Placebo Analgesia

One of the best workshops I attended at the IASP congress in Milano was on placebo analgesia, run by three very classy speakers: Luana Colloca, Ulrike Bingel, and Regine Klinger.

I learnt lots about the neurobiological mechanisms of placebo and nocebo, and came away with lots to ponder.  This is what I learnt:

The rostral anterior cingulate cortex (rACC) is active in the placebo response, and may be part of the descending pain-modulating system, particularly given its connections with the periaqueductal grey (PAG)—a structure known to be involved with descending pain modulation.  Interestingly the coupling between these structures is inversely correlated with pain-related activity in the primary somatosensory cortex.  And the white matter tract integrity between the rACC and the PAG is correlated with one’s capacity for placebo analgesia.

And this is what’s really got me thinking:

According to findings from neuroimaging studies, we create a nocebo response (the opposite of placebo, i.e. the expectation of increased pain leading to increased pain) faster than we do a placebo response.  I wonder why this is.

One of the speakers suggested that perhaps hidden treatments, i.e. harnessing the power of placebo analgesia, are more transferrable to the clinic than we realise.  Think of all those patients who leave the clinic setting unaware of, for example, what medications they’ve been put on or what spinal manipulation they’ve undergone.  While contentious I can understand where the speaker was coming from, and I agree with her in that we need to somehow use the power of placebo analgesia to our advantage in the clinic.

So it was controversial (there was nearly a fight), but I bet I’m not alone in saying I don’t mind a bit of controversy if those involved have informed opinions.  And that’s what this was: a well delivered presentation of the latest research, along with honest, brave and very well-informed opinions.

Flavia Di Pietro

Flavia Di Pietro Body in MindFlavia Di Pietro is a PhD student in the Body and Mind Research Group, Sydney. She is investigating the development of Complex Regional Pain Syndrome (CRPS) after wrist fracture. Specifically, Flavia’s PhD involves brain scanning people who are in a higher than usual amount of pain in the first 3 weeks after the fracture, and then following them for a few months. Her question concerns whether or not there are changes in brain activation patterns that emerge before the CRPS does and if so, what do they tell us about the condition? Here is Flavia talking about what she does and a link to her published researchBiM author’s downloadable PDFs can be found here.


  1. Very interesting discussion. The concept of the neuromatrix is undergoing revision so I won’t reference it. In terms of the myofascial group, Jay Shah is a very entertaining and engaging speaker. He references that he had 6 years of chronic low back pain ‘cured’ by one paraspinal injection (I don’t know if this was to a trigger point (active or latent) or what). It does highlight that altering the afferent barrage can have effects on some ‘susceptible’ patients – especially if it allows them to return to meaningful function (probably the best modulator) – what happens to the nucleus accumbens – pFC chronic connection if the person does receive a much searched for reward ? Flavia, I was wondering if all the patients in your fracture research group are receiving high doses of vitamin C (definitely a better than placebo response in preventing CRPS post fracture) or at what point giving them vitamin C is no longer helpful or if just telling them that vitamin C is helpful, is helpful. Just curious.

  2. Flavia Di Pietro says

    Hi Vince,
    Thanks for your interest. What these three did was present a collection of research in this area. There is evidence that certain people are more ‘prone’ to the placebo effect than others. What I found most refreshing about this workshop was that it did not talk about placebo as negative, or as something we necessarily always need to eradicate in the research or clinical world. Clearly we always want to know the effects of a treatment we provide. But in many instances, as we were reminded, placebo analgesia is difficult (and perhaps not always crucial) to eradicate. You bring up an interesting point, if I am correct here in my interpretation of the second part of your post – that someone’s beliefs and attitudes can indeed influence the amount of relief they will receive from some treatments. Thanks again for your input!

  3. Flavia Di Pietro says

    Hi Tim
    Thanks for your comments. You always contribute interesting stuff!
    Regarding the “hidden treatments”: in the same way patients’ expectations are deliberately altered in some trials (e.g. “this will really help”; “you will feel better with this pill or manipulation”), could we, hypothetically, say something similar in the clinical setting in order to perhaps enhance the effects of the treatments we are providing? It may seem unethical until we’re reminded, as we were in this workshop, that so many people seeing health care professionals are largely unaware of the treatments they are receiving anyway. Very interested in the Glimpse Effect. Thanks again for your input Tim.

