Road Safety Part 2: Medical aspects of fitness to drive - Spring 2003

Introduction

Although pre-existing ill health makes only a relatively small contribution to road accidents, both the declaration and the assessment of medical fitness-to-drive is part of the licensing process. In addition, specific advice given by a person's clinician informs the driver on whether he or she can drive safely. This programme of research aims to improve the quality of evidence underlying both licensing decisions and medical advice to ensure that risks to road safety are minimised while drivers are not unnecessarily restricted.

Recently completed projects

DVLA database development and analysis

The current DVLA databases hold a large amount of information. In addition to the main database of all licensed drivers, the DVLA also maintains a separate database of relevant medical information on drivers who have notified the DVLA of certain medical conditions. This Drivers Medical Database is used to control and administer the process whereby certain individuals may be issued with restricted-term driving licences for medical reasons, or have their licence revoked. This database was set up as an administrative tool only, and hence it is at present not useable for other purposes.

It is nowadays highly desirable to use such an important pool of information, which is unique in Europe, both for research purposes and for analyses that will help inform future decision making regarding medical fitness to drive. For this reason, a scoping study was commissioned to examine the following issues:

• Determine the needs of DVLA staff, and in particular the needs of Medical Advisers.
• Assess the medical information already stored in the current database.
• Explore the scope for modifying the existing system to provide the outputs required.
• Recommend ways that research could be extended by appropriate data linkage and/or skill set management.

This project aims to advise on the collection and analysis of medical information relevant to fitness to drive. The aim is to provide the Medical Advisers of the DVLA with systems that will help organise, plan and monitor the medical data of nearly 40 million UK drivers. The system will be used to predict needs, resources and future workloads and to provide data for internal and external research programmes.

The project used individual interview and workshop sessions to identify users' needs. In addition, the current database was explored in depth and a list of recommendations was compiled.

Summary of main recommendations:

• Create a stand-alone version of the Drivers Medical Database for research purposes, detached from the operational data store.
• Use this database to test hypothesis, and not to generate them.
• Review data capture strategy and processes as soon as possible, taking into consideration future research and policy needs as well as today's operational needs.
• Explore further linkages with external data sets
• Set up a small group of internal and external specialists to determine the most appropriate coding strategy to address the ambiguities in the current scheme and link it with external reference schemes like ICD-10.
• Acquire the primary skill sets required to conduct research using the Driver Medical database: academic healthcare expertise, specialist IT experience, epidemiological skills, project management skills.

Next steps: At present the DVLA mainframe IT system is being modernised. When this is completed, the recommendations of this study will be implemented to enable medical data analysis.

The role of clinical status and performance relevant to driving: a review

The number of medical conditions that may have a potential detrimental effect of driving safety is large. For several medical conditions, the value of clinical and experimental information to the licensing decision-making process is often uncertain. In order to focus the development of the medical aspects of fitness to drive research programme, and to place the studies within the programme in the context of current knowledge, the current review was commissioned.

The main aim of this project was to review the scientific studies examining the relationships between medical conditions and aspects of performance that may be relevant to safe driving.

The review would provide background information that will enable us to set priorities for future research on medical aspects of fitness to drive, in an informed way.

Summary of main findings:

• There was significant uncertainty in identifying safe and unsafe drivers with medical conditions. Group data implied the presence or absence of an increased risk of accidents, but failed to reveal individuals or sub-groups that were at fairly high or fairly low risk.
• The clinical examination (together with other tests - as in the assessment of cognitive impairment) remains an important part of the process of assessing fitness to drive. It not only diagnoses and evaluates the condition under review, but may also reveal other significant medical factors. Nevertheless, it is clear that there are limitations to the usefulness of the clinical data in the prediction of safe driving.
• It is unlikely that the use of simulation in the assessments of patients will prove practicable in the foreseeable future. It is, therefore, necessary to determine the circumstances under which individuals should be afforded an on-the-road test. This is particularly so in the assessment of patients with mild cognitive impairment.
• The review highlights the need for a further, specialist, appraisal of studies that have attempted to relate visual field defects and accident involvement. In addition, it recommends that studies are carried out to identify individuals on insulin who are unlikely to be impaired while driving due to an inherent instability of their condition.

Next steps: The findings of this study are guiding the development of the medical aspects of fitness to drive programme

The role of risk analysis and sensitivity analysis in guiding fitness to drive decisions

Road transport is inevitably associated with a level of risk. Driving restrictions are imposed on certain groups of individuals whose medical condition is perceived to pose an unacceptable level of risk.

