Opiates and Driving Ability
1 Overview
1.1 Heroin (diamorphine) is a narcotic analgesic drug, one of a family of opiate drugs which also includes morphine, codeine and pharmaceutical products such as dextropropoxyphene and dihydrocodeine. All these drugs act on the same neurochemical system, relieving pain, or the distress caused by pain, and relieving stress so the user leaves behind his or her cares and worries.
1.2 The effects of heroin can last for 2-6 hours, depending on dosage and the tolerance of the individual user. The effects of a recreational dose include an initial "rush" (effect of diamorphine proper), followed by a feeling described as being wrapped in cotton wool (once heroin is metabolised into morphine). Addicts will feel an initial sense of relief or release, but with tolerance and physical dependence, the drug simply helps them feel normal, rather than intoxicated.
1.3 Drug interactions: Opiates interact with alcohol to increase the depressive effects on respiration and mood. Cannabis can potentiate the analgesic effect of opiates.
1.4 The effects of drugs on driving have been studied using laboratory tests of psychomotor performance and cognitive function, simulator studies, and assessment of road accidents and driving records. IDMU is currently engaged in research
2. Laboratory Studies of Psychomotor & Cognitive Function
2.1 In a study of cancer patients, Vainio et al found no significant effect of morphine on "...intelligence, vigilance, concentration, fluency of motor reactions, or division of attention. Of the neural function tests, reaction times (auditory, visual, associative), thermal discrimination, and body sway with eyes open were similar in the two groups; only balancing ability with closed eyes was worse in the morphine group. These results indicate that, in cancer patients receiving long-term morphine treatment with stable doses, morphine has only a slight and selective effect on functions related to driving."
2.2 Pickworth et al found hydromorphone to have no effect on circular lights, digit symbol substitution, and serial math tasks or card-sorting tasks. Oxycodone, a mu-opioid receptor agonist, was found to cause "increased reaction time and impaired vigilance, attention, body balance and coordination of extraocular muscles".
2.3 Hill & Zacny reported "Psychomotor impairment was ... and absent with morphine (which)... produced dose-dependent decreases in pupil size.." In an earlier study, Zacny et al found "morphine had no effect on psychomotor functioning." in healthy non-using volunteers.
2.4 Sjogren et al, studying patients receiving high-dose morphine therapy, found "Vigilance/attention, psychomotor speed, and working memory were significantly impaired in chronic nonmalignant pain patients." In a study of cognitive and psychomotor function, O"Neill et al reported: "Morphine had one major effect, which was to increase the accuracy of responding on the choice reaction time task, at every assessment. Morphine produced some sporadic effects in other tests and an increase in subjective calmness. These data show that oral morphine may enhance performance in some measures of cognitive function", in an earlier study of cancer pain management O"Neill had reported "opioids do have effects on cognitive and psychomotor function, and although many of these effects diminish once the patient is on a stable dose... the relationship between measurable effects and the performance of everyday tasks such as driving is unclear."
2.5 Walker et al found morphine and codeine "...did not affect performance on Maddox-Wing, digit-symbol substitution, coordination, auditory reaction, reasoning, and memory tests. Dose-related decreases in pupil size (miosis) were observed following codeine and morphine. ...These results suggest that oral codeine and morphine ... have only modest effects on mood, produce few side effects, and do not impair performance." Zacny et al found "morphine produced minimal psychomotor impairment.", and that "morphine "...did not affect performance on the Digit Symbol Substitution Test." However by contrast Petry at al found "Morphine produced significant dose-dependent effects in DSST performance... and pupil diameter." in occasional drug users, whereas Zacny et al, studying healthy non-using volunteers, concluded "Some aspects of psychomotor performance (reaction time, Digit Symbol Substitution Test and Maddox Wing) were impaired by morphine; however, eye-hand coordination was not. Miosis was induced by morphine. Most effects of morphine were dose-related, some effects peaked soon after morphine injection (e.g., increased stimulated and high ratings) and dissipated gradually, whereas other effects did not peak until later into the session (sedation or exophoria). Our results are fairly consistent with other studies examining morphine effects in healthy volunteers, and also indicate that the profile of morphine effects differ between healthy volunteers and those with a history of opiate dependence."
