• HEALTH POLICY & PRACTICE
• CLINICAL NEUROLOGY

Being able to drive is integral to many aspects of life in many parts of the world, and is regarded very much as a right rather than a privilege. It is therefore not surprising that conversations during consultations about limiting an individual’s eligibility to drive can be challenging and sometimes heated. In this edition of Practical Neurology, Rees et al discuss many of these issues from the perspective of the UK national licensing panel (Driver and Vehicle Licensing Agency, DVLA).1

When creating and implementing policy, there are several substantial challenges.

The first of these is to strike the appropriate balance between protecting the rights of the driver on the one hand and those of other road users and pedestrians on the other.

The second is to take heterogeneity adequately into account. Drivers are not uniform automatons, but vary in many ways (eg, age, driving experience, time spent driving per day, risk-taking behaviours, etc), which influences the risk they pose to themselves or others. Similarly, people with medical conditions that might affect their ability to drive safely are also heterogeneous. This heterogeneity poses a major challenge to those creating policy, which must be fair, consistent and easy to implement. In fact, there is an inevitable tension between fairness and ease of implementation, as the granular detail required to take heterogeneity into account can result in policy and guidance that is too complex to implement and to keep up to date. There is also the risk of unintended consequences when policy is transcribed into legislation, the interpretation of which might not be as intended.

The third is to account for uncertainty, as it is not always possible to give a definitive diagnosis. For example, an incapacitating event might not have been witnessed, and even those that have can be difficult to differentiate among a range of possibilities. What degree of certainty is required to prevent an individual from driving?

So how do we decide what level of risk is unacceptable? And how do we make a judgement that an individual poses sufficient risk behind the wheel that they should lose their driving licence? The easier place to start is with incapacitating events that might recur, such as seizures. What size of seizure recurrence risk is acceptable? And what level of risk, if used as a standard, would not penalise people who have had a seizure when compared with other groups who are legally allowed to drive? While this decision can be informed by epidemiological evidence about risk, it ultimately requires a judgement to be made. This was considered by a European Union (EU) working group in 2005,2 which recommended that ordinary (group 1) drivers should be prevented from driving if their risk of a seizure in the next 12 months exceeded 20%. Fortuitously, that was also the standard that the DVLA was working to at the time, and continues to do so following the harmonisation of EU driving regulations.

Once a standard has been agreed, the epidemiological evidence can be used to inform the duration of time off driving. The general approach is one of conditional risk: on the condition that an individual has been ‘X’ months without a seizure, what is their risk of a seizure in the next 12 months? This approach underpins the guidance for those with a first unprovoked seizure,3 where people are allowed to drive after 6 months if there are no risk factors that increase their risk (relevant abnormalities on EEG or brain imaging), or 12 months if there are risk factors that increase their risk. A similar approach is taken for newly diagnosed epilepsy and for anti-seizure medication withdrawal.4

The standard that is recommended for group 2 drivers (heavy goods vehicles, etc) is necessarily stricter, requiring a seizure recurrence risk of less than 2% in the next 12 months without anti-seizure medication. The problem is that there is very limited epidemiological evidence to inform policy, the main challenge being that there are no prospective studies of sufficient duration to allow reliable risk prediction and stratification. This has resulted in an element of ping-pong between the DVLA and clinicians, when the onus is put on the clinician to indicate whether an individual’s risk is above or below 2% in the next 12 months. Looking forward, to improve consistency of approach, the DVLA is moving towards stating an indicative duration of time off driving, as described by Rees et al. 1

A much more difficult problem is to create a standard and guidance for people with progressive or degenerative neurological conditions, where there are many fewer epidemiological data to inform judgements and guidance. At what point does an individual with Alzheimer’s or Parkinson’s disease have to stop driving? What are the risk factors that help to stratify this risk? Rees et al describe the DVLA’s approach to this,1 but there remains the challenge of ensuring that any assessment tools are used in an equitable and reproducible way.

It is clear that the DVLA panel operates in a challenging space, attempting to provide and implement guidance. It is difficult to judge whether a future that includes autonomous vehicles will make that task any easier.