When is a Placebo Acceptable in the Clinic?

We recently published a paper with the mouthful title “Placebo Use in Pain Management:  A Mechanism-Based Educational Intervention Enhances Placebo Treatment Acceptability”[4].  Before I get to the the findings from that paper, I’d like to provide some context of where this paper fits in the larger placebo analgesia literature.

Placebo is not nothing.  I’m convinced of that.  Many academics are beginning to understand that.  There is plenty of evidence for that conclusion[1,5,7,9,10]. Placebo analgesia is a powerful, physiologically active process.  However, the lay public, and most patients consider placebo to be inert[2].  Related to that belief, most of the public and patients with chronic pain whom we have surveyed consider the use of a placebo deceptive, and view deception as highly undesirable[3].  However, the unacceptability of a placebo is lessened if the placebo is found to be effective[2]!  This suggests that placebos might be more appropriately employed in the clinic if we bring our patients to an understanding similar to our scientific understanding.

So how do we make placebo analgesia more acceptable to patients?  First, placebo is highly used, even in its deceptive formulations [4,6,8].  Furthermore, I strongly suggest to you that placebo or its active ingredients of expectation and conditioning are part of every single treatment that we use with patients.  The evidence overwhelmingly supports that claim.

Forget the term placebo.  How do we use expectation and conditioning to treat people with pain?  How do we do that ethically, without deception, and increase acceptability to patients?

The Kisaalita et al. paper employed an online education about placebo to do just that[4].  In this paper we randomized patients with chronic pain to two conditions.  In one condition they received an online “education” about the empirical evidence for placebo analgesia.  We told this group what we know about brain activation, opioid mechanisms, expectation, and conditioning.  The other group received control education about chronic pain, but no information about placebo.  We assessed placebo acceptability before and after these educational interventions.   As predicted, the patients with the placebo mechanism-based education found placebos to be significantly more acceptable than at pre-intervention, and they found them more acceptable than the control group.  In general, we find that the most acceptable placebo is one that works, is non-deceptive, and has few negative consequences[3].

So what? Should we give everyone a sugar pill and tell them it’s a miracle cure?  Should we sell snake oil?  No, emphatically NO.  We are not suggesting anything of the sort.

These results provide us with an ethical foundation to inform our patients about the full range of active ingredients in our treatments.  We can now truthfully inform them that those treatments we already use (I’m presuming you use empirically validated treatments) work in more than one way.  We can let them know that through positive expectations (yours and theirs), and through classical conditioning, we can even enhance the effectiveness of a treatment.  And we can do it without deception.

What should you do as a treatment provider?  First, pick treatments that have been empirically validated.  Then be enthusiastic about them and the expectation for success.  Engage the patient in the same enthusiasm, and tell them that you are doing that to increase the efficacy of the treatment….because that is also an empirically validated treatment!  Double the value for the money!  We are still working on how to employ classical conditioning into treatment, and have some as yet unpublished data on how to do that.

About Michael Robinson

Dr Robinson is a Professor of Clinical and Health Psychology, Anesthesiology, and Physical Therapy at the University of Florida. He is the Director of the Center for Pain Research and Behavioral Health at the University of Florida.


[1] Craggs J, Price D, Robinson M: Enhancing the placebo response: functional magnetic resonance imaging evidence of memory and semantic processing in placebo analgesia. J Pain , 2014.

[2] Kisaalita N, Robinson M: Analgesic placebo treatment perceptions: acceptability, efficacy, and knowledge. J Pain , 2012.

[3] Kisaalita N, Roditi D, Robinson M: Factors affecting placebo acceptability: deception, outcome, and disease severity. J Pain , 2011.

[4] Kisaalita NR, Hurley RW, Staud R, Robinson ME: Placebo Use in Pain Management: A Mechanism-Based Educational Intervention Enhances Placebo Treatment Acceptability. J Pain 17:257–69, 2016.