2.6 Hanks et al, studying healthy volunteers, found "morphine produced significant impairment at 1 hour on tests of secondary memory retrieval (delayed word recall and picture recognition sensitivity). CFFT was reduced for the whole observation period (6 h) achieving statistical significance at 4 hours. Morphine 15 mg produced a significant improvement in accuracy on the choice reaction time test at the 2, 4 and 6 h assessments. These results show minimal impairment of cognitive and psychomotor function after single oral doses of morphine and with possible improvement in one test." However, also with healthy volunteers, Kerr et al reported "morphine... caused significant impairments of some but not all elements of cognitive and motor function. The time needed to encode and process serially presented verbal information increased and the ability to maintain low consistent levels of force decreased during the morphine infusion. We also assessed verbal recall 3 hours after the morphine and saline infusions. Delayed recall of information presented during the morphine infusion was significantly impaired. Our results demonstrate that morphine can interfere with cognitive and motor performance at plasma drug concentrations within the usual therapeutic range." In a general review of the effects of painkillers on occupational health, Payne concluded "all classes of analgesics may impair... neuropsychiatric functioning, which may influence job performance in specific instances."
2.7 Bourke et al found "Morphine did not impair psychomotor function (Trieger Dot Test (TDT) and... Continuous Performance Test (CPT))" Saddler at al compared the effects of alcohol and morphine, finding "Ethanol produced a significantly greater deterioration in motor skills", with no effect of morphine on reaction time.
2.8 Bradley & Nicholson studied the effects of codeine on visuo-motor coordination, dynamic visual acuity, critical flicker fusion, digit symbol substitution, complex reaction time and subjective mood. They reported "The effect on visuo-motor coordination was limited and was dose related and linear, and performance was altered on visuo-motor coordination with 60 and 90 mg codeine, and on dynamic visual acuity with 90 mg codeine (P less than 0.05). No other effect of codeine was detected." Saarialho-Kere et al found "Codeine ... failed to affect performance in objective tests (body sway, digit symbol substitution, flicker fusion, Maddox wing, nystagmus) "
2.9 Conley et al, studying the effect of cold-water immersion on the effects of morphine among naive users found "Morphine impaired psychomotor performance during one of the warm-water immersions, but not during the cold-water immersions." In a comparison study of the cognitive effects of morphine and hydromorphone, "morphine had less adverse consequences", Beauford et al found no significant effect of morphine on psychological test scores.
2.10 Nasal butorphanol in a high dose was found to "impair psychomotor performance for up to 2 h, and produce subjective effects for up to 3 h. The smaller dose had no psychomotor-impairing effects, but had subjective effects (including increased ratings of "sleepy"). All three active drug conditions included miosis (pupil constriction)."
3 Cognitive Function
3.1 Much of the research involving the cognitive effects of opiates has focussed on methadone maintenance patients. Methadone has been found to adversely affect cognitive ability, specifically memory impairment, and also "information processing, attention, short-term visual memory, delayed visual memory, short-term verbal memory, long-term verbal memory and problem solving."
3.2 Ornstein et al.reported "heroin abusers were impaired in learning the... intra-dimensional shift component... tests of spatial working memory... failed to show significant improvement between two blocks of a sequence generation task after training and additionally exhibited more perseverative behavior on this task... profoundly... impaired on a test of pattern recognition memory sensitive to temporal lobe dysfunction.", and Eiber et al noted "Opiate addicts showed a decrease in episodic autobiographical memory but an increase in semantic affective memory and objective modalization."
3.3 Numerous studies have investigated the effect of maternal use during pregnancy on cognitive function of children, mostly finding no effects once social circumstances are controlled for, Goddard et al finding "childrens' behaviour and cognitive skills were not adversely affected", while Fabris et al found "No long-term neurologic or cognitive deficits are directly associated with heroin or methadone use"
3.4 Castaneda et al, studying patients with dual diagnosis of psychiatric disorders and drug dependence, reported "Heroin addicts reported that heroin improved some of their psychiatric symptoms and all of their cognitive dysfunctions." Studying groups of individuals dependent on different drugs and controls, Amir et al found heroin addicts to make more errors in tests of cognitive impairment than controls.
