Intended for healthcare professionals

Editor's Choice

Communicating risk: the main work of doctors

BMJ 2003; 327 doi: (Published 25 September 2003) Cite this as: BMJ 2003;327:0-f
  1. Richard Smith, editor (rsmith{at}

    “Dr Smith, your serum potassium is at the upper limit of normal.”

    “What does that mean?”

    “Nothing really. You shouldn't worry.”

    “Well, why did you tell me?”

    “We thought you wanted to be kept informed.”

    Many doctors are not good at communicating about risk–yet increasingly it is one of their central tasks. Readers have asked us to produce this theme issue because they would like to be helped to do better. This is an issue, I suggest, that deserves perhaps two hours' reading–as opposed to the more usual 30 minutes. All doctors–including those in laboratory based disciplines and public health–have to communicate risk to people.

    This has become especially important because of the changing nature of the doctor-patient relationship. When doctors made decisions for patients–as many still do–they didn't need to communicate risk. The doctor would decide on a treatment and then help the patient feel good about it, perhaps–with good intentions–slightly exaggerating the benefits and playing down the risks. Numbers were not involved. Even the “calculation” of the risk benefit ratio was internalised: doctors' experience told them what to do.

    Increasingly this is not good enough. There is a need for numbers, and many doctors don't feel easy with numbers. “Can you,” asks Tze-Wey Loong, “explain why a test with 95% sensitivity might identify only 1% of affected people in the general population?” (p 716) My guess is that not one BMJ reader in a thousand could answer that question, but the numbers are in many ways the easy bit. The communication is the harder bit.

    There is an increasing array of aids and tools for presenting the numbers (p 736 and p 741), although they can't overcome the problems of uncertainty and of moving from populations to individuals. The low point in risk communication in Britain was a government minister feeding his young daughter a hamburger and assuring the population that beef was “perfectly safe” (p 726). Uncertainty was swept aside, the public was patronised, and trust badly damaged.

    Trust is the key to communicating risk–as it is to so much. Lying destroys trust, but deluging patients with numbers doesn't build it. Several contributors point out that we don't think about risk rationally. A risk is a combination of a probability of something happening (where statisticians might be able to help you but often can't), a feeling of the dreadfulness of that event (which is very personal), and a context for the event. To improve communication of risk, write Andy Alaszewski and Tom Horlick-Jones, doctors must build trust, be aware that patients have many other sources of information (including some they may trust more than doctors), and be sensitive to the psychological and social factors affecting patients (p 730). Things good doctors do all the time.

    To receive Editor's choice by email each week subscribe via our website:

    View Abstract