Intended for healthcare professionals


Talking to patients about cancer

BMJ 1996; 313 doi: (Published 21 September 1996) Cite this as: BMJ 1996;313:699
  1. Robert Buckman, Medical oncologist
  1. Sunnybrook Regional Cancer Centre, Toronto, Ontario, Canada M4N 3M5

    No excuse now for not doing it

    An extraordinary thing about the medical profession is that it managed to avoid telling the truth to patients for so many centuries. In 1672 the French physician Samuel de Sorbiere considered the idea but thought that it might seriously jeopardise medical practice and concluded that it would not catch on.1 This approach—recognising the need to act but finding excuses not to—is still in evidence today, even though there has been a considerable change over the past 35 years. Two papers in this week's journal remind us that what is comfortable for clinicians is not necessarily what patients want.

    In 1961, a landmark paper by Oken showed that 90% of surgeons in the United States would not routinely discuss a diagnosis of cancer with their patients.2 However, several studies subsequently showed that a growing proportion of cancer patients wanted to know their diagnosis. The proportion of patients wanting to know has varied in different studies, depending on the locale and the sample (Northouse has done an excellent overview3, but has often been in excess of 90%.

    Medical practice has gradually changed to meet patients' needs, particularly in the United States, and nearly 20 years after the Oken study, Novack et al repeated the survey. They showed that by then, there had been a sea change in clinical attitudes, with more than 90% of American physicians saying that they would tell their patients if they had cancer.4 In view of this, the crucial question in truth telling should perhaps have moved on from whether to tell patients, to what to tell them—and how. Sadly, this is not yet the case—certainly not in many areas of Europe and in Britain. A survey of European gastroenterologists in 1993 showed that nearly 60% of responders did not routinely tell their patients of a cancer diagnosis if the patient did not ask.5 A survey of British general practitioners and hospital consultants in the early 1980s showed that 75% and 56% respectively did not routinely tell their patients.6 By contrast, in the United States (and to a slightly lesser extent in Canada), medicolegal practice has now enshrined the patients' rights to truth, not only in the codes of ethical practice but in case law.

    Of course, this has disadvantages; being told the truth (without regard to your own wishes, the manner in which the news is imparted, or the underlying nature of the doctor-patient relationship) can be just as upsetting as being lied to. Centuries of systematic insensitive deception cannot be instantly remedied by a new routine of systematic insensitive truth telling.

    It is not hard to understand the main reasons for avoiding sharing bad news with patients. It is unpleasant to be the bearer of bad news and to have to support and assist patients as they absorb and understand the nature of their medical situation. Traditionally, clinicians have found two main justifications for keeping patients in the dark. Firstly, the facts might upset them. This is undoubtedly true, but (legal considerations aside) that line of reasoning has never been acceptable to any other profession in which news might be bad—accountants, bankers, magistrates, stockbrokers, building societies, and so on. Secondly, they presumed that patients don't really want to know.

    Two papers in this week's BMJ remove the last vestiges of rationality from both arguments. Meredith et al (p 724) studied 250 patients attending an oncology centre in Scotland; 79% of patients wanted as much information as possible and 96% specifically wanted to know if their illness was cancer.7 Almost all patients wanted to know the chance of cure and the side effects of treatment. In Benson and Britten's smaller study (p 729), patients reported feeling that they had certain rights to have information about their diagnosis and that they should have a say in who else was told.8 All patients felt that family members should be told if the patient had given permission, but nearly two thirds felt that if the patient did not wish the information given the family should not be told, and that information should not be released without consent except in certain circumstances.

    Important as these results are, they cannot tell us how to break bad news. Almost all of us still feel some (often much) discomfort during an interview about cancer, and perhaps anxieties about technique are behind most arguments for not telling the truth. Most of us have not been taught how to break bad news sensitively and supportively, and as in all branches of clinical practice we all have a tendency to avoid doing any procedure if we are uncertain of our expertise. Yet over the past decade or so, such expertise has become available to almost anyone who wants to acquire it. There are courses in many medical schools,9 there are books10 and videos11 that describe in detail the practicalities of breaking bad news (including ways of finding out whether the patient wants to know or not), and there are now courses for practising oncologists that use role play and video recordings to show how to learn and improve the techniques.12 There is even some material available for patients to help them get the best out of communications with the doctor,13 although we should not need our patients' help in order to help them.

    Nowadays there is no excuse for physicians who simply don't want to perform this important part of our job. It needs to be done and it can be done. It has been said that if the breaking of bad news is done badly, patients and their families (or often their lawyers) may never forgive us, but if it is done well they will never forget us. As this week's two papers show, our patients want, expect, and need the truth from us. There are simple, practical techniques that we can learn and use to tell our patients the truth if they want to know, and we no longer have any excuse for avoiding it.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    12. 12.
    13. 13.
    View Abstract