3.5 Various studies have investigated cognitive impairment among drug users infected with HIV, however few have attempted to associate impairment with drug use whilst controlling for HIV status, and most consider the effects to be due to the virus rather than the drug. Del Pesce et al found HIV infection was associated with cognitive impairment in intravenous drug users compared to seronegative users. Silberstein et al reported "seropositive IVDAs may show evidence of impaired neuropsychological function even in the absence of AIDS related symptoms and are consistent with the hypothesis of the early neurotropism of HTLV." Concha et al, studying neuropsychological performance of drug users, reported "Effects of the frequency of reported past use of marijuana, heroin, cocaine, barbiturates, and alcohol were not statistically associated with performance on the tests."
3.6 Cipolli & Galliani, using Rorschachs ink blots, found long-term heroin addicts to perform worse than addicts of shorter duration, considering their results to "support the hypothesis that cognitive functioning is impaired along with addiction time" Roszell et al, comparing patients receiving antidepressants and methadone maintenance, noted "There were no significant differences between groups on cognitive measures." Miller reported "A neuropsychological review of systems is likely to show a pattern of impairment in substance abusers that involves the integration of different cognitive functions for effective problem solving."
3.7 Keiser et al found that a group of heroin addicts performed better on the "positive Digit Span scatter" test than neurotic/depressive patients, whilst Lombardo et al found no differences in cognitive function between low and medium-dose methadone patients. However Gritz et al reported "Methadone subjects performed significantly poorer on several tests of learning and immediate recall compared to abstinent subjects."
4 Driving Performance
4.1 Meijler considered the Dutch ban on driving for opiate addicts to be unjustified: "There is no scientific basis for such a measure, however. On the contrary, current evidence indicates that e.g. cancer patients using 209 mg morphine daily for three months do not differ significantly from a control group with respect to thinking abilities, alertness, concentration, reaction speed and dividing attention. For obvious reasons utmost care must be observed with the use of morphine by traffic participants. But a rigorous prohibition of driving for patients requiring chronic alleviation of severe pain needlessly restricts their mobility."
4.2 O"Neill considered "the relationship between measurable effects (of opiate drugs) and the performance of everyday tasks such as driving is unclear." Jonasson et al, studying analgesic use among drivers suspected of driving under the influence of drugs, concluded : "analgesics containing dextropropoxyphene or codeine are not drugs of primary interest in this specific population."
4.3 Heishman et al studied the accuracy of field impairment tests, finding that trained officers were able to identify the correct class of drug in under one third of test cases.
4.4 In Denmark, Neilsen et al found "The frequencies of accidents in cases with morphine or methadon were lower than in the material as a whole while the frequency of accidents for dextropropoxyphen was higher". Smith, writing in the British Medical Journal, considered that patients taking stable dosages of morphine should be able to drive safely. By contrast, Sticht et al described two fatal traffic accidents following heroin consumption, in one of which the concentration was such that the driver was risking a fatal overdose.
4.5 Chesher, reviewing the evidence in 1985, stated: "The behavioural pharmacology of intravenously administered heroin suggests that any drug induced deficit in driving performance is not due to any effect on psychomotor function, but might be expected from the effect of the drug on mood states. Methadone, as used in treatment schedules for narcotic dependence produces no significant effect on measures of human skills performance. Epidemiological data are contradictory though the suggestion is that the involvement of the narcotic analgesic drugs in road crashes is unlikely to be a source of significant concern"
4.6 IDMU"s 1998 and 1999 drug user surveys found the overall accident rate for the survey respondents as a whole to be 0.608 per 100,000km (898 accidents in 147.8 million km), close to the national average. Frequency of heroin use was assessed as experimental (less than 10 times), occasional, regular, and daily. The accident rate for all heroin users was lower than the group as a whole at 0.507, although occasional users showed a higher than average rate. A lower proportion of regular or daily heroin users drove than respondents as a whole.
