Tue 31 Jan 2006
Dr Adam Winstock, who also presented on this subject at the recent APSAD conference in November in Melbourne.
A whistlestop tour through the epidemiology of self-reported recent substance-affected driving in Australia showed prevalence to range from 3-9%, varying depending on such factors as substance availability, availability of other transport, and the age group and cultural mores. Preferred substances vary greatly, MDMA, cannabis and alcohol being most commonly used by clubbers, and stimulants by truck drivers. Illicit opioids are generally the least commonly used by drivers. Not surprisingly, young men are the group most likely to engage in substance-affected driving.
The risk of substance affected driving is not only due to intoxication, but also to more subtle effects such as altered judgement and risk perception, and post-use effects such as the alcohol 'hangover'. Other important factors include the behaviour of passengers (even when the designated driver is not substance affected), driver inexperience, road and vehicle conditions. Such factors tend to aggregate, as in ......a group of drunken teenage males in a rickety car hooning on a dirt road on a dark and stormy night.....
The legal concept of the 'culpability' of certain substances was defined: "If drugs contribute to road accidents then drivers found culpable for crashes will be more likely to have drugs in their bodies than drivers deemed non culpable". Culpability is clearly demonstrated with benzodiazepines and alcohol, but not for cannabis, opioids and stimulants.
The 'culpability' of alcohol is striking, with a dose-related risk of accidents with rising blood alcohol level (BAL). Permissible blood alcohol concentrations vary from zero (Bulgaria, Turkey), 0.02 = 20 mg/L (Sweden) and 0.08 = 80mg/L (UK, Austria, Spain). Dr Winstock deplored this high permitted level in the UK.
Driving under the influence of alcohol is strongly associated with other alcohol related problems, particularly dependence with rate of accidents being 0.5/year for all drivers, 2.5 for binge drinkers and over 3/year for those with alcohol dependence. Further, there is an overrepresentation of alcohol dependent drivers who are caught with very high BAL .
Driving simulator studies show impairment of driving skills by cannabis, however there is a wide variation between individuals. The impairment appears to be less in true on-road conditions, possibly because cannabis affected drivers are capable of adapting by driving more carefully. There is some evidence that chronic use of cannabis causes impairment of driving unrelated to current intoxication.
Gamma hydroxybutyrate (GHB), like alcohol, produces dose-related psychomotor impairment. It is especially prone to cause gross impairment (including vomiting and amnesia) owing to the narrow 'therapeutic window' of its desired effects. Testing for this drug is difficult as it may be detected in small amounts in the body as a normal metabolic product. We were told that it is used therapeutically in some countries for alcohol withdrawal management.
Stimulants such as amphetamines may possibly improve some aspects of driving by increasing vigilance and stamina, and reducing fatigue. However judgement may be impaired at higher doses, with over-confidence, risky or reckless behaviour. Depending on the dose and a person's reaction to the substance, stimulants may have other risks, including altered vision from dilated pupils, the possibility of perceptual disturbance or paranoia. MDMA has been shown to increase the rate of accidents on a driving simulator (De Waard 2002).
Given that Attention Deficit Disorder (ADD) is itself a risk for driving accidents, it is an interesting question whether people with this condition ought to be provided with stimulant drugs that might in some cases improve their safety on the road.
With prescription medicines, both the underlying condition as well as the effects of the medicine need to be considered, as well as interactions with other substances. An example is opioids, where the stabilised opioid-tolerant person will generally show no signs of psychomotor impairment from their medication. However there may be increased impairment when combined with alcohol or other sedative-hypnotics, and one needs also to consider any associated personality or other psychiatric disorders.
In Germany, it was reported, a patient on opioid replacement is considered impaired until proven otherwise and facilities exist for formal testing. In New South Wales, the situation is rather complicated, and Dr Winstock took us step by step through his experience of trying to achieve clarity about legal and professional responsibilities of the health professionals. A useful website for understanding these is Austroad's Information for Health Professionals section.
In general it is, in the first instance, a driver's responsibility to report to the relevant Drivers Licensing Authority (DLA) any medical condition which may impair their ability to drive. In NSW, for example, the DLA is the Roads and Traffic Authority.
If a patient is "unable to appreciate the impact of their condition, or to take notice of the health professional's recommendations due to cognitive impairment or if driving continues despite appropriate counselling and is likely to endanger the public, the health professional should consider reporting directly to the Driver Licensing Authority" and "in the Australian Capital Territory, New South Wales, Queensland, Tasmania and Victoria ..... health professionals who make such ....without the patients consent but in good faith that a patient is unfit to drive, are protected from civil and criminal liability ".