[5] Letzen J, Craggs J, Price D, Robinson M: Effective connectivity predicts future placebo analgesic response: A dynamic causal modeling study of pain processing in healthy controls. Neuroimage , 2015.

[6] Price D, Fillingim R, Robinson M: Placebo analgesia: friend or foe? Curr Rheumatol Rep , 2006.

[7] Price DD, Craggs JG, Zhou Q, Verne GN, Perlstein WM, Robinson ME: Widespread hyperalgesia in irritable bowel syndrome is dynamically maintained by tonic visceral impulse input and placebo/nocebo factors: evidence from human psychophysics, animal models, and neuroimaging. Neuroimage 47:995–1001, 2009.

[8] Robinson M, Price D: Placebo analgesia: Widening the scope of measured influences. Pain , 2009.

[9] Sevel LS, Craggs JG, Price DD, Staud R, Robinson ME: Placebo analgesia enhances descending pain-related effective connectivity: a dynamic causal modeling study of endogenous pain modulation. J Pain 16:760–8, 2015.

[10] Vase L, Robinson M, Verne G, Price D: Increased placebo analgesia over time in irritable bowel syndrome (IBS) patients is associated with desire and expectation but not endogenous opioid mechanisms. Pain , 2005.

Commissioning Editor: Associate Professor Claudia Campbell; Associate Editor: Dr Tory Madden


  1. A quick reminder to folks who are commenting on BiM. As a courtesy to our authors and readers please keep your comments ON TOPIC and BRIEF.

  2. Gerry Daly says

    ” At present, we are grabbing at smoke because we lack the definition consistency. ”

    Someone like me, obviously a skeptic, would probably want to rephrase that as….’ At present, we allow it to be used as a smokescreen because our complicity in allowing ‘placebo’ to have an enigmatic lack of definition, might seem to have an advantageous purpose’. Meaning…….due to definition confusion, it is a useful unchallengeable last resort treatment, where other more evidenced science-based approaches might be failing. In a way, placebo relaxes the pressure on the science to come up with answers, and everyone get’s to go home thinking…’Oh well, at least that might work ‘. That’s not a criticism, I’d expect that from anyone faced with a medical puzzle that confounds analysis. But, placebo can blur the edges of good treatment research and trialing, and even good assessment of outcomes.

    “Good outcome false reasoning”….very true of many great medical innovations, which were initially conceptualised through ‘unblinkered’ observation of every day events. Only after the initial ‘eureka’ moment was the science evidence added in support. Got to give credit there for pure intuitive observation as the catalyst, and the science thinking being relegated to a back seat to allow more licence for visionary adventurism. Pasteur noticing that some German milkmaids seemed to have immunity to bovine diseases, leading to his ideas about vaccination, springs to mind there. Perhaps the ‘science’ frowns upon such approaches nowadays.

    ” and possibly even harder to have patients accept the idea. ”
    That’s the bit for me that calls out for a return to basic understandings, and the need for clear evidence-based definition which a patient might understand. Placebo effect has obviously gained some recognition in the professional circles, but I’m not sure that has translated itself back on to the public in a meaningful way. It is maybe still regarded as an ‘interventionist’ tool, rather than as a ‘patient’ tool.

  3. Interesting discussion, indeed. I like the sentence, Forget Placebo.
    Ultimately we may be dealing with a patient’s belief system interfacing with a therapist’s belief system and both those belief systems being experientially and educatively biased.
    Then the physiology kicks in. The reason patients, therapists, and researchers may be having a problem defining a consistent explanation for the placebo effect are the multivariant influences. A one size fits all explanation is unlikely. Also, the discussion is plagued by the use of the term ‘placebo’ itself, because the definition is changing and not everyone uses the term ‘on the same page’. A discussion becomes fraught.
    Addressing experiential learning by new experience may show promise, as doing so may influence over protective or maladjustive belief systems. A tailored approach for each individual patient and therapist.
    But are we, as clinicians, well enough informed about our own, and patient bias, and even if so do we possess sufficient broad cross professional skills to adequately understand and then to provide relevant repeatable interventions? Are we there yet? I do not think so. However, the discussion is very necessary.