Accident rates among users of Heroin
Freq
No Drivers
Mean accids
Total Number
Mean km/ 5yrs
Total km/5yrs
Total accids
Accid. Rate
% users drive
Never
1835
0.44
2480
51994.16
128945517
807
0.626
74.0
Exp
156
0.38
203
53758.39
10912953
59
0.543
76.8
Occ
24
0.46
26
59261.54
1540800
11
0.717
92.3
Reg
13
0.46
22
47854.55
1052800
6
0.568
59.1
Daily
10
0.50
16
62000
992000
5
0.504
62.5
Ex-users
79
0.44
102
63670.46
6494387
35
0.535
77.5
Total
306
0.38
407
56352.19
22935341
116
0.507
75.2
4.7 The study also asked respondents whether they had had accidents under the influence of particular drugs. Of 245 such accidents reported, only 5 involved heroin. However because the low incidence of heroin use, this suggested a slightly higher risk compared to the incidence of use of other drugs, and did not approach statistical significance.
4.8 In the 1994 IDMU study, heavy polydrug use was associated with a significantly higher level of accidents.
5 Culpability analysis studies
5.1 Crouch et al considered that in 50 out of 56 cases where drugs or alcohol were found, these contributed to the accident, however it is unclear to what extent drugs other than alcohol were increased culpability.
5.2 An Australian study of Drummer, investigated over 1000 accidents, using risk analysis to compare the relative accident risks of alcohol, cannabis and other drugs, finding alcohol (p<0.0001) and multiple drug use (p<.05) to significantly increase the accident risk, whereas cannabinoids reduced the risk ratio to a degree which approached statistical significance (p=.065). Opiates were present in 3.2% of cases, with a slightly increased level of culpability - an odds ratio of 2.0 (range 0.7-6.5). Although this result did not approach statistical significance, such an increase cannot be ruled out due to the low incidence of opiate positives.
6. Testing & Pharmacokinetics
6.1 Meneely found that ingestion of poppy-seeds (25g) in cakes generated opiate-positive urine tests (300 ng/ml cutoff levels), although "no subjects were found to exhibit symptoms of opiate impairment." Gjerde et al found that morphine levels in blood to have resulted from metabolism of codeine and/or ethylmorphine, stressing the importance of analysing body fluids for a spectrum of opiate drugs where these are suspected to have caused impairment.
6.2 Niedbala et al found "93.6% agreement between oral fluid and urine" when testing for opiates, and concluded "oral fluid may be a reliable matrix for opiate detection". Vandevenne et al, reviewing detection times, reported "Experimental data for total morphine using a cut-off of 300 ng/mL suggest a detection time of 1 to 1.5 days for relatively low doses of heroin (3-12 mg) administered via i.v., IN or i.m. route." Cone et al found heroin levels in the blood to peak at 5 minutes after nasal spray or intramuscular injection, identifying the need to monitor blood levels of "heroin, 6-acetylmorphine, and morphine"
6.3 Smith et al investigated the reliability of increased immunoassay cut-off levels in detecting known doses of morphine - 8 subject received intravenous doses of 3, 6, and 12 mg heroin HCl and four smoked 3.5-, 5.2-, 10.5-, or 13.9-mg doses of heroin (base) - and eliminating false positives, finding "Detection times for morphine using the 300-ng/mL cutoff assays was approximately 12 h for low dose and 24 to 48 h for higher doses of heroin. For the two 2000-ng/mL cutoff concentration assays detection time was about 12 h."
6.4 In the USA, workplace testing cut-off levels for morphine were increased from 300ng/ml to 2000ng/ml in November 1998, to avoid false positives arising from medicinal codeine preparations and poppy-seed foodstuffs. Monoacetylmorphine has been proposed as a urinary marker for recent heroin use. The legal limit of plasma morphine for driving under Belgian law is 20ng/ml, equivalent to 10ng/ml in blood, levels of 500ng/ml were found in saliva.
7 Summary - Opiates and Driving
7.1 Although opiates can cause impairment in na"ve users, most addicts are unlikely to be significantly impaired when driving under the influence of maintenance doses of heroin.
7.2 Many addicts receiving stable doses of pharmaceutical diamorphine are able to function normally - the heroin-addicted doctor is the classic example.
7.3 Use of larger dose, sufficient to produce significant intoxication, may impair performance.
7.4 Addicts are more likely to be impaired when withdrawing from opiate drugs.