    Tim Cocks Reply:

    Hi Flavia

    Thanks for your response.

    There’s a brilliant Brian Science Podcast (No. 77) with Dr Fabrizzio Benedetti, the author of ” Placebo Effects: Understanding the mechanisms in health and disease.” and “The Patient’s Brain: The neuroscience behind the doctor-patient relationship” (both on my ever growing list of books to read). Similarly, Dr Benedetti talks about studying placebo effects (and he makes a point that it is plural) and his interest in not so much eliminating them from research but trying to understand the multiple underlying mechanisms. He talks about the placebo effects recorded at the level of a single neuron in patients with Parkinson’s Disease given saline injections and verbal suggestion of motor improvement and the idea of unconscious placebo effects and the immune system – just fascinating stuff.

    Clinically, it reminds me that everything we say and do can have placebo/nocebo effects, especially if we’re perpetuating Rene’s horrible mistake.



    Julian Campbell Reply:

    Fabrizzio Benedetti has also published this review which hopefully explains the background to the work presented in Milano.
    I have had to concede that most of my medical interventions are purely “placebo” according to his definition.
    The challenge is to do better than placebo, otherwise I would stick to selling snake oil.

  4. Usual mixed bag of responses. The term psychosocial has long since stopped meaning anything to me. Psychology is physiology. Society impacts on an individual and equally changes physiology. Tissue injury changes physiology. They’re all the same thing!! Anyway…Dr Mick Thacker has been talking of research done by a group that shows the specific effect of any drug is improved if the Dr explains what the drug whilst writing the script. He extrapolated this to the manual therapies over a decade ago, recommending therapists explain what there treatment does while doing it. Understand a thing and it’s no longer a non-specific effect, right?

  5. Teresa Merrick says

    Very interesting! Also, google material by Sarno, especially “Mindbody Prescription” and “The Divided Mind”.

  6. I think this is a problem where definitions are very important. We can have pain that has existed beyond the healing period,but are all chronic pains equal? I think that there is pain which persists due to a problem which is present because the irritant is still there and once you remove the irritant then the pain can be reversed but there are pains which exist even after all therapies known have been exhausted and all that can be done is to affect the psychosocial aspects and have the coping mechanism improved. Some pain seems to be unable to be reversed. Is this because we do not know the real irritant or is it a different type of mechanism creating this chronic pain we just don’t know how to reverse.

  7. I’m a kiwi living in boston and treating a lot of people in pain. I love reading the bim banter, but I’m a little perplexed. Flavia et al talk about the ACC and PGM, but we’ve known about this stuff for ages. It’s long been published in the wonderful books ‘Muscle Pain’ and associated research papers, by Mense and Gerwin et al.

    Muscles cause pain. Chronic pain. Myofascial pain. Trigger point pain. Lots of pain. The good news is that it’s mostly reversible. Google Kraus-Weber, or Norman Marcus, or Travell & Simons, or Jay Shah, or Sidhartha Sikdar, or Dommerholt and Gerwin to uncover the background for my boldness.

    Pain, retreat,

    Tim Cocks Reply:

    Hi Stew

    I beg to differ.

    Muscle’s do not “cause pain” – never have, never will.

    Phrases like “myofascial pain” and “trigger point pain” commit the great Cartesian error ( !) and in the context of a modern, neuroscience based understanding of the biology of pain are indefensible.

    That anyone can accurately, reliably and consistently palpate “trigger points” is extremely questionable (see the systematic review by Lucas et al, Clin J Pain, 2009;25:80-89).

    “Evidence” that purports to support trigger point treatment is full of poor methodology – such as this one – “The immediate effect of soleus trigger point pressure release on restricted ankle joint dorsiflexion: A pilot randomised controlled trial” by Grieve et al 2011 in the Journal of Bodywork and Movement Therapies.

