What did you expect?

Hands-up who thinks a patient’s expectations influence how well they do in treatment?

By Steve Kamper

Nearly everyone? That’s no surprise. Research recently published by a group in the US reported on the relationship between expectation and outcome in a sample of back pain patients receiving physiotherapy. This is by no means the first time this link has been reported; numerous previous studies have reported similar findings across many different types of conditions. This is also this same relationship that provides the most widely-accepted explanation for placebo effects (sometimes called non-specific treatment effects). Outside the research setting, clinicians recognise the association between expectation and outcome and try to pass on a feeling of confidence and positivity, even if, at times they know the treatment itself is ineffective [1].

It all sounds pretty rosy, just pump up the expectation volume and you get extra bang for your treatment buck. But what if the expectation/non-specific effect is all you are getting? Even some medications that we ‘know’ are effective analgesics can have negligible effects when administered without the patient’s knowledge [2]. Given that placebo interventions attempt to control for patient expectation; maybe this is why RCTs report such small effects for so many commonly used treatments [3]. However, if we buy into this argument, are we reducing our clinical practice to little more than 21st century witch-doctory? For someone like myself with a background in physiotherapy, the thought is a little disconcerting. And I’m not alone, whole therapeutic professions have had to defend themselves against this accusation [4].

On the other hand, does it matter at all? If the people who come through the door (mostly) end up getting better and they are satisfied with the service, is their time/money not well spent? [5] As long as the patient knows what they’re getting and how much it’s going to cost and the clinician seeks to attain to the best outcome, everything’s hunky-dory. Maybe…

Regardless of where you stand on the specific effects of any particular treatment though, it is probable that disregarding the effects of patient expectation will result in suboptimal outcomes for patients. Just as apparent though is that, for the most part we’ve bugger-all idea about how to use this information. Much has been locked up in the black box marked ‘Placebo’ which, it could be argued has been/continues to be a hindrance to understanding and harnessing non-specific effects (see [6] for an excellent discussion on this topic).

There is no shortage of questions that are awaiting our attention. Are the patients with high expectations of doing well the ones that just aren’t very sick? Are ‘expectations’ just measuring how optimistic someone is generally? How do people come by their expectations? Can clinicians change them, and if so, how? How do expectations influence outcome, is it a direct effect or indirect via something else, like improving treatment adherence?

The idea that we can piggy-back enhanced expectation effects onto whatever treatment we choose is a seductive one, it offers the promise of additional benefit for lots of patients with all different types of conditions. If we can just find the right key (or sledge-hammer) to open the placebo black box we just might find that it is full of tasty goodies.

About Steve Kamper

Steve’s career as an Environmental Scientist was cut short due to an inability to grow dreadlocks or a convincing beard; he changed to Physiotherapy after being told he looked handsome in a polo shirt. He is currently enjoying the fancy restaurants and 4-day weekends that accompany the life of a PhD student at the George Institute for International Health in Sydney. Steve’s research to date has involved investigation into subjective outcome measures and placebo effects, particularly in patients with whiplash and low back pain. Leisure time is spent playing soccer, running and doing push-ups in his Speedos at the beach.


[1] Tilburt JC, Emanuel EJ, Kaptchuk TJ, Curlin FA, & Miller FG (2008). Prescribing “placebo treatments”: results of national survey of US internists and rheumatologists. BMJ (Clinical research ed.), 337 PMID: 18948346

[2] Amanzio M, Pollo A, Maggi G, & Benedetti F (2001). Response variability to analgesics: a role for non-specific activation of endogenous opioids. Pain, 90 (3), 205-15 PMID: 11207392

[3] Machado LA, Kamper SJ, Herbert RD, Maher CG, & McAuley JH (2009). Analgesic effects of treatments for non-specific low back pain: a meta-analysis of placebo-controlled randomized trials. Rheumatology (Oxford, England), 48 (5), 520-7 PMID: 19109315

[4] Singh S, Ernzt E (2008). Trick or treatment. Alternative medicine on trial. Bantam Press, UK.

[5] Hush JM, Cameron K, Mackey M (2010). Patient satisfaction with musculoskeletal physiotherapy: A systematic review. Under review at Physical Therapy.

[6] Nunn, R. (2009). It’s time to put the placebo out of our misery BMJ, 338 (apr20 2) DOI: 10.1136/bmj.b1568

All blog posts should be attributed to their author, not to BodyInMind. That is, BodyInMind wants authors to say what they really think, not what they think BodyInMind thinks they should think. Think about that!


