In Brief: New Evidence Shows Expectations Influence Pain

In the current issue of Science Translational Medicine, Oxford neuroscientists in Irene Tracy’s lab have published a new study of the placebo effect with dramatic results.

In their experiment, test subjects were subjected to pain in the form of heat, while inside an fMRI brain imaging machine, and asked to rate their subjective feelings of pain. After the pain was induced, a powerful opiate analgesic drug, remifentanil, was administered by a covert injection, leading the subjects to report a slightly lower level of pain. Next, they were informed that a drug had been given, and their reported pain fell much more. And finally, they were told (falsely) that the drug had been withdrawn, upon which their pain returned to the level it was at before the drug was injected.

The experiment is significant for three reasons: first, it provides the strongest evidence so far that our expectations of benefit significantly contribute to our experience of suffering. Second, it shows that the benefit of a powerful modern painkiller can be completely eradicated if the patient believes she is not getting a drug. And finally, the experiment’s fMRI data showed that the positive effects were associated with the same pattern of brain activation as inert placebos, but that the negative effects were associated with activity in unrelated parts of the brain that are associated with increase in pain due to anxiety. This last point proves that the total elimination of the drug’s effect was not simply due to the removal of the beneficial placebo effect—rather, our minds have the ability to entirely block the beneficial action of a real analgesic drug.

The experiment has deep implications both for the neuroscience of pain and for philosophical questions in phenomenology. But it has immediate implications for the ethics of clinical practice as well. It is essential that we start to understand the power that our expectations have over our experiences. Prescribing a patient a pill for pain is simply not enough on its own; in fact, if the patient expects the pill not to work, it might be better not to prescribe anything at all.

  • Facebook
  • Twitter
  • Reddit

6 Responses to In Brief: New Evidence Shows Expectations Influence Pain

  • matt baum says:

    I would be interested to know whether endorphins (endogenous orphins) are secreted in response to the expectation of pain relief and thereby act via the same mechanism as the opiate. Also, would be important to find out whether the return to baseline subjective pain happens because the anxiety cancels out the opiate effect competitively (and thus could be overcome by more opiate) or non competitively (in which it effectively short circuits the opiate analgesic pathway). Interesting results regardless.

  • Anthony Drinkwater says:

    “If the patient expects the pill not to work, it might be better not to prescribe anything at all.”

    To take an analogy :
    Child on scramble with parents shows severe signs of anxiety on crest of hill
    Parent attaches rope to child, child’s confidence improves and the family continue their scramble
    Older brother shouts “mum’s not really holding on the rope, you baby!”
    Young child returns into deeply anxious state

    Conclusion : if the child expects the rope not to work, it might be better not to use it

    This, of course, might be true; but does it follow ?

  • Peter Wicks says:

    An even more interesting question from my perspective (than whether it is better not to use a pill that the patient doesn't expect to work) is how one can harness the placebo effect for positive ends, in a way that avoids deception. By "deception" I mean trying to convince others, or indeed our future selves, of something we (currently) believe to be false.

    The answer in my view is to focus on positive, self-fulfilling ideas. If there is one lesson to be drawn from this research, it is that beliefs such as, "This is going to make me/you feel a whole lot better," are to a very significant degree self-fulfilling. The trick, I think, is to formulate them in such a way as to maximise the placebo effect while avoiding falsehood. For example, a belief such as, "I'm never going to feel unhappy ever again" is probably going to turn out to be false, no matter how intensely I believe it, so it's probably best not to focus on this type of idea. An alternative view would be that a degree of falsehood might be a price worth paying for the benefits of the placebo effect, but such an approach can be dangerous.

  • Julian Savulescu says:

    This is indeed striking and important research.
    I liked Anthony's analogy.
    I think it is also important to put this kind of research into perspectives It occurs in an artificial experimental settings using very minor noxious stimuli. Back in 2001 I had a very badly broken leg with a compartment syndrome requiring several operations. For the first couple of hours, I received essentially no pain relief. I was very focussed on making sure I did not lose my leg. When I finally got a big dose of morphine, I suddenly realised I had been in severe pain. It was an amazing relief. One of the great triumphs of modern medicine is the control of pain. After that point, I never experienced pain from this fracture or its complications. That was not due to my expectations or mental state, but due to the excellent pain control of the analgesics that were employed.

    I am sure that our mind has a great influence over the pain we experience. But we do know a lot about the physiology of pain and have made great achievements in pain control. Pain is not "all in your mind." It is dangerous, as Anthony suggests, to conclude that analgesics should not be prescribed if a patient does not expect them to work. Our expectations can also be mistaken. If a patient did not expect anaesthesia to work, this would be no reason not to provide anaesthesia.

    Our attitude to life may influence our experience of it. But it does not fully determine it.

  • Peter Wicks says:

    I think I agree with all of this, although when I read, "That was not due to my expectations or mental state, but due to the excellent pain control of the analgesics that were employed," I did think, "How do you know?" I think it's overwhelmingly likely to be true but I guess you'd need a controlled experiment to be 100% sure.

    Another caveat I want to add is that the extent to which an analgesic (or other drug) is effective independent of the expectations of the patient will presumably depend on the power of the drug. So while I agree that it is dangerous *in general* to conclude that analgesics (or indeed other drugs) should not be prescribed if a patient does not expect them to work, there may be good arguments for withholding such treatments in cases where the actual chemical effect of the drug is incremental (as in paracetamol) rather than determinant (as in general anaesthetic). One such argument could be that it gives the patient a (justified) sense of control over treatment, which is likely to be therapeutic in itself.

  • Dennis Tuchler says:

    The big ethical issue, raised in Peter Wicks's first comment is this: may the physician administering the anesthetic lie about its effect, in order to harness the maximum self-anesthetic response of the patient through raising expectations?

    Many years ago, physicians in the US and, I suspect, the UK lied to their patients in order to prevent distress or enhance the patient's confidence in the doctor's work. This is no longer the ethical situation in the medical profession. Does the role of patient expectation in the success of anesthesia make it proper to return to the old standard, at least when the belief induced by the lie tended to enhance patient comfort and perhaps the patient's progress to a better situation?

    What if the physician is a bad liar, and the patient, catching the lie, now becomes distressed and thereby causes the treatment to fail? Malpractice?


Subscribe Via Email