    Gerry Daly Reply:

    ” Also, the discussion is plagued by the use of the term ‘placebo’ itself, because the definition is changing and not everyone uses the term ‘on the same page’. ”

    Very true. To some it is a psychological procedure, to some it is a neutral control in trialling, to some it is a means of placating doubts, and to some, on a purely subjective level, it is means for evaluating natural healing. For me, it’s the possible effect it might have which should determine its meaning….and that has to be the question of how the options of intervention and ‘disguised’ non-intervention play out in real healing terms. Without a real healing outcome, the whole ‘placebo’ question becomes invalidated…..in other words, if it only works sometimes, we’re probably looking in the wrong place for what instigates it. So, it seems we tend to view it as an ‘intervention’, rather than as an ‘awakening’ of a patient’s confidence in their own self-efficacy. As I said earlier, it hints at credit being claimed for processes that are really beyond the interventionist powers or abilities of the operator. The only question for any operator is….if placebo is proven to be a viable alternative to other treatments, what might be the best method for ‘unlocking’ it, or better still, for allowing the patient the best means for unlocking it. Personally, I tend to think that outcomes would probably not be much different whether we call it ‘placebo’ or ‘reliance on natural healing’, so, claiming it as an advised procedure seems a bit like ‘bending’ reality.

    Mark Quittner Reply:

    Again, interesting concepts. We all struggle for accurate definitions to assist both us and our patients to place discussion upon common ground. For the assessment purpose, we need a consistent framework by which we can compare outcomes. At present, we are grabbing at smoke because we lack the definition consistency. Science, less so than snake oil, has been caught out by noting change after interventions then falsely tagging the change to the intervention. (Example: Miasma – the horrible smell in the Thames River outside the British Houses of Parliament thought to be causing disease and the subsequent separation of sewerage from the water supply reduced both smell and disease. Good outcome false reasoning.)

    I believe we can eventually tag the reliable recruitment of ‘placebo’ effect to treatment once an understanding of the mechanism/s has occurred. Then we can teach each other how to assist our patients to utilise the technique. Not as straightforward as understanding the benefits of exercise upon our physiology, and possibly even harder to have patients accept the idea.

  4. Gerry Daly says

    Well, I think you’re on to something there, Alison. Especially with chronic conditions with no medical healing expectations, who knows the intimate dynamics best ? It’s a source that is seldom explored because of the reverence for the science. Home remedies and self-efficacy have been blind-sided by the textbook approach. The great innovations in science have also had an opposite effect on ‘listening’ to the problems.

  5. Thanks Gerry, I agree , science based intervention has been the catalyst so far. It does not allow for much patient involvement and choice. Have we put biases on these therapeutic treatments and on their capabilities?
    Have we become the healer instead of the facilitator? What would happen if we guided and the client led, helping them to move into their threat response with safety without the need for drastic intervention. They may just see how amazing the body is at correcting itself.

  6. Gerry Daly says

    “Are you saying that if you were in deep sleep, and someone applied a flame to your arm, you’d not wake? Come on now.”

    No. What I’m saying is that the pain experienced before falling asleep becomes non-existent during deep sleep. Any new threat arising whilst asleep will cause a wakening. For instance, one could go to sleep with a broken arm, and not feel pain whilst asleep….but a mere pin-prick to a finger, whilst in that state, would register and cause an awakening. It only seems strange because different rules apply to an existing pain experience, than to a new emergent threat. That’s a phenomenon which could easily be tested for validity.

  7. Clinically finding the low effective treatment to add your trust inducing pitch to is still riven with your own inherent biases.

    I work from a position of knowing I am biased and trying to mitigate that bias. Sheesh. Cover that in do no harm. Very tricky. To create that link of confidence without fostering passive coping. To educate without adding unhelpful beliefs that work their way into unhelpful behaviours.

