Australians are living longer,1 at least partly because of better treatment of one or more medical diagnoses.2,3 Drivers are also more likely to continue to drive as they age.4 Age-related conditions such as osteoarthritis, dementia and cardiac disease can impair driving. Additionally, treatment medications can potentially impair safe driving.5,6
From the 1950s, the general practitioner (GP) was recognised as the best-positioned person to judge medical fitness to drive (FTD) because they had a long-term relationship with their patients.7,8 Long-term relationships with a GP have been shown to improve patient outcomes,9,10 but this has not been studied with respect to driving assessment. The first guidelines were developed in the United States in 19597 in response to the ever-increasing road fatality rate. The first edition of the current Australia guidelines was published in 1998 with further editions every five years,11 with the latest being in 2022.12
The current driving guidelines imply that the GP, with extensive knowledge of the ageing driver, in equivocal assessments, will make a judgement to balance whether the patient’s long driving history and self-regulatory driving practices compensate for age-related deterioration, resulting in continued safe driving.12 In case of uncertainty, the guidelines recommend on-road assessment by a specialist occupational therapist (SOT), noting that this will come at a cost to the patient (currently over $1000). The SOT conducting the road assessment is the ‘Gold Standard’ for driving assessment in cases of mild cognitive impairment, whether for early dementia13 or chemically induced cognitive impairment as a side effect of opioid use in patients receiving palliative care.14
The aim of this study was to elucidate factors that could lead to a more standardised and fair assessment of drivers undergoing mandatory driving medical assessments.
Methods
In Toowoomba, a regional city in southern Queensland, senior doctors at general practices were contacted. Information was emailed for the practice GPs’ perusal. The principal investigator (PI) knew the senior doctors professionally but did not know the others they recommended. Recruitment was a mix of convenience sampling and snowballing. One doctor declined, citing lack of time but referred a colleague who participated. All participants read and signed participant information and consent forms.
Ten voluntary participants contacted the PI to arrange an uncompensated semi-structured interview at a time and place of their convenience. The inclusion criterion was the performance of yearly mandatory driving licence medical examinations for either mandated age requirements or disease-specific conditions such as diabetes or epilepsy, and the prescription of opioids to potential drivers. A consent form was signed prior to an audiotaped interview between just the PI and participant. The interview guide was based on the PI’s extensive experience in driver licensing and driving impairment, with review by other members of the research team. The interview approach was practised by the PI with advice from a highly experienced qualitative researcher. The Standards for Reporting Qualitative Research checklist was also used to guide the design.15
The participants were doctors aged between 28 and 65 years, with the average age of 46 years. The range of years since graduation was 5–40 years, with an average of 21 years, and general practice experience ranged between 1 and 36 years, with an average of 16 years. Six were male and four were female and they were from five different GP practices. Half were senior GPs, Fellows of The Royal Australian College of General Practitioners (FRACGP); each had been in general practice for over 25 years. The remaining five were GP registrars (GPRs), qualified for 5–7 years and were undertaking GP training for 1–3 years as part of a four-year training program to attain the FRACGP. All registrar–patient consultations occurred independently with the registrar having the opportunity of debriefing with their supervisor on a weekly basis or immediately for urgent advice.
The interviews took place between November 2019 and November 2021. First, the PI explained his lifetime interest in road safety. The interview durations were between 28 and 56 minutes, with an average of 40 minutes.
Questions included:
- How do you make a licence assessment?
- Do you discuss driving as a topic in other routine appointments?
- Have you referred any patients for further driving assessment (ie an occupational therapist)?
- How did that go?
- Have you ever had to take a licence from a patient?
- Tell me about that?
For a copy of the complete interview proforma, contact the corresponding author.
A rich description of the topic was sought from the participants’ interviews, and the transcribed text was coded using reflexive thematic analysis (RTA).16 The PI’s reflexive position was informed by 28 years as a rural GP conducting driving license medical assessments; police-requested assessments of impaired drivers; and collection of blood specimens for drug analysis from impaired drivers and motor vehicle crash victims. In a further 12 years as a palliative care specialist, he has often been asked by referred patients medicated on opioids about their rights to drive. The PI’s approach to health is expecting all persons to get a fair share of abstract societal wealth, including freedom and independence.
Field notes were taken during the interviews, with additional material entered on re-listening to and re-reading the interview transcripts. Aliquots of dialogue, interpreted with an RTA approach in the data, were used as codes. These were then grouped to make subthemes and themes following in-depth discussion with the whole research team. The six-step method for RTA16 by Braun and Clarke – data collection; initial codes; searching for themes; defining themes; naming themes; reporting – was used, all seen though the reflexive lens of the PI and the team. This process was iterative and recursive, as described in a worked example;17 however, it focused on important themes rather than saturation.
