Shared Medical Decision Making: Between Autonomy and Authoritarianism

.
One area that has been a “hot” topic of conversation in the medical literature, among “patients”, and even among physicians, is “shared decision making”. This means that decisions about a person’s health are made with input both from the person (or their decision makers, if they are unable as in the case of children and people with dementia) and the physician. An initial response might be “Why shared? Why should not the patient make the decision?” A physician may be involved in explaining the options, discussing the positives and negatives, risks and benefits and costs, of each alternative, try to put the issues and words into “lay” language, even suggest what s/he (the physician) would do (or thinks s/he would do!) in the same situation, but should not the decision, ultimately, be up to the person?

Alexander Kon, in a recent article in the August 25, 2010 issue of JAMA, “The shared decision making continuum”,[1] helps us to understand this. He begins by describing the trajectory of decision making, moving from a long-standing “paternalistic” physician-centered model to, in the 20th century (and I would say late 20th century) a patient-centered “autonomy based” standard in the US. He notes that in the US the pendulum is swinging back toward the middle, to shared decision making, and that it is also moved in this direction in many other countries that never moved from paternalism to autonomy. Kon, as promised, presents a continuum, from 100% patient-driven (autonomy) through “physician recommendation”, “equal partners”, and “informed nondissent”, to 100% physician-driven decision making. This makes sense, as rarely, even in the most paternalistic systems, and certainly even in the most “autonomous” that have developed, is the decision making truly at either extreme. Kon goes into some detail about each option, which are best understood as stages; however, “informed nondissent” may require some further explication. In this model, the physician takes on the bulk of the decision making, and explains it to the patient, who has the right to demur (dissent). Given that information is required to be able to dissent, for this to be different from the extreme of “physician decision”, the patient has to be in possession of all pertinent information, and also has to be aware of the ground rules – that “nondissent” (silence) implies tacit agreement. “Patients must understand that they are welcome to veto the decision and if so, their wishes will be honored and they will receive excellent care.”

Kon further discusses the uses of different approaches in different settings for the same patient-physician relationship. He suggests that “physician responsibility” may be appropriate for “value-neutral” decisions, but goes on to caution that what seems value-neutral to the doctor may not be to the patient and provides several examples of this. He suggests that a very narrow definition be applied – such as “what size endotracheal tube* to use.” Kon diverts a bit from the framework he has developed when he says “Patient preferences must guide the approach used,” for in saying this he is taking a stand against a truly physician-centered approach. I agree with him, but it must be acknowledged that there are many settings and many physicians, both in the US and abroad, who use a physician-centered model, and to whom the question of “what is the patient’s preference?” never occurs.

Of course, even in these settings “patient preference”, and indeed “patient decision making” plays a major role. Kon’s examples of the size of the endotracheal tube used, the hand in which an IV might be inserted in a child, and even decisions regarding end-of-life, are understandable given that he is in the field of pediatric critical care. However, for the primary care physician – or indeed any physician or practitioner caring for “outpatients” (the medical term for “people who are not in the hospital”), it is clear that patients all make decisions all the time.

Some part of ambulatory care, in which a “procedure” of some type is to be performed (an immunization, a Pap smear, a minor surgical procedure, even an x-ray or lab test) resembles inpatient (hospital) care in that it is done by the health care provider and the role of the patient is to agree or not agree to it. However, in most of ambulatory care the physician acts solely as a consultant, suggesting behaviors that would be beneficial, suggesting what medicines to take (and not take) and how often and for how long. The advice, for medication at least, is transferred to a prescription that serves the shared conception that the doctor is actually doing something. In fact, of course, it requires the patient (or their agent) to take the action of bringing that paper to a pharmacy, paying money to fill it, and taking it – as prescribed – for the suggestion to be turned into an action that might improve the person’s health. Even in the most physician-centric model: “Here,” says the physician, writing a prescription, “take this three times a day,” without any discussion of whether the patient wants to, has questions, is concerned, or anything else, the patient has autonomy and ultimately makes the decision. This may be because the patient has a different understanding of what “three times a day” means (up to three times a day? Every day? When I have a symptom? What symptom? For how long?). Or it may be because the patient is getting different advice from someone else (family member, friend, other health provider), or because the patient sees, correctly, the physician as advisor rather than dictator.

None of this gets to the issue of cost, and who pays, which further complicates the question. Sometimes there are treatments that the physician, or the patient, or both agree, is necessary and/or appropriate, but which the patient cannot afford. Sometimes the physician, or the patient, or both agree to because it is affordable (or the insurance company is paying for) when it is not necessary and/or appropriate (or even because the physician will be reimbursed for it). Thus, well-insured people may get treatments they don’t need (and may even harm them), because the doctor, patient or both want to try it, while others may not get treatments they definitely need despite doctor, patient or both recognizing that need.

I agree with the concept of shared decision making, not because I am opposed to patient autonomy, but because of the risk that physicians will, under the rubric of autonomy, not fulfill their obligation to make sure that the patient has the benefit of all the physician’s knowledge and understanding in making their decision. To say “here are the choices, it is your decision” about something as complex as many health decisions are, may seem a more efficient use of the physician’s time, but it is irresponsible. In addition, there are many circumstances (in addition to cost) in which a patient decision to have a treatment should not be honored (when the likelihood of benefit is small or nil, and/or the likelihood of harm is significant). Ultimately, people should make the decision on whether or not to take a recommended treatment, but it is the responsibility of the provider to ensure that they have all necessary information that makes the process “shared.”

*This is a tube inserted through nose or mouth into the airway to help someone to breathe, usually attached to a “breathing machine”

[1] Kon, A, “The shared decision making continuum”, JAMA 25Aug2010;304(8):903-4.