Intended for healthcare professionals


The role of risk communication in shared decision making

BMJ 2003; 327 doi: (Published 25 September 2003) Cite this as: BMJ 2003;327:692
  1. William Godolphin, professor of pathology (wgod{at}
  1. University of British Columbia, Vancouver, BC, Canada V6T 2B5

    First let's get to choices

    I want to feel like a rational and autonomous person, even when I'm ill. Doctors ought to use their power (legal and knowledge) not only to relieve suffering but to enhance patients' autonomy.1 A prescription for this is shared decision making, a middle ground between “nanny knows best” paternalism and rampant consumerism–an ideal that aims to reconcile the fact of professional power with the ethic of informed choice.

    Laws are leaning towards informed choice. For example, the supreme court of Canada in 1980 ruled that doctors have a legal obligation to disclose, unasked, whatever a reasonable person in that patient's particular position would want to know before making a decision. The ethical positions of medical guilds have more or less followed suit: “Duties of a doctor [are to] give patients information in a way they can understand; respect the rights of patients to be fully involved in decisions about their care.”2 Informing patients and involving them in decisions therefore seem to be a doctor's duty.

    It is within the “black box” of the medical encounter that patients most value information and the doctor's help in interpreting it. A consistent theme in studies is that patients want more information than they get. How often they mean risk information is less clear.

    Risk communication is risky. We are predictably fallible and prone to biases in our judgments of risks and our use of information.3 There are relatively few medical problems for which good risk information is available. Uncertainty changes (as women taking hormone replacement therapy will have noticed recently). Risk evidence rarely includes psychosocial outcomes, although these are important to individual patients. The autonomous patient may not make a “better” medical decision, nor will better decisions guarantee better outcomes for the individual.

    Informed and shared decision making has been called the crux of patient centred care4–an interesting choice of words. On a rock climb, another high risk activity, the crux is the hardest move, the one on which the whole enterprise depends. Shared decision making, of the informed sort, is difficult, and evidence shows that it rarely happens. For example, options and patients' understanding of information and role are rarely explored.57

    This deficiency is often attributed to attitude: “Doctors… provide patients with too little information about the side effects of prescribed medicine… Shortfalls could be corrected by a change in attitudes, rather than a change in structures.”8 But the time available in the consultation and the training of doctors (for example) are functions of structure.9

    Formative medical training, when students are “professionalised,” tends to be in acute care. They are taught to be responsible in settings where choices are few and patients' autonomy is limited. They are rewarded for being confident and getting the “correct” answer. This is unlikely to foster a predisposition to, or practice of, offering choices or enhancing patients' autonomy. They are taught interviewing and history taking but not much about giving patients information or risk communication. Most of our communications skills are habitual and learnt from role models–hard to change even when we wish. Are the most influential role models and opinion leaders also competent at shared decision making?

    The things a doctor ought to be able to do to engage in shared decision making have been proposed.10 The story begins after the problem is identified. It includes explicit clarification of roles; information preferences; discussion of options with reference to the patient's characteristics and the evidence; response to the patient's ideas, concerns, and expectations; and partnership building. A statement of options seems to be a good starting point that may lead to the other elements.

    Interventions that stimulate patient questions result in more involvement.11 Patients have difficulty asking questions–and they attribute this mostly to doctors' traits. They feel intimidated, are concerned about using the doctor's time, and fear that assertiveness will jeopardise rapport.12 A patient's question is a teachable moment. A testable hypothesis is that a doctor's conscientious and judicious search for and offering of choices will stimulate questions from the patient and lead to better information exchange and more involvement, perhaps even a role for risk communication.

    What would happen if “We have some choices and they are…” was in the doctor's habitual script, and “What's the evidence for that, doctor?” in the patient's?


    • Competing interests None declared.