This study received human research ethics approval from the Queensland University of Technology as part of a broader project: ‘How do doctors and pharmacists perceive their roles in ensuring safety in driving for both the patient and the general community?’ (Ethics approval number 1900000374).
Results
The participants readily recounted important aspects of the medical examination such as musculoskeletal examination and visual acuity. Senior GPs only saw familiar patients. GP registrars were often in the more challenging situation of consulting ‘new’ patients.
The main themes that emerged from the analysis were:
- Relationships
- The big worry
- Standards, equity and fairness
Relationships
Relationships: Trust leading to judgement
The importance of a long-term relationship when it comes to something as contentious as driving adjudications can be seen clearly in the dialogue with the senior GPs. These doctors had all been in general practice for between 25 and 35 years and had closed their ‘books’ to new patients, only doing mandatory driving licence medical examinations on patients seen regularly for over 10 years.
Closed practice, Yes, I know them very well. After 25 years you do know them well. (GP1)
For years I don’t see new patients. So, the patients I see are the patients that I know very well. (GP3)
Don’t have any problem filling out a (licence) form for a patient I have known for a decade or two. I know my patients well. (GP5)
New relationships
The younger doctors highlighted the difficulties of being in a new relationship when trust is not fully matured.
I think if you’ve seen a patient a few times it’s harder to say no to someone you’ve started to get to know regarding this as a conflict of interest, going on to compare the stress with a one-off patient contact: for example, if you see a patient in ED, you’re not going to see them again. (GPR1)
No. If they were (a new patient) I would probably apologise and not do it. Someone that knows them has to do it , adding ‘You would hope that you would have seen them a few times before’. (GPR5)
Stress on the new relationship with potentially restricting driving privileges is shown in the following quotes:
People’s independence revolves around their car. (GPR1)
Isn’t it important for everyone to have a licence? (GPR2)
He did live on his own as well – that was the last piece of independence that he felt we were stripping away from him. (GPR3)
You can’t live without driving. (GPR4)
But it’s hard, it’s hard to take away someone’s independence. (GPR5)
The big worry
The big worry: Assessment of age-related mild cognitive impairment and dementia
Alert cognition is important for safe driving. Senior GPs drew on experiential knowledge of the driver, whereas GP registrars often relied on the Mini Mental State Examination (MMSE).
Senior GP1 does not do any formal mental assessment, similar to most of the senior GPs. Consulting with the driver for over 20 years gave a good basis for judgement:
I don’t do a cognitive as in a Mini-Mental State Examination (MMSE) or anything like that. I choose to do that in passing. So, I make my own assessment as we go. All these patients are basically patients that I’ve known for years and years. (GP1)
GP4 similarly goes on her long-term knowledge of the driver:
I go by my gut feeling on knowing them. The other thing is that for years I don’t see new patients. So, the patients I see are the patients that I know very well. (GP4)
The development of ‘gut’ feeling is not an option for the younger GPRs. Most registrars said they would use the MMSE to assess cognitive impairment, which is a well-known tool.
So in terms of – I know it’s not always the most reliable in terms of interconnective cognitive impairment but that’s one I use the most. So that’s what I’m most comfortable with. (GPR1)
GPR5 also goes for the most familiar test:
Probably an MMSE. I think that’s probably what I would – at least my first line. I think it’s an okay screening tool. (GPR5)
Functional assessment for mild cognitive impairment and early dementia: The gold standard
The gap in the assessment for driving with cognitive impairment is the absence of on-road assessment. Many GPs make the comment that they do not see the patient drive.
(regarding on-road dementia driving assessment): It’s surprising how many people pass because, apparently, with early dementia, people still maintain their driving skills which I was unaware of. They can’t remember my name when they come in, but they are still able to drive. I can’t believe that. (GP3)
Cost restrictions also play a part in functional assessment:
(Regarding cost) A major issue. (GP1)
The trouble is the OT assessments are expensive. (GP3)
Yes. I’ve had every few years, someone will say that because they can’t afford it (on-road assessment) or don’t want to pay for it. We take the licence off them. (GP4)
They will do anything to keep their licence. They will part with the dough. (GP5)
(Regarding cost) It’s so exorbitant. (GPR1)
I know (cost) that’s a barrier, definitely is a barrier. (GPR2)
But it’s a fee and they have to pay for this. (GPR5)
Three of the senior GPs and one of the junior GPRs expressed surprise when a licence, conditional or open, was given to a driver after SOT review. GP5 sends 4–5 patients a year to a SOT:
I’m surprised, no one has failed yet. (GP5)
Certainly, there have been a couple of times at least when I was surprised that the OT passed the person. (GP1)
(regarding dementia patient’s SOT review) I was a bit surprised because I thought he may have failed. (GPR3)
Medico-legal concerns
Both senior GPs and GPRs were worried about making driving assessments.
