Expertise and Autonomy in Medical Decision Making

Written by Rebecca Brown.

This is the fourth in a series of blogposts by the members of the Expanding Autonomy project, funded by the Arts and Humanities Research Council.

This blog is based on a paper forthcoming in Episteme. The full text is available here.

Imagine you are sick with severe headaches, dizziness and a nasty cough. You go to see a doctor. She tells you you have a disease called maladitis and it is treatable with a drug called anti-mal. If you take anti-mal every day for a week the symptoms of maladitis should resolve completely. If you don’t treat the maladitis, you will continue to experience your symptoms for a number of weeks, though it should resolve eventually. In a small number of cases, maladitis can become chronic. She also tells you about some side-effects of anti-mal: it can cause nausea, fatigue and an itchy rash. But since these are generally mild and temporary, your doctor suggests that they are worth risking in order to treat your maladitis. You have no medical training and have never heard of maladitis or anti-mal before. What should you do?

One option is that you a) form the belief that you have maladitis and b) take the anti-mal to treat it. Your doctor, after all, has relevant training and expertise in this area, and she believes that you have maladitis and should take anti-mal.

An alternative option is for you to refrain from believing you have maladitis, and to refuse the recommended anti-mal. Your doctor might, after all, be wrong – doctors sometimes make mistaken diagnoses, and recommend inappropriate treatments. Sometimes people experience bad outcomes as a result of following medical advice.

One reason in favour of accepting what your doctor tells you is that, if you do this, it is more likely you will form a true belief. This is an epistemic reason, since it has to do with knowledge and beliefs. The doctor, in this instance, counts as an expert with regard to your illness and the appropriate treatment for it. She is therefore more likely to hold true beliefs about whether or not you have maladitis than you are. Some people go so far as to say that, when you encounter someone you regard as an expert on a given topic, you should immediately adopt all of their beliefs on that topic. The reason is the same as in the maladitis case: the expert will be more likely to hold true beliefs (on the topic on which they are an expert) than you are, so if you adopt all of their beliefs (on that topic), your beliefs will come closer to the truth.

It doesn’t need to be the case that the expert has entirely true beliefs. It might be that there are a few matters within their domain of expertise where you have a true belief and they have a false belief. But since it is very hard to be sure when this is the case, you shouldn’t pick and choose which of the expert’s beliefs to adopt as your own. You just have to accept all of them on the basis that, on average, they will be closer to the truth than your existing beliefs.

One implication of this is that, if you care about having true beliefs, you should avoid trying to do your own research, or reasoning from first principles on matters where you lack expertise. Instead, you should just adopt the beliefs of an expert. You shouldn’t try to supplement the expert’s beliefs with your own beliefs – this will only make them less likely to be true.

In the domain of medicine, then, it looks like patients should just adopt their doctors’ beliefs on medical matters (unless the patient themselves is a medical expert, but I shall set such cases aside). Why, then, is there so much emphasis on patient autonomy? This is the idea that patients should be self-governing. Patients should be provided with information and perhaps advice about medical decisions they need to make, but there is value in them being the one to make those decisions. But from the discussion above, it looks like patients should just always hold the same beliefs as their doctor.

On this account, if patients exercise their autonomy to form different beliefs from those held by their doctor, they are more likely to be wrong. Is there inevitably a trade-off between epistemic value (having true beliefs) and ethical value (being autonomous)?

Well, philosophers suggest that having true beliefs is actually quite important for autonomy. This means that having completely mistaken beliefs will make it very difficult for you to be autonomous. This is because autonomy involves living your life as you choose. But if you have false beliefs then, whilst you might think you are making choices that will promote the things you care about, you might actually be failing to do so. Imagine you choose white ice cream thinking it is vanilla (which you like) but discover it is lemon sorbet (which you hate). Your autonomy in this choice was limited by your false belief.

In the medical context you will need at least some true beliefs in order to be able to make autonomous decisions. And the best way of getting true beliefs is by deferring to medical experts – e.g. your doctor. We might think about relying upon what doctors tell us in order to form our beliefs is similar to relying upon any other kind of evidence. So, if doctors are experts then their beliefs count as good evidence for what we should believe.

Great! This suggests there is no conflict between deferring to doctors’ expertise and being autonomous decision makers. But there is a problem with this picture: it assumes that doctors (and other medical professionals) really are experts in the relevant way. Is this the case?

In some sense, yes. Doctors receive medical training and have professional experience that means they have a better understanding (on average) of matters relating to medicine and health than patients do. But there are a few reasons for thinking that doctors’ expertise is weaker than we might expect.

  1. The fact/value distinction
  2. Evidence and understanding
  3. Communication skills

First, to the extent that doctors are expected to be experts in medicine, this is usually understood to be an objective, scientific discipline. But there are many ways in which medical research and practice involves many ‘value’ judgements, ones which patients might (reasonably) disagree with. It is well understood that patients bring relevant knowledge to decision making about their care. What is less frequently acknowledged is that even ‘scientific’ judgements – how many participants to recruit to a clinical trial, what scale to use to measure changes in depression – involve reflection about values. It is not clear that clinicians (or clinical researchers) are well-placed to make judgements about these kinds of values.

Second, doctors’ understanding of what we might take to be ‘medical facts’ might not be as good as we expect. Whilst they are probably better at reading clinical trial evidence than the average patient, they might not be that good at it. For instance, doctors receive minimal statistical training and often lack basic numeracy. This hampers their ability to, for instance, interpret screening results appropriately.

Third, doctors may lack the skills needed to communicate the significance of medical evidence to patients. Poor judgement regarding informational needs can lead to missing out relevant information or ‘dumping’ unhelpful and irrelevant information on patients.

Although trust in doctors and healthcare professionals is generally high, it seems that their expertise is probably more circumscribed than we would typically expect. Whilst deferring to experts is perfectly consistent with autonomous decision-making, deferring to experts who are not as expert as we think might be more of a problem.

A couple of solutions present themselves here: First, we could make doctors better experts! This would involve more training in interpreting clinical evidence, statistics, communicating risk information to patients, and so on. This will of course have costs in terms of the expanded training needs of doctors. Second, we could encourage more realistic expectations of doctors’ expertise, to avoid a mismatch between patients’ beliefs about the extent of doctors’ skills and evidence, and the reality. What doesn’t seem realistic (or desirable) is to try to avoid patient dependence on doctors’ expertise entirely. For most patients, most of the time, the best we can do is defer to our doctor.

  • Facebook
  • Twitter
  • Reddit

One Response to Expertise and Autonomy in Medical Decision Making

  • Pavel Novak says:

    According to law the doctor as an expert is responsible for the treatment that must be on the proper expert level.
    Doctor is obliged to inform the patient about the treatment, its risks, advantages/disadvanteges, alternatives and about limitations within which the treatment will influence the patient’s life.
    Doctor is also legally obliged to make medical records included the informed patient’s consent/dissent.
    If doctor accomplishes all these legal duties then it is exclusively patient’s decision about his/her health.
    So despite the doctor bears a lot of legal duties during medical treatment providing, the relationship “doctor-patient” is still private and health is private value.
    The patient must decide about treatment that is on proper expert level recommended by doctor.
    So again if the doctor observes all his/her duties then it is patient’s full responsibility to decide.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use the <em>, <strong> and <blockquote> tags. Links have been disabled to combat spam.

Notify me of followup comments via e-mail. You can also subscribe without commenting.

Recent Comments

Authors

Affiliations