  1. Nicksterdam says

    Steve, you raise some very interesting questions, which made me think of one (only one, it is Monday morning) of my own… Which treatments do you think these expectations influence?

    I can imagine that positive expectations about prognosis or treatment strategies can be helpful for many different patients and conditions, but what if they have positive expectations about harmful treatments? I’ve been conversing with a few nutritionists lately, who seem to battle daily with patients who seem to have misguided expectations about which weight loss strategies work (for instance). What I mean is, even if you have positive expectations that eating only cheeseburgers will help you shed the kilos, it probably still wouldn’t work, right? Do these expectations mostly pertain, then, to treatments where there is a minimal “active”/”biological” component (e.g. physiotherapy)?

    Any thoughts on this?

    Another great post (by Steve, not me), I’m loving the variety of contributors and articles provided by this blog. Keep up the good work!

    Steve Kamper Reply:

    It’s an excellent question Nickster, in short; I’m not really sure, but in the spirit of the Federal election campaign underway here in Australia I will bang on for a bit on somewhat related issues and hope it passes for an answer. I don’t think that expectation is the only factor at play but probably acts as an adjunct to whatever the biological (in the sense of the ‘bio’ part of the biopsychosocial model) processes are going on (in the treatment and in the condition). So to your example; maybe the cheese-burger believer will lose a little more weight on the cheese-burger diet than a non-believer, but in all likelihood still end up pretty tubby. Related to this, perhaps the potential for effect of expectation with respect to any condition (or treatment) is in some way proportional to the size of the biological part of a condition. To the example again; maybe cheese-burgers along with cheese-burger belief will be less successful for obesity, than for fibromyalgia. I’d recognise here that I have no idea of how, or in fact even if it is possible, to divide a condition up in this way. Another complicating issue here is choice of outcome – it seems plausible to me that some outcomes would be more sensitive to expectation effects than others. To the example a final time then; maybe cheese-burgers will improve your quality of life but not be so good for shifting the spare tyre.

    I’m hungry.

  2. Good piece. I think anyone considering the patient’s experience will think about and ask about their expectations. We all hold beliefs based on our past experiences, understanding of our place in the world and a host of other influences.These beliefs will surely impact upon any expectation that we may have about treatment and life in general.
    As I sit here watching the cricket highlights, I expect that the Aussies will put up a bit of a fight. If they don’t I’ll be disappointed. This expectation comes from 30 odd years of watching Australian cricket.
    Expectation is inherent and part of processing thoughts unless you are in a state of mindfulness. This is a wonderful state of non-judgement, holding oneself in the present moment. There is some great writing on this including Thich Nhat Hahn (http://www.amazon.co.uk/Peace-Every-Step-Mindfulness-Everyday/dp/0712674063/ref=sr_1_1?ie=UTF8&s=books&qid=1279918680&sr=8-1) and Eckart Tolle. From a clinical standpoint I do not find it surprising that mindfulness is gaining acceptance, especially in chronic pain treatment programmes. Attention on the painful area and mindful ‘non-judgement’ means that the signals are being acknowledged but not valued, hence threat reduction. I have no empirical evidence only case studies!
    Asking about patient expectations is routine. Reducing the threat is routine. How do we individualise it? Perhaps recognise belief and see expectation as an ally that can help us determine the best course of action.
    Mmm, could get waffly now so signing off… I now expect Pakistan to win.

  3. Darcia Dexter says

    Great post. As a Feldenkrais Practitioner going into my 15th year, I’ve seen many a chronic pain client where the expectation is exactly correlative to the outcome and willingness to participate in their own wellness. You had me at your bio…one of the best I’ve seen. Thanks! DD

    Steve Kamper Reply:

    Thanks for your interest Darcia, glad you liked the piece. I have a question about the patients in whom you see such a strong relationship between expectation and outcome. Some of the research that has been done in the area makes the distinction between expectation related to outcome, and expectation that is specific to a certain treatment, a sort of interaction effect. So the question becomes; does it matter what treatment people with postive expectations get, if their expectation is driving their outcome? In your opinion do the patients you mention expect to get better before they see you (almost regardless of what you are going to do), or is the expectation generated by your explanation of the treatment? The obvious answer is that it is a bit of both, but I wonder if you think one or other is more important.