Your name is on the paper so you’re responsible if they have an accident. (GPR3)
That’s something that’s pretty dangerous if you make the wrong call (on passing the driver’s medical). (GPR5)
I’m not sure that I’m the appropriate person to judge a person’s safe driving or not. (GPR2)
We all cringe and we’re all terrified that it’s going to be one of ours (when an older driver crash has been reported in the area). (GP2)
Senior GPs mitigate their medico-legal risk by only seeing well-known patients:
At the moment I don’t worry about it too much because I know my patients so well. I feel that I’m going to jump in before they have problems, one would hope. But it is a risk, I don’t do driving assessments, I don’t know what these people do (when driving), I’ve got no idea. If a strange person came in and said I want my license form, I’d say I’m not doing it. (GP5)
Standards, equity and fairness
Standards: Variability of the driving medical examination in experienced hands
The Austroads government-sanctioned resource12 gives good guidelines on some diseases, but in general, GP judgement is required.
When it comes to standards, senior GP1 comments on the variable practices in his clinic:
I just know this in our group, for example, it’s very haphazard what all doctors do. We’re getting a lot of variation. (GP1)
GP2 goes beyond the medical examination, considering regularity to maintain driving skill:
If they only drive every three months, I do not give them their licence back. They’ve got to be driving two to three times a week on known routes at 10 o’clock in the morning. I would ask them to avoid peak times. (GP2)
GP5 takes a different perspective, in that the medical examination is about fitness, not ability, to drive. Often couples will come together for their licence medicals.
One of them mightn’t have driven for a decade but will insist on having a medical for a licence. What would that person be like if they ever got behind the wheel? Should I fill that form in? I don’t know, but I do. They’ve got no reason why they can’t – there’s no medical reason why you can’t fill the form in. It’s just their choice not to drive. It’s for emergencies, doc. (GP5)
GP3 already knows who will be re-licenced and who has been forewarned of license restriction or cessation:
80%, when they turn up for their driver’s license form, it’s a 5-min consultation. That’s what they do. The others (forewarned) – you have to enter a very difficult conversation. (GP3)
Standards: Variability in opioid advice
GPs will often prescribe the most common potentially driving impairing medications, opioids.
Say you took someone from a Buprenorphine 10 to a 15 patch? I wouldn’t stop them – I would be happy for that person to drive, but be very, very careful the first 24 hours after they’re on the 15. (GP1)
GP4 takes a completely opposite approach, recommending that older patients cease driving once prescribed a Buprenorphine patch:
Oldies on the patch I wouldn’t let drive. (GP4)
GPR4 commented on the time to wait until driving was safe after taking opioids:
Yeah. Just more how you feel. Yeah. Similarly, on long-term opioids, I would just say it’s going to affect your driving for the next couple of days. Yeah, see how you feel. (GPR4)
GP5 and GPR1 commented on the long-term and short-term use of opioids:
I’ve got patients on opioids, and they’re established on them, they stay on them. As long as they don’t change the dose, they’re fine. (GP5)
(no driving) for prn (as needed) two hours probably. (GP5)
I would normally say you need to leave at least 12 hours for it to run out of your system. (GPR1)
Senior GP3 makes an overall comment about the approach of the driving guidelines towards medication-impairing driving:
The book is very clear on epilepsy and it’s pretty clear on diabetes, but on medication, it’s quite vague, and I think that pushes a lot of, yes, responsibility and probable guesswork back onto the GP. (GP3)
No GPs recounted receiving focused education on mandatory driving medical examinations.
