Issue 25 / 14 July 2014

RECENTLY I wrote about the placebo effect, asking whether there were ethical ways for health professionals to harness its undoubted power in their therapeutic relationships with patients.

There are no easy answers given that placebo appears to work best when the patient believes they are receiving an effective treatment, implying some level of deception is required for maximum benefit to be achieved.

But what about placebo’s evil twin, the nocebo effect — where just the suggestion of side effects is enough to bring on negative symptoms?

A viewpoint article published in JAMA last week suggests contemporary medical practice may unintentionally contribute to nocebo effects, particularly through requirements to disclose potential side effects of treatment to patients.

Just as our expectations of positive effects from a treatment can lead us to experience a placebo benefit, warnings about possible side effects can lead us to experience the harm of nocebo, German neurologist Dr Ulrike Bingel writes.

Although there has been comparatively little research on the nocebo effect, a 2012 review implicated it in symptoms as varied as wheezing, headache, sleep problems, back pain, nausea, constipation, itching and vision problems.

In one study examining the use of finasteride in benign prostatic hyperplasia, men who were warned the drug could cause sexual side effects (erectile dysfunction, decreased libido or ejaculation disorders) were significantly more likely to experience those symptoms (44%) compared with men who did not receive a warning (15%).

Another small study found gastrointestinal symptoms were induced in 26% of people with confirmed lactose intolerance as well as 44% of those reporting symptoms of lactose intolerance but with negative test results, when they were given sugar pills they believed contained lactose.

Interestingly, in her JAMA article Dr Bingel suggests switching from a branded drug to a chemically identical generic can also be associated with an increase in adverse effects, perhaps because patients believe the generic drug is somehow inferior.

Nocebo is not just responsible for unwanted symptoms. When patients hold negative expectations about a drug’s performance, this can also lead to a reduction in efficacy, Dr Bingel writes.

In one study of rizatriptan for episodic migraine, for example, the drug provided significantly less effective pain relief when it was falsely labelled as a placebo.

So where does this kind of research leave doctors and other health professionals?

Dr Bingel believes it is time clinicians took the nocebo effect seriously.

“Physicians should be aware of ways by which they unintentionally induce nocebo effects and be familiar with strategies to prevent or minimize nocebo effects”, she writes, suggesting this should also be part of the curriculum for future doctors and other health professionals.

Her suggested strategies for doctors basically come down to establishing good communication with patients, including emphasising treatment benefits alongside any discussion of possible adverse effects and avoiding use of technical jargon or language that is likely to induce fear.

Many doctors probably follow those principles already, even if they don’t think of them specifically as anti-nocebo measures.

The authors of the 2012 review go further, though, suggesting doctors could ask patients whether they want to be informed about potential non-serious side effects, in light of evidence that being told about them might actually increase their likelihood of experiencing them.

The old saying “What you don’t know, can’t hurt you” has obvious flaws, but it perhaps holds some truth when it comes to the nocebo effect.

Jane McCredie is a Sydney-based science and medicine writer.

4 thoughts on “Jane McCredie: Negative power

  1. Randal Williams says:

    A mentor of mine used to say ” ask the patient how much information they want’ . Some want to know everything while others find too much information anxiety-provoking. This is particularly true of potential complications of surgery. If the patient doesnt want to know I record this in my case notes. It applies also to side effects of any treatment, but some are too important not to mention eg potential gastric side effects of NSAIDs. Good communication includes establishing the level of detail the patient wants.

  2. Paul Dessauer says:

    An interesting example of nocebo from an Australian “naturalistic experiment”;

    When a batch of methadone syrup was released that accidentally contained less caramel (colouring agent), 1/3 of the patients reported experiencing opioid withdrawal symptoms.

    When it was tested, the methadone syrup was found to contain exactly the same dose in mg of methadone per ml of syrup. It was just much lighter. It looked like the normal syrup diluted 50/50 with water.

    It appears that patient perception that the syrup was diluted was enough to precipitate physiological withdrawal responses in 34% of the patients.

    Administration of a lighter-coloured methadone liquid. Byrne A, Hallinan R, Love A. Drug & Alcohol Review [letter] 2002 21;4:405

    On nocebo, see also ,

  3. Keith Charlton says:

    On rare occasions I have applied a placebo treatment for spinal pain therapy (de-tuned ultrasound…. the ticker was ticking solicitously, but the machine was off). I justify this ethically by declaring to the patient that since nothing else is working (medication, exercise, physio, chiro etc) there is another approach yielding about 30% success rates in such circumstances, and it’s quite free of nasty side effects. It seems Beecher’s et al estimate of placebo in pain works. About 30% of the time.

  4. Dr Philip Dawson says:

    the placebo effect is quite useful where the patient insists on some potentially dangerous treatment the doctor doesnt want to give. You can ethically suggest anything you like to the patient, eg try calcium  and vitamin D tablets they might help your back pain. You are not lying , they “might”, and they are certainly better than methadone for the patient!

    Equally the nocebo effect is very useful if you dont want a patient to take  a dangerous drug eg valium. If they are so upset about something eg bereaveent,separation etc but you suspect they are the type of person who may become addicted, it is quite reasonable to say you are going to give them something to calm them down and help them sleep, but they mustnt take them more than a week because they might stop working (tolerance), and they might become addicted to them. This is generally enough that they only take them a couple of days, and often they come back with stories of others they know who are addicted to them, and “I dont want to end up like that, doctor”.

Leave a Reply

Your email address will not be published.