Medical illness and fitness to drive – the doctor’s dilemma

As physicians, we believe that we have a fiduciary duty to our patients; that our prime directive is to serve them in good faith and to advocate for them. At the same time, society expects us to protect them as well in matters of public safety. When these two roles come into conflict, as they sometimes do, the physician is placed in a very uncomfortable position.

One issue that poses this conflict in very stark terms is the assessment of medical fitness to drive.

In 7 Canadian provinces, 6 US states and most Western European countries, legislation has been enacted that requires physicians to report to the regulatory authorities any patient whose medical condition may make them unsafe to drive. If the licencing authorities then determine that the medical condition precludes safe driving, a suspension of the patient’s driver’s licence is issued.

In Canada (Alberta, Nova Scotia and Quebec excepted), provincial legislation is in place requiring us to report patients with medical illnesses that may make them unsafe behind the wheel. Because the operative word in most legislative language is “may” – everyone may be at an increased risk – the Canadian Medical Association has developed guidelines to aid physicians in this somewhat unpleasant task. Everything from vision impairment, cognitive impairment (usually dementia) and mental illness through to syncope, seizures, and heart failure are on the list. The guidelines provide, on a condition-by-condition basis, recommendations as to who should be suspended from driving. Physicians can consult with these guidelines to aid their judgment and support their decisions.

As a co-editor of the CMA Driver’s Guide (the 8th Edition is now available at http://www.cma.ca/driversguide), I can tell you that developing the guidelines was exceedingly difficult. In many instances, evidence was lacking and we had to make recommendations based on consensus and general principles. There are no RCTs comparing patients with medical illness who are randomized to drive or not to drive. Even if there were, the measurement of risk at the individual patient level is a very inexact science – there are just too many variables.

The other difficulty, of course, is that we are all acutely aware that by making an effort to comply with legislation and protect society, we are necessarily working against our individual patients. The official view is that driving is a privilege, not a right. But we all know how important driving is to people – taking a person’s licence away adversely affects their sense of independence and can cause considerable economic hardship. Indeed, many patients earn their living as truck drivers, taxi drivers, or they need to be able to drive in order to go to work, buy groceries, or attend social and family functions. It seems that driving should be something more than just a revocable privilege, even if it can’t be deemed a full-fledged right.

Given these competing interests and potential harms, the onus is on physicians to be thoughtful in our fitness to drive assessments. But I would also like to advance the argument that the scientific community has some work to do. We need more evidence so that we can make decisions with more confidence.

The first thing that needs to be sorted out is this: What is the magnitude of the problem? Of all motor vehicle accidents, how many are caused by a driver with a medical illness? More to the point, how many are caused by a medical illness that could have been foreseen?

The limited data we have from police report databases is that less than 1% of all accidents are caused by some sort of medical event behind the wheel. Of those, most were as a result of a sentinel medical event; in other words, they occurred in drivers who would have been previously assessed as having been “low-risk” and fit to drive.

We need better data, but this alone makes me wonder what the efficacy of mandatory physician reporting really is. Even if we could prospectively identify every single driver who was destined to to cause an accident because of a medical event (we can’t), we would evidently only reduce the total number of MVAs by 1% or so. Is that worth the harm we do to the lives and livelihoods of thousands of suspended drivers? Is it worth the cost of maintaining the expensive Ministry of Transportation infrastructure to carry out these evaluations and licence suspensions? What is the cost per life year saved? We don’t know the answers to any of these questions.

Sometimes I even wonder if we are doing more harm than the good that is intended, at least in some instances. In my practice, I see a lot of patients with syncope. Some require licence suspensions according to the guidelines. (They are never very happy about that!) The guidelines allow for resumption of driving if a certain period of time goes by with no recurrence.

There is some pretty good evidence that in mandatory physician reporting jurisdictions, patients with intermittent conditions like syncope or seizures will deceive their physicians. When they know that their licence will be affected based solely on the answer to the question, “Have you had any episodes in the past 6 months?” they have a powerful incentive to lie. When this happens, the patient not only continues to drive, but they are suboptimally treated for their potentially disabling condition! Does mandatory physician reporting paradoxically make the roads less safe in these cases?

And what about the notion that drivers may “auto-adjust” their behaviours based on the knowledge that their condition makes driving a bit more hazardous? We hear stories from our senior patients all the time about how they don’t drive on the highway anymore, or at night; or they only drive to and from the store. Most of us have a “risk set point” wired in our DNA – if we sense that the risk is heightened, we adjust other behaviours to bring that risk back to baseline. Indeed, Don Redelmeier’s recent NEJM paper (http://www.nejm.org/doi/full/10.1056/NEJMsa1114310) showed nicely how a simple warning to patients from their physician about their risk to drive significantly reduced subsequent MVAs – largely without the aid of formal Ministry-sanctioned suspensions.

Finally, there is a fundamental question of fairness. What is the acceptable risk we should accept? Many of you will be surprised to learn that no one has ever defined this. Clearly, we don’t demand that all drivers should have the same risk – it’s well known that 18 year old men and 80 year old women are at higher risk for MVAs – the insurance industry understands this well – they charge higher rates as a result. Why are these drivers at higher risk allowed to drive whereas patients with some medical conditions are not?

Let’s make a few comparisons. Smokers are 1.5 times more likely to be in an MVA than non-smokers are. For migraine sufferers, the number is 2.5. Diabetes makes you 1.8 times more likely. Do you take a tricyclic antidepressant? Your risk is 2.3 times higher than average.

And how about this one: women with epilepsy are less likely to be in an MVA than men without epilepsy.

Maybe we should suspend the licences of all men!

You can see what I am getting at. Many of us belong to a demographic group that is at higher risk than some other group. Where should the cutoff line be? After all, if we suspended everyone from driving, there would be no accidents at all.

This isn’t to say that mandatory physician reporting is the wrong approach. It’s only to say that we don’t know if it’s the right approach. We don’t know if it’s working and we don’t know what it costs.

Until we have better data, physicians can take comfort in the published guidelines – we’ve done our best to give you the best advice with the evidence we have. Visit http://www.cma.ca/driversguide. I’d love to hear your feedback!