No formal education: Just working with our group (of GPs), it’s what all the GPs would do over the years. (GP1)
Equity and fairness: Mismatch of roles
Registrars with limited GP training find driving license assessments stressful:
It’s an uncomfortable position. I don’t like it. I never did medicine to be a policeman. (GPR1)
I’m not sure I’m the most appropriate person to judge whether a person can drive or not. I would be more reassured with a more practical assessment (on-road) for driver’s licences. Nobody slips nicely into the criteria (of the Austroads guidelines). (GPR2)
(A GP registrar of 18 months standing, with no education and no exposure to this skill in five years of hospital training): Within my first week of GP training I started doing the license drivers medicals (sic). But I think if I’ve got a little bit of training, even now, it would be much more easier for me. (GPR4)
Trainee GPs will be sent new patients, or at least new to them, requiring a driving medical assessment. Senior GP3 takes the registrars’ experience in the matter further:
I will comment, our registrars hate this, hate driver’s licence. They’re green, our registrars; they’re young, they’re green, these are difficult conversations, especially if you’ve got to restrict them (driving privileges). (GP3)
A number of GPs commented that junior doctors might not continue this practice:
The younger GPs won’t put up with this. It’s all grey. … There will come a time when GPs in Australia won’t do it anymore. (GP4)
Younger doctors won’t do this. They are all so risk averse. (GP5)
Discussion
Inequities exist for elderly drivers and GPs. On-road testing has shown to have low agreement with physician’s opinions;18 in this study, it is the GPs underestimating driving ability. Recent Australian research reveals that many GP-referred drivers requiring an SOT on-road assessment had been suffering mild cognitive impairment (MCI) for some time, suggesting the diagnosis had been missed or overlooked for a number of years.19 Drivers who are faced with driving cessation often cannot afford to pay for the SOT review. Senior GPs have the benefit of regular informal assessments and prewarning their long-term patients of impending driving restriction or cessation, easing the impact of this life-altering imposition. Senior GPs in this study did not see new patients; avoiding the medico-legal stress of de novo driving assessments was at least part of the reason.
New patients, often the most difficult to assess in terms of safe driving ability because of the lack of background from a long-term relationship, are still sent to the trainee registrars, who have no formal training and very limited ‘on-the-job’ training. The Austroad guidelines inform on specific conditions such as diabetes and epilepsy; however, in terms of age-related MCI and opioid cognitive impairment, it is vague, leaving the assessment to the GP. Additionally, it has a disclaimer that neither Austroads or the authors accept legal responsibility with their application.12 The most common cognitive test GPRs used (ie MMSE) is a poor predictor of acceptable on-road ability,20 as many patients with mild dementia still drive safely.21,22 Numerous studies have recommended better education of GPs together with validated practice assessment tools.23–25 Additionally, GPs are expected to assess without observing driving.25 For conditions or treating medications that impact driving safety, there is a legal expectation for the assessing GP to advise the patient to self-report to the driver license authority. The legal expectations of the guidelines are quite far-reaching,26 with Beran and Devereux suggesting some GPs are unaware of their full legal responsibility.
What this research adds
The current research supports previous findings about the complexity of this GP role; however, the possible inequities flowing from GPs to patients (variable standards and lack of affordable on-road assessment) and senior GPs to trainee GPs (with unvetted expectation to assess new patients) is new territory (Figure 1). These consultations are stressful,27,28 with participant GP3 observing ‘I think it’s one of the hardest conversations we have in general practice’. Caution is recommended with trainee GPs already under many life stresses and prone to mental health issues.29 The comments made by GPRs indicate they are aware of the enormity of the decision to take away driving privileges. Considering that they have no formal education in this field, it is inappropriate for them to be put through this stress and unfair to the patient. The fact that senior GPs, after a lifetime of ‘on-the-job’ training, will not conduct de novo license medical assessments speaks to how difficult these consultations can be.
GPRs see a different clientele than their supervisors, particularly with respect to new patients.30 Research on the apprenticeship model of GP training stresses the safety of the patient but rarely mentions registrar safety. The concern over the lack of formal guidance with respect to legal liability expressed by many GPs31 is fertile ground for the practice of defensive medicine. The advice given by participants on opioids and driving was highly variable and often discordant with current opinion. It is therefore relevant as opioid impact on a driver should be discussed at the mandatory assessment.
Figure 1. Inequitable pressures on trainee general practitioners.
Limitations
The difficulty of recruiting time-poor, mid-career GPs, noted in a similar study, is a limitation;32 however, the mix of participants in this study highlights the issues for trainee GPs.
Conclusion
The current driver assessment role, which is conducted by GPs is hampered by lack of defined guidelines for dementia, cognitive-impairing medication and undefined medico-legal liability. Senior doctors adapted by only testing familiar patients, and new patients were assigned to registrars. The development of better assessment tools, standardisation of education, improved GPR support, defining GP legal liability and affordable on-road testing would make a fairer and safer system for both patients and GPs.