the present study support a high level of agreement between researcher ratings (Swinburne University) and police officer ratings (Victoria Police) on impairment using the SFSTs. There was statistically significant agreement between raters on the presence of 17 out of 23 SFSTs errors. Primarily, non-agreement was recorded for errors that involve eye signs, such as Lack of Smooth Pursuit (LSP), and Nystagmus at 45 degrees (N45). These errors were reported to be present more often by the researchers than the police officers. This non-agreement can be explained by the police officers having to score SFST performance via video footage. With respect to identifying impairment on each individual test (HGN, WAT and OLS) and impairment on the SFST battery as a whole, there was statistically significant agreement in all cases between police officer ratings and the researcher ratings. This result supports that, when administered correctly (by a trained individual), the SFSTs are reliable tests of impairment, and that it is unlikely that scores on the SFSTs would alter if administered by a different rater. Overall, the results suggest that SFSTs have reliable scoring procedures and that test results are accurate and replicable. In conclusion, the use of the SFSTs is a moderately good predictor of driving impairment and the consumption of THC only and THC together with alcohol. In the absence of reliable and accurate physical tests of THC plasma levels and driving ability, the SFSTs can provide relevant information concerning drug intoxication and driver fitness, in particular associated with the consumption of cannabis. [Executive summary] | |||||||||||||||
Publication details | |||||||||||||||
| |||||||||||||||