    Best pragmatic evidence informed opinion piece on placebo I have ever read.
    Thank you.

  8. Gerry Daly says


    In a sense, it might be said that the science has inadvertently relegated the autonomic responses to a lesser importance, which in turn impacts upon the public’s confidence in those essential health-ensuring self-healing systems. That confidence can be restored with intelligent and meaningful informing.

  9. Gerry Daly says

    “only we have put limits on the capacity of the body to heal and unwittingly passed this information on.”

    That’s a good and revealing way to put it. Over the last two centuries, the public have been bombarded with the wonderful medical advances achieved, mostly at the expense of a relative reporting of failing treatments for chronic issues. As such, it has become the default mindset that reliance on the science is something of a last resort for most people. However, when a chronic condition kicks in, and the science hasn’t yet caught up, that ‘reliance’ becomes meaningless…in fact it becomes an obstruction to any determination to be as self-reliant as possible . The mindset vacuum created by having to re-think the reliance on the science, usually only encourages a dependency on short term solutions, whether medicinal or surgical….simply by virtue of patients not wanting to lose faith in the science. So, instead of increasing self-efficacy prospects, what tends to happen is the patient gets caught up in nurturing a growing mistrust which is only placated by entering further cycles of medications and hope for possible surgical solutions. The science, when it doesn’t live up to desired expectations, can leave patients with chronic conditions somewhat stranded in no-man’s land…and that creates its own lateral demands which also require attention.
    People tend to see the treatments, whether medicinal, surgical, or therapy, as the catalyst for any healing to occur. The reality is that most meds and surgery actually work by increasing the threat and thus initiating a more robust autonomic response. It’s the autonomic response which does the healing. That’s not to undermine, in any sense, the corrective intentions of many procedures, but considering most procedures would be ‘too risky’ without a guaranteed autonomic response, I think the attention should be shifted to those responses, and the public informed accordingly. As it is, the focus of the public tends not to go beyond the ‘intervention’, and subsequently there has been a general depletion of self-reliance on the corrective abilities of the body itself. Having a dependency on the science, which doesn’t always deliver, is a very vulnerable place for any chronic patient to have to negotiate.

  10. Yes I agree Gerry, it only appears a placebo to the therapist/doctor/patient because through applying medical knowledge/science only we have put limits on the capacity of the body to heal and unwittingly passed this information on.

  11. Gerry Daly says

    ” If you could remain conscious during deep sleep, you’d experience pain without suffering. ”

    I would probably rephrase that as….”If you could remain conscious during deep sleep, you’d realise that pain doesn’t exist in that state, and that there is no requirement for pain to express itself as part of the ongoing healing processes in that state. ” Some differences there, but I can’t see the point to the existence of any pain without discomfort or suffering……that’s what it does when we’re conscious. It might be difficult for some to grasp that pain is a conscious only event, and also, that it’s attributes are mostly consciously detrimental, in the sense that it achieves nothing except some discomforting or distressing awareness of a threat. In all likelihood, pain only seems to have one definable attribute….to restrain normal conscious reactions to an ongoing injury/threat. As such, there would seem to be no requirement for pain when consciousness is ‘switched off’, and that’s what seems to happen.

    Drifting off the subject matter there….back to placebos. There will be patients who are susceptible to suggestion, in whatever form. But, I don’t see that happening in the larger demographic……the average mindset would be too resistent, too skeptical, to allow an assumption of second or third party control to influence their intuitive decision making. We protect these instincts as though our life depended on it. Any hypnotherapist would attest that the patient needs to be consciously ‘willing’, and that, by virtue of the exclusion of those who don’t display willingness, becomes questionable, ethically. If it doesn’t have the possibility of working for everyone, it simply doesn’t fulfill the criteria of a ‘general practice’. Otherwise, it’s fine that it is explored in interested groups.