The basis on which current decision making regarding this level of risk are made needs constant updating. To examine these issues, we carried out an investigation into the usefulness of the risk analysis approach to preventing road accidents related to drivers' medical conditions.

This review aimed to:

• Assess the scope for using available clinical and performance data in risk analysis directed at preventing road accidents related to medical conditions.
• Establish the extent to which quantitative or qualitative analysis can be used to derive equitable, evidence-based standards to apply to decisions on medical aspects of fitness to drive
• Help to inform the future direction and priorities for research on medical conditions and accident risk in driving.

The application of risk analysis to the decision making process has 2 stages:

1. The assessment stage involves the estimation of the risks associated with a particular medical condition, and has to be based on the available information. It involves the exposure to risk, the probability of impairment or incapacity while at the wheel, the probability of an accident resulting from impairment or incapacity and the likely outcome of such an accident.
2. The management stage relates to the level of risk that is acceptable or can be tolerated.

Decisions may be related to a single individual or to the community.

The risk analysis methodology was applied to medical fitness to drive, and a possible line of approach in conditions such as epilepsy and hypoglycaemia was outlined.

Summary of main findings:

• The application of the risk assessment approach to medical fitness to drive is limited because of the lack of credible evidence, at least in some areas, on which to base calculations. In some cases the approach cannot be justified because the information is so unreliable that the approach may lead to invalid conclusions.
• Further experimental studies are needed to fill some of the gaps in present knowledge. These are not only required to establish the impact of impairment on driving performance, but also the probability of impairments at different levels of severity. In this context the identification of subgroups with little or no increased risk will enable the medical licensing rules to be relaxed with minimum impact on accident rates and a positive impact on personal mobility. There is a need to obtain reasonable estimates of accident risk for individual groups.
• Sensitivity analysis is needed to determine how strongly the outcomes depend on parameters that can only be imperfectly estimated.

Even if the impact on risk would be, in the future, accurately assessed, further issues to be addressed are the levels of increased risk that can be accepted, either overall or in relation to specific groups of individuals.

As a sensitivity analysis was recommended, this was additionally carried out. It aimed to evaluate the relationship between a medical condition and the accident risk associated with it, and especially to examine how the calculated accident risk varies when parameters related to the medical condition change. This would enable us to determine how strongly outcomes, such as accident risk, depend on parameters that can only be imperfectly estimated.

Sensitivity analysis allows us to assess the impact from changes in the assumptions of a particular model, and hence to test a model's reliability and robustness. The issue of sensitivity is related to the uncertainty in the true values of the parameters, such as the frequency of a certain critical event occurring in someone with a medical condition (e.g. the timing of seizures in an epileptic person).

The analysis can take the form of calculations based on specific sets of parameter values. The outcome (e.g. accident risk) can be examined by using a range of values for each parameter in model. Epilepsy, severe and mild hypoglycaemic reactions and the administration of anxiolytics and zopiclone were considered in this sensitivity analysis, and a range of parameter values (optimistic, typical, and pessimistic) was used for each condition.

The main conclusions of this analysis are:

• In the majority of cases there was considerable uncertainty concerning many of the parameters.
• Limits on the outcome variable (accident rate) tended to be extremely wide, and often varied by a factor of 10 or more.
• The analysis helped to identify those areas where more precise information would be valuable:
• A more precise assessment of risk would be possible in the case of epilepsy if more information was available on the consequences of an incident at the wheel and the probability of a serious outcome.
• In hypoglycaemia, an accurate medical assessment of the risk of a hypoglycaemic event occurring in each individual or group of individuals would enable the risk analysis to be based on a rule such as the 2 & 20% rule.
• The uncertainty in the calculations related to anxiety and the treatment of anxiety would be reduced if more accurate information was available on the average duration of an individual prescription.

Next steps: The findings of this study helped focus the development of the medical aspects of fitness to drive programme

Visual fields and driving: A review of selected papers

A critique of the recent literature that examines the link between visual function and car accident data was carried out. The review focussed on visual field defects, and examined the clinical and statistical evidence, taking into consideration the methodological strengths and weaknesses of both cohort and case control studies.

The review concluded that:

• Although the studies reviewed suggest that there is an association between visual field defects and an increased risk of crashes, it is not clear what aspects of the visual field are essential for safe driving.
• The studies often lacked crucial information on the visual capability of the driver and the circumstances surrounding the accident.
• Accident data alone cannot establish the link between the characteristics of the visual field and fitness to drive. Such characteristics need to take into account the binocular nature of the visual driving task. The association between visual field defects and increased risk of accidents, together with an estimate of what constitutes an acceptable level of risk, should form the basis of policy on the visual field requirements for fitness to drive.