    I am not a clinical researcher, but I have huge issues with a study that concludes that trigger point therapy is effective when the control group received no sham intervention or even have a therapist touch them at all. Would have been interesting if the control group had received at least some form of touch to anywhere on their lower leg, away from the purported “latent trigger points” and then had their ROM re-tested.

    The real kicker is in this sentence from the authors themselves “These findings are clinically relevant, although the treatment effect on ankle ROM is smaller than a clinical significant
    ROM (5′)” which seems to contradict an earlier statement that “This study identified an immediate significant improvement in ankle ROM after a single intervention of TrP pressure release on latent soleus MTrPS”

    Which is it, significant improvement or not clinically significant??

    Further, how do you know that a muscle is the “cause” of pain and not responding nocifensively to the vast array of input on the left hand side of the neuromatrix (from which the term “trigger point” has been removed in the updated model)?

    Conflating correlation with causation (your discs/muscles/joints are causing pain) has led to much harm, suffering, unnecessary surgery and entire professions built on logical fallacies.

    Isn’t it time we moved on from “issues in the tissues” thinking? Isn’t it time to embrace neuroimmune, neuroendocrine, whole person models based on our deeper understanding of pain biology?

    Google Moseley, Butler, Thacker, Wall, Melzack to uncover the background for my boldness.


    Anonymous Reply:

    Straight and to the point. Good for you. I totally agree

  8. “According to findings from neuroimaging studies, we create a nocebo response (the opposite of placebo, i.e. the expectation of increased pain leading to increased pain) faster than we do a placebo response. I wonder why this is.”

    Could this be the brain/person “erring on the side of caution”? Makes a kind of sense that it’s safer (biologically more advantageous perhaps) to be “better”/quicker to increase pain in the anticipation of a potentially or actual painful or dangerous situation as a protective response.

    From another (protective) perspective, it makes sense for the brain/person to “jump to the conclusion” and output pain – if it gets things wrong erring/jumping towards a nocebo response, the worst case scenario is pain – if it gets it wrong the other way – ie placebo analgesia the outcome could be harm/damage or worse…

    Reminds me of a story I heard a bloke tell once, about a sarong, a snake bite and 4 day coma – reckon he could have used a bit of nocebo (but then he wouldn’t be able to tell that great story…)

    A potential drawback with this might occur when the individual’s thoughts and beliefs lead to them anticipating that EVERYTHING is going to be painful.

    Perhaps similarities here to Daniel Eagleman’s idea of the “Glimpse Effect” He reckons (according to research he has carried out in his lab) that when people very quickly glimpse a face, they are more likely to consider it as attractive then when compared to having as much time as they like to rate the attractiveness of the same face.

    He suggests that this is biologically advantageous in that by erring towards more attractive, one is more likely to chase/follow a possible mate with attractive genes – a quick double take, at very little cost, might even be enough.

    On the other hand, if you mistake an attractive possible mate as being unattractive then its “sayonara to a potentially rosy genetic future”.

    Could the same cost/benefit reasoning be at play with placebo/nocebo?

    Flavia, could you expand on the idea of “hidden treatments” in the context you used it. Is the suggestion that placebo analgesia is enhanced when the recipient is unaware of ‘treatment’ occurring?

    It seems to me that the word/idea of a placebo can be a bit loaded- have a bunch of baggage hanging off it; I’ve seen it used often to denigrate a treatment or modality, ie “treatment X does not work, it is *only* a placebo” (perhaps with a bit of a sneer). However, I think it has to be one of the most interesting areas of study, not just in regards to pain but in a whole range of health fields.

    Another great post Flavia – thanks- nothing like a bit of neuroscientific fisticuffs!

  9. thank you for this. i am just preparing my lecture on this specific topic and I am very interested in your thoughts. i like to provoke a bit of controversy in class

  10. Nice precis Flavia. Did their research relate some assessment or measure of personality ‘type’ or ‘grouping’ to better understand whether the effect achieved was placebo or nocebo. For instance is the ability of the patient to grasp exactly what medicines they have been given a pre-condition to how effective any attempt to manage pain will be received and interpreted?