    The placebo effect must be self-generated, that’s the only way I can see it might unlock any healing potential. For me, it shouldn’t require any conscious effort, because that would make it vulnerable to typical conscious skepticism, undermining its purpose. And that suggests to me that the best way to encourage a possible placebo effect might be to educate patients about the inherent powers and abilities of the autonomic systems. Let’s take the personalities out of the equation, and just focus on what does the actual healing. People, in general, need their confidence in autonomic healing restored….what used to be regarded historically as ‘divine intervention’ has been replaced with medical theory, all at the expense of a lack of awareness about any individuals own healing potential.

    EG Reply:

    1) “As such, there would seem to be no requirement for pain when consciousness is ‘switched off’, and that’s what seems to happen”.

    Are you saying that if you were in deep sleep, and someone applied a flame to your arm, you’d not wake? Come on now. As for dream sleep, pain is certainly possible in REM. The pain in NREM appears to be expressed as basic withdrawal movements.

    2) “Let’s take the personalities out of the equation, and just focus on what does the actual healing”.

    But it’s the person that creates the change, not the process. The really powerful healers through history were all relatively free of self-referencing. I understand you want to keep religion out of the picture, but when you strip religions bare, you find at the core are teachings about how to retreat from the self. If you want, you can ‘scientize’ it and call it a process of switching off the default mode network. We scientists have to realize we are late on the scene when it comes to the question of human suffering.

    If healing was just about knowledge, we could publish a book or website and distribute it worldwide. It would work to some degree (no doubt), but the difficult cases need personal interaction. Self-healing is possible, but much harder to complete.

    EG Reply:

    Self-Correction… it *is* about the process, it’s just that only certain people have mastered the process (so it does become personal). And I’m not referring to myself because master healers don’t post on blogs! But meditation seems essential. Healing emerges from certain mind states which can be transferred. No doubt about that at all, because I experience that side of things and have performed my own experiments.

    Gerry Daly Reply:

    When I say …’Lets take the personalities out of the equation’, what I mean is ‘Let us remind the operators of their exact role in the equation between their ability to heal, and the autonomic ability to heal’. For me, the operator’s role is totally dependent on the guaranteed autonomic response…..so, the hierarchy is obvious. Any machinations within that hierarchy should be kept well clear of client patients, for fear of exposure. The question facing any patient is ….’what do I trust most, an operator’s insistence that they can manipulate autonomic responses, or just the autonomic responses themselves ?’. I’m sure I wouldn’t be alone in choosing the latter, unless I knew for certain that the former could out-compete the latter.

  12. Gerry Daly says

    Just some extra thoughts……

    As someone who tends to see consciousness as a possible threat, or restriction, to autonomic healing functionality, I have to ask myself what exactly the placebo effect might be. It might seem to be a ‘contained’ natural healing method which, under the right circumstances, can be released to fulfill its probable function. I’m continually drawn to an understanding of some ‘conflict’ between consciousness and the autonomic systems, which restricts autonomic reactions until there is conscious acceptance, or until consciousness no longer has the ability to interfere in the autonomic processes…as when sleeping, for instance. Our mysterious placebo effect might just be exactly what happens when we sleep….the autonomic systems get a free run with no complications restricting their processes. So, it’s maybe not so much a question of ‘what is a placebo effect’, and more a question of ‘are autonomic processes restricted under some ‘consciously-controlled’ circumstances ?’.

    EG Reply:

    Gerry, I agree that the placebo is a truly stunning phenomenon. Imagine telling someone “this is morphine” and injecting them with saline and getting just as powerful pain reduction as the real thing. Even more stunning are those rare people who can say “your pain is now gone” and in a matter of seconds, finding exactly that – gone. Such things are possible.

    Keep in mind we have two issues – pain and suffering. Pain is the realm of the chemist. Suffering is the realm of the alchemist. When it comes to suffering, conciousness is not the problem, self is. Just because self arises only during consciousness, doesn’t mean that self *is* consiousness. You can be conscious without self (see video below). Learn the difference!