Next steps: The findings of this review were used to develop future visual field and driving research within the medical aspects of fitness to drive programme

Excessive daytime sleepiness: a workshop

Although much research into the effects of fatigue and sleep depravation has been carried out, there is uncertainty concerning the nature and contribution of the medical causes of excessive daytime sleepiness to accident involvement.

This project brought together experts from several fields, with the aim of formulating recommendations concerning the implications of the medical aspects of excessive daytime sleepiness to driving safety.

Sleep experts from the disciplines of neurology, respiratory medicine and psychology were invited to present their clinical and research views and discuss the implications of medical aspects of excessive daytime sleepiness to driving safety.

The panel of experts present made the following recommendations:

• Disorders of sleep of medical origin are a significant factor in Sleep Related Accidents. Medical Practitioners should be encouraged to ensure prompt investigation of drivers with Excessive Daytime Sleepiness.
• It is the responsibility of individual drivers not to drive when sleepy, to take appropriate action should they become sleepy and to inform the regulatory authority if their sleepiness is due to a medical condition. It is not the responsibility of the physician to inform the regulatory authority of the diagnosis, except in exceptional circumstances.
• The investigation of neuro-degenerative (including Parkinson's Disease) and neuromuscular disorders should include an adequate sleep history. Sleep difficulties in these disorders require further attention.
• Group 2 drivers should be provided with information on sleepiness and the risk of accidents. This could be provided at the time of initial applications for a licence and on renewals
• The medical status of individuals involved in Sleep Related Accidents should be established
• The contribution of medical causes of Excessive Daytime Sleepiness to Sleep Related Accidents needs further study.
• The prognosis and accident record of drivers with the complaint of Excessive Daytime Sleepiness (but who are presently considered safe to drive) should be studied.
• Cataplexy should be assessed as a potential risk factor in accidents. A follow up of patients with narcolepsy with respect to safe driving should be carried out.
• The prevalence and significance of symptomatic and asymptomatic sleep apnoea in Group 2 drivers should be determined.
• Support is needed from the regulatory authority for the use of smart CPAP machines (that provide checks on compliance by recording 'mask on' time) in appropriate circumstances.
• Drivers and medical practitioners should be made aware that it is current medical opinion that surgical procedures are unlikely to be of wide benefit in the management of sleep apnoea, and that the use of drugs that promote wakefulness needs careful supervision.
• The regulatory authority should consider how appropriate literature for patients attending sleep centres can be provided.

Next steps: The DVLA are currently taking further the recommendations of this workshop

Cognitive impairment I: a workshop

The aim of this project was to identify a set of issues regarding fitness to drive in those who suffer cognitive impairment that are important in clinical practice and within the licensing system.

A programme of research into driving with cognitive impairment of various causes would subsequently be based on these recommendations.

A Steering Group of experts was set up, and discussions between the members led to the identification of several issues that needed to be resolved in the approach to the driver with suspected cognitive impairment. The following aspects that need further exploration were identified:

Clinical issues

• The value of the clinical and driving history including the value of information from relatives.
• The value of the diagnosis for neurological disease in the prediction of the natural history of the disease, and the prediction of optimal frequency of re-assessment.
• The value of the assessments carried out by occupational therapists (other than those involved in on-the-road assessments) concerning cognitive and relevant physical capabilities.

The behaviour of the driver

• The importance of the assessment of behaviour, perhaps in the context of challenging social and domestic situations as a surrogate for driving behaviour.
• The need to determine self-awareness on the part of the driver regarding their level of disability and driving skill, and evidence of insight on the part of the driver.
• The usefulness of a 'brief screening instrument' that would have considered the potential of 'surrogate measures'.

Assessment of cognitive function

• The need to assess sustained and divided attention, global cognitive function and judgement, and the detection of early deficits in these domains.
• The consideration of tests that are acceptable to the driver (face validity), and consideration of the acceptability and feasibility of conducting such tests to the medical assessor.
• The value of currently available 'paper and pencil' tests.
• The need to assess memory.

Assessment of driving skill

• The potential usefulness of currently available assessment rigs in the Mobility Assessment Centres.
• The role of the Mobility Assessment Centres in evaluating driving behaviour in those deemed as neither fit nor unfit to drive safely by the clinical or neuro-psychological assessments. Particular attention to methodology and measurement.
• Consideration of driving assessments in the validation of the proceeding clinical and neuro-psychological assessments.

