Shared Decision Making and the Experience of Partnership in Primary Care

The idea of “shared” decision-making (SDM) has always made me uncomfortable for several reasons. First, to the extent that SDM is based on a consumerist model of health care, the SDM model seems to lose the importance of relationship. Ideally, patient-physician relationships are not adversarial – they are not 2 parties with different interests presenting their cases before some external standard. Rather, they present multiple opportunities for dialogue and deliberation.1 The decision, then, is only one point during a relationship. It may be that the decision becomes a non-decision – the course of action is obvious to both. Even though this characteristic is most obvious in primary health care settings, all clinicians have ongoing relationships with patients, even if only indirectly through their relationships with other health care professionals. To call such decisions purely “autonomous” is a misnomer – they are autonomous because they occur in the context of a relationship.

Second, it is not clear whether the point of decision is in words or actions because the relationship between decision and action is tenuous. All clinicians have cared for patients who seem to agree with the plan only not to follow through. The patient his or herself may actually believe that the decision is the right one in the moment, but then seems false when placed back in his or her home environment. Patients, as long as they are cognitively and physically intact, ultimately make most decisions.

Third, there must be accessible high-quality information and the capacity to understand and retain it before talking seriously about a shared decision.2 But, information transfer is so flawed that even those decisions that appear shared by subjective and objective criteria may put patients in a position where they are still dependent on the physician’s judgment, willingness to listen and engage, and caring for the patient.

Fourth, SDM is an attempt to level the playing field between clinicians and patients. But, knowledge and power is always assymetrical. It is how that power is used that connotes ethical practice, but let’s not delude ourselves that the power differential is eliminated. And, there are different kinds of power – patients have the power to threaten physicians with lawsuits, the power to not take medications, and the power to bring up topics for discussion. But these manifestations of power are opaque to most research methods.

Finally, patients want to trust their physicians, and value that trust even more when they are extremely ill.3 Dependency is a dirty word to most of us who come from northern European backgrounds, but when frightened and in pain, most of us are desperate to trust, and at times, to be relieved of the burden of deciding. But to deny the value of patient activation is to fall into the opposite trap of promoting mindless paternalism.

The study by Saba et al4 is a landmark in research on patient- physician communication. It is a warning about patients’, clinicians’ and researchers’ capacity for self-deception. Perhaps, as a patient, I feel that I have participated in a decision. But, on viewing a recording, perhaps there was not even a decision made during the encounter. My impression is based on a global sense that this is a good doctor and that I trust him or her. Surveys of airline quality have shown similar patterns – the airline rated highest in the USA was also rated highly on its meal service – even though that airline does not serve meals. Correspondingly, perhaps, as a physician, I feel that I have done a good job because I may feel that I have provided choices. But factors beyond the clinical encounter – health literacy, social status, unconscious prejudice, deferential behavior in the face of power – make those choices not seem real to the patient. Acquiescence is the result. There appears to be no way that researchers can spot this self-deception without asking the patient, and even then, the patient may acquiesce to what he or she feels that the researcher wants to hear. In a study also using process-recall, we found that researchers’ views on interactions between patients at risk for HIV and their physicians rarely corresponded with those of the patient and the clinician.5 This dilemma cannot be fully resolved by creating more sophisticated coding systems.6

The implications for communication research are both clear and expensive. Mixed methods studies are more likely to approximate the nuanced realities that really influence patients’ behavior and outcomes. Outcomes research will determine whether the expense of such studies can be justified by virtue of providing data that is more trustworthy and able to guide principled action.

Reference List

(1) Emanuel EJ, Emanuel LL. Four models of the physician-patient relationship. Jama 1992;267(16):2221-6.

(2) Woolf SH, Chan EC, Harris R et al. Promoting informed choice: transforming health care to dispense knowledge for decision making. Ann Intern Med 2005 August 16;143(4):293-300.

(3) Wright EB, Holcombe C, Salmon P. Doctors' communication of trust, care, and respect in breast cancer: qualitative study. BMJ 2004 April 10;328(7444):864.

(4) Saba GW, Wong ST, Schillinger D et al. Shared decision making and the experience of partnership in primary care. Annals of Family Medicine 4(1):54-62, 2006 January;-Feb.

(5) Epstein RM, Morse DS, Frankel RM, Frarey L, Anderson K, Beckman HB. Awkward moments in patient-physician communication about HIV risk. Annals of Internal Medicine 1998;128(6):435-42.

(6) Epstein RM, Franks P, Fiscella K et al. Measuring patient- centered communication in Patient-Physician consultations: Theoretical and practical issues. Soc Sci Med 2005 October;61(7):1516-28.

Competing interests:   None declared