    Self is a bundle of memory-thoughts, all linked to the body and its desires/fears. When when self temporarily recedes (as in deep sleep or twilight anaesthesia), suffering cannot exist. If you could remain conscious during deep sleep, you’d experience pain without suffering.

  13. Gerry Daly says

    The entire ‘placebo’ question is perhaps being perceived from a ‘Who’s in control ?’ perspective, as most things are. As a possible means for healing, or just encouraging healing, it’s probably only natural for operator’s to wish to claim credit for something which might only be a patient’s willingness to attend to a medical problem with positive self-efficacy. Isn’t that a bit like taking credit for the activities of the autonomic systems. As the article implies, encouraging a patient towards a mindset which allows for the possibility of a placebo effect, is the ethical question because, if it doesn’t work, there arises a ‘responsibility’ issue with the operator’s methods. Placebo mindset encouraging, whilst maybe ineffective, is unlikely to have any negative effects on the presented condition, so, as such, it is probably ethically sound. Only if it is encouraged at the expense of some other treatment should the ethics be reconsidered.
    However, the means of encouragement do need to be questioned….whether that be suggestion, neuro-nudging, or an insistence that healing might be uinlikely without a pro-placebo mindset. These differing methods can affect a patient’s capacity to choose independently of an operator’s opinion. Neuro-nudging might seem the most ‘devious’ method, where a patient is encouraged to believe that they, themselves, have arrived at a pro-placebo mindset without any ‘suggestion’. At the same time, neuro-nudging might be the most effective method because it kickstarts a self-efficacy mindset. For any placebo to work, it seems there must be a conviction or a self-convincing belief that it is likely to work. Such convictions or self-convincing usually has better traction if it happens on a subliminal level, rather than as a consciously acquired belief. So maybe, some subtle neuro-nudging, without the patient being aware it’s being attempted, would offer up best prospects for the patient becoming convinced. If some good outcomes evolve from that….great, if not, no harm done.,

  14. Above, when I said “…then I don’t”, it should have been “then I don’t have to”.

    Whilst capable of answering lots of questions and allaying doubts, I don’t want to. I find it boring and it is usually detrimental to the outcome (usually, not always). I’m still congruent at this point. So I may experience a feeling of disinterest or irritation about being quizzed relentlessly. Full congruence requires outwards expression, but in this case it would create tension, so it needs to be withheld to some degree. So long as I am fully aware of my irritation, it’s all good. I tell the client I need to concentrate and I do need to concentrate quite intently. Still congruent. There’s a desire to avoid unecessary time wasting and achieve optimal outcome. I do this is a way which suits me.

    All this is a lenghty way of saying “If I can be as authentic as possible, it benefits both myself and the client maximally”.

  15. “First, pick treatments that have been empirically validated. Then be enthusiastic about them…”.

    It’s hard to be enthusiastic when you know that even the best physical treatments have very low power to create change which lasts. We don’t have that luxury unfortunately.

    So if a therapist decides to demonstrate a high level of enthusiasm for a treatment which has low-level scientific support, that will make him highly incongruent. Incongruence creates distrust and it’s game over very rapidly after that. Rogers has even described incongruent therapists as ‘dangerous’.

    The physical aspect of treatment is a necessary ingredient which has little power on its own. To remain congruent to this, I don’t talk much about the physical treatment, nor do I focus on it. The most I’d say is something like “this will help settle the nerve endings so that the muscles can relax. Then the pain goes away”. Very bland. I don’t elaborate because I don’t want to.

    For a long time I struggled with how to approach those clients who are very analytical and deep thinkers. They often seem to be engineers, not sure why. But you know they’ve been on Google for several hours before they even make an appintment and they have endless questions and doubts. I usually tell them I have to concentrate during the treatment and will answer their questions at the end… then I don’t. If I’ve remained congruent and present, the body-mind will recognize this at a deeper level and questions and doubts will just vanish. Then it’s just a matter of rebooking for a review. Simples.