This leaves the health professional with flexibility and at the same time a grey area of legal responsibility. It was pointed out that a family doctor may have very detailed knowledge of a person's psychosocial situation which may help making differentiated judgements but might also involve a conflict between the interests of their patient and society at large.
In general situations involving imminent risk and involving commercial drivers call for more decisive action from health professionals.
In determining whether to report someone to the Driver Licensing Authority, Dr Winstock advised considering:
- the risks associated with disclosure without the individual's consent or knowledge, balanced against the implications of non-disclosure;
- whether the circumstances indicate a serious and imminent treat to the health, life or safety of any person.
He reminded us of the formula: Risk = (likelihood of the event) x (the severity of consequences).
Unfortunately, anomalies do arise. Dr Winstock was advised by a NSW Roads and Traffic Authority spokesperson that a driver need only self-report a medical condition that was chronic, therefore the commencement of methadone treatment need not automatically be notified. However, if a person stayed on methadone treatment for over 12 months, notification would be appropriate. This is exactly the opposite of what an experienced clinician would advise - a patient early in methadone treatment should be advised not to drive until the dose is stabilised, a person stable on methadone maintenance can be expected to be perfectly fit to drive. In cases where illicit use compromises driving safety (regardless of the treatment status of the patient) notification by the patient should be recommended.
Dr Winstock's algorithm "What should do we do in practice ?" is based on the need to protect ourselves, the patient and the community where a driver has an impairment to driving.
- Give feedback to the patient on the legal and financial obligations and implications for them, as well as your own legal/professional responsibility.
- Invite the patient to notify the DLA, or inform them of your intention before you do so yourself.
- Document your advice, including that you have told patient what they should do.
- Request feedback and document patient's reported action/inaction re-driving at next appointment.
- Assess immediate risk if patients continue to drive.
- Document and seek second opinion from colleague/DLA.
- Advise patient that unless they notify the DLA you will.
The case studies illustrated how these principles might be applied in practice.
In the first case, a patient on methadone maintenance developed psychotic depression and was commenced on risperidone and citalopram. The patient was notified to the DLA, and a subsequent medical report was that her akathisia did not affect her psychomotor skills and she was fit to drive. However, another doctor started her on large doses of diazepam for the akathisia. Interest centred on the inappropriateness of using diazepam for akathisia, but also the question of how a doctor best determine whether there was any impairment from this combination of medications. Suggestions included examination 3-4 hours after supervised methadone and diazepam dosing.
A second case study dealt with a patient on methadone maintenance known to use benzodiazepines who was observed to stop on the wrong side of the road then reverse across double lines to park on the correct side. He claimed that he did not usually drive as he was unlicensed and his car unregistered. Although the patient showed no signs of benzodiazepine intoxication, the behaviour could reasonably be described as disinhibited, and the responsible doctor chose to deal with this by giving very firm warnings, including a future possible notification of police or the Department of Community Services (responsible for the safety of children in NSW). Debate centred on whether the doctor should have confiscated the car keys, immediately informed the police, notified the DLA (even though it has no powers over unlicensed drivers) and the question of how best to follow up the patient for compliance with acceptable standards.
In another case a patient using a cocktail of medications including sedative hypnotics, tricyclic antidepressants, opioids and a major tranquilliser, was offered inpatient management of her problems. She agreed to be admitted but insisted she had first to drive home to cook dinner. In the face of this woman's signs of current intoxication, the doctor acted firmly to take her car keys (which the patient accepted in good spirit), justified by the immediate and serious risk to her and the public.
This led to discussion of possible protocols such as requiring anyone entering an alcohol detoxification treatment to agree to notify the DLA of their impairment. It was pointed out that draconian measures by health professionals could push a person out of the therapeutic relationship, losing the opportunity for constructive engagement.
Finally, some public health challenges including road side drug testing were discussed:
- Legal status of drugs is unrelated to driving risk.
- Nor does the mere presence of drugs imply impairment.
- Limitations of road side testing, including the number of false negatives.
- Visible, frequent, random testing may maximise exposure to enforcement and testing may alter perception of risk and lead to a positive impact on people's decision whether to drive.
Office tests such as heel-toe, past-pointing, Rombergs and examination for lateral gaze nystagmus are good for alcohol but much less good for other causes of impairment (the latter is a good party trick as well). Dr Winstock briefly described other more sensitive measures of impairment such as head sway measures that may become more widely used in the future.
A telling note was struck when Dr Winstock asked how many participants at the seminar routinely asked about driving in every clinical drug and alcohol assessment. At the end of the seminar participants were left with a heightened awareness of this need and a practical framework for acting to protect themselves, the patient and the public.