Next steps: The recommendations have been used in the formulation of the forthcoming research programme on Cognitive Impairment and Driving.

Ongoing research

Peripheral visual field defects and driving

This is a study of the driving correlates of peripheral visual field defects. The association between these factors will be studied with the aim of improving the specification of standards for visual field defects and the methods to be used in measuring them.

The study includes the investigation of the most efficient testing methods for visual field loss and the correlation between visual field loss and driving performance measured by a range of techniques including simulated driving.

Anticipated completion date: Summer 2003

Insulin-treated diabetes and driving commercial vehicles

This is a three year multi-centre research programme that examines the relationship between the clinical features and treatment of diabetes, and the control of glycaemic levels.

The aim of the study is to identify, if possible, those drivers with insulin treated diabetes for which a case-by-case clinical assessment would be warranted. The results of the study should improve the accuracy of identifying those diabetic drivers who are at low risk of sudden loss of capability to drive because of hypoglycaemia. The results may enable some diabetic persons currently prevented from driving to resume their driving entitlement.

This project examines the package labels and the inserts for several medications to identify inconsistencies. It also examines the way in which warning labels are presented on packaging.

Anticipated completion date: Summer 2004

Central scotoma and driving

Our on-going research into the driving correlates of visual field defects is extended by this two-year in-depth study of the relevance of central visual field defects to the ability to drive safely. The study will evaluate the existing methods of examining central scotoma and develop more accurate measures, and will assess the association between performance on these measures and performance on several tests related to driving.

The aim of this project is to improve the basis on which standards for central visual field defects are set, and to refine the methods used in measuring central vision.

Anticipated completion date: Summer 2005

Forthcoming research

Cognitive impairment and driving II

Based on the recommendations of the Cognitive Impairment Phase I project, the Department initiated a further study into the driving implications of cognitive impairment.

The aim of this two-year study is to examine the role of clinical information, neuropsychological and behavioural assessment, and the assessment of driving skill on decisions of fitness to drive in persons with cognitive impairment of various causes.

The attitudes of health professionals to giving advice on fitness to drive

Much progress is being made on providing evidence on which fitness to drive decisions are based. However, this information does not seem to be well disseminated to medical practitioners and is not incorporated effectively in advice given to patients during consultations. This project will evaluate the awareness of risks and their mitigation, the opportunities for advising, and barriers to giving advice. It will also examine ways in which fitness-to drive advice could be incorporated better into clinical consultations and what support measures are needed to achieve this.

The results of this project will be used to develop support measures and systems that improve on current practice.

Anticipated start date: Autumn 2003

Expert consensus workshop on irregularities of cardiac rhythm

At present, dealing with irregularities of cardiac rhythm is very difficult in terms of licensing decisions, as little research evidence on which to base decisions is available. In addition, in recent years new treatments have made a considerable difference to prognosis. This highlights the need for an update on the facts regarding the frequency of occurrence of such events and the likelihood that they will occur while driving and potentially lead to accidents. In order to gather the up to date research, an expert workshop will be organised.

Participants will present position papers and discuss the driving - related issues. The results will be used as a basis for examining the current licensing decisions made in patients with these medical conditions. It will also provide the up-to date evidence and will highlight areas where research is needed.

Anticipated start date: Autumn 2003

Systematic review of risks of further acute vascular events

Acute cardiac events (heart attacks, unstable angina, interventions such as angioplasty and bypass grafting) are relatively common medical problems in the general population. They carry a risk that in the near future a further similar episode will occur. Information regarding the duration between one event and the next, and about the nature of the next event is important, as it could help in identifying those drivers that and are highly likely to suffer another event in the near future, possibly while driving. A compendium of the epidemiological and actuarial research evidence regarding these conditions is not available at present, but would greatly improve the basis of making decisions in this large driver group.

The results will provide a summary of the evidence of probability of an event occurring over a short period after another event has occurred. This will allow us to stratify drivers with specific conditions within the acute cardiac events group into low, medium and high risk of recurrence, and this will improve the basis of licensing decisions substantially.

Anticipated start date: Autumn 2003

For more information about any of these projects please contact:

Dr Liliana Read
Research Manager Road Safety Division,
Department for Transport
2/09 Great Minster House
76 Marsham Street
London SW1P 4DR

E-mail: Lily.Read@dft.gsi.gov.uk
Tel: 020 7944 2054