Can Biases Impact How Effective You Think Your Therapist Is?
All you need to know about the placebo effect, natural history and regression to the mean.
Placebo, natural history and regression to the mean may sound irrelevant, but they are actually super important and have completely changed my understanding of healthcare. They are three separate, but often related, effects that your therapist doesn’t like to talk about. Three effects that your therapist isn’t really responsible for, but can make them look really good at their job. Let’s find out some more about them.
The placebo effect is a beneficial effect from treatment, which is not related to the active ingredient of the treatment itself. So, feeling better due to the expectation that the treatment is going to make you feel better, not due to the treatment itself.
A nice example of the placebo effect can be seen in a clever study that compared the effect of the price of medication on movement ability in participants with Parkinson’s disease. Both groups received the same medication, one group thought the cost of dosage was $150 whereas the other thought it was $1500. Surprisingly, the more expensive group had a significantly higher improvement in their movement, due to the expectation that more expensive medication would have a better outcome!
Another reason for looking at placebo in research is to see if the treatment is any better than changing beliefs and expectations alone, that is, is there an active effect at all? This is important as health care professionals want to be delivering treatments that are shown to be better than just placebo.
Physiotherapists (and other health care professionals) took a financial and ethical sting when placebo research started to investigate the widely used modality, therapeutic ultrasound. Therapeutic ultrasound is when a therapist places an ultrasound jelly on your skin over the sore spot, and rubs an ultrasound probe over that part of the skin. The idea is that the pulsing sound waves heat the tissue and help with inflammation and pain. But, does it perform better than the just thinking that it helps? The answer is probably not.
Clever researchers have looked at this in many domains, but this study I think highlights how influential the placebo effect can be. What they did was take a group of people and randomly place them in a genuine or sham therapeutic ultrasound group for the treatment of knee osteoarthritis. Importantly, neither the participants nor the therapists knew which group they were assigned to. Here’s the interesting part, both groups had a significant improvement in pain, stiffness, functional activity, walking time, disability, depression and anxiety, with NO difference between them. Meaning, the placebo effect exists and that therapeutic ultrasound (at least in this setting) is no better than changing beliefs and expectations around their knee and pain!
The sceptic Voltaire famously said “the art of medicine consists of amusing the patient while nature cures the disease” and I think that Voltaire was really on to something here.
Natural history refers to how an injury/illness naturally improves over time without intervention. Our bodies are wonderfully complex and clever and they know how to deal with injury or illness. In the vast majority of cases: bones remodel and strengthen, ligaments heal, cartilage grows, discs resorb, muscles stitch themselves together leaving functional scars and tendons re-align through doing nothing but getting on with life whilst respecting the body as it heals. In the case of pain, provided you’re generally active and healthy and the injury isn’t substantial, pain will pass without the need to see a health care professional for help at all.
But, why is this relevant for healthcare and why was Voltaire on to something? It’s relevant because health care professionals and their patients are guilty of associating the treatment they received with the outcome they experienced. Here’s how it goes: You go to a healthcare professional with a problem, they tell you that the problem can be fixed with method X and your problem improves and you associate your improvement to method X. The elephant in the room is that your problem may have just improved over time anyway, without help.
The graph here shows the results of a study comparing dry needling, sham dry needling and natural history for pain after a total knee replacement. As you can see there’s a very slight difference in pain at the 1 month mark with true dry needling helping a bit, but at the 3 and 6 month mark there is no difference; “amuse the patient while nature cures the disease”.
Let’s be clear here, I’m not saying that healthcare professionals can’t have an impact, I’m just saying that we don’t do most of the work (your body does) and that sometimes our odd treatments seem to be helping, but really they aren’t doing a lot more than reassuring you that it’s going to get better and that you’re doing the right thing for that to happen.
The slightly concerning thing about natural history is that a treatment may look like it’s helping, but in reality it could be slowing the natural healing process. To the healthcare professional and the patient all they can see is that they are getting better, perhaps slowly, but it is improving! The catch 22 is it might have improved faster if we didn’t get our meddling hands in there.
Regression to the mean
This one is very similar to natural history but tends to be related to people with longer-term issues. Basically it means this: when you have problems that fluctuate up and down in severity over time, you are more likely to visit a health care professional when things are bad. What we don’t realise, is that when things are bad they are generally on their way back to an average of being not so bad, or not noticeable at all. So you’ll probably notice an improvement in symptoms, despite any health care intervention.
The graph above shows the interaction between pain and time. Over time there is a normal ebb and flow of the intensity of pain and this ebbing and flowing is often the result of many complex, interacting factors. If you look at the gold star you can see that pain is at its worst, this is generally when people seek treatment for their pain. Just looking at how the pain naturally ebbs and flows over time you can see that it would have made its way back to the mean (or the average) anyway. Unfortunately, due to the same biases we have when it comes to placebo and natural history, we associate feeling better with the treatment we received, and not the normal ebb and flow of our pain. Also, if you look at the blue star you can see that they stopped treatment when the pain was at its best, naturally the pain started to get worse and in this situation the person may mistakenly associate the worsening of symptoms with stopping the treatment.
Another problem with regression to the mean is that you can have natural fluctuations of pain when doing things that are actually helpful for you. For example, you can start exercising when your pain is good, and have a natural swing to your pain being quite bad. You may then start to associate helpful behaviours, like exercise, with your pain.
How to avoid these biases
Unfortunately, I don’t know if it’s possible to completely avoid these! Even those who have rigorously studied cognitive biases like Daniel Kahneman, author of the popular book “Thinking Fast and Slow”, admits to having very little control over the outcome of his own biases, which is a little bit disheartening. That being said, by having an awareness of them I believe you can be a tad bit savvier in choosing your healthcare. When you receive a spinal manipulation, start drinking liquorice tea or do the “exercise to top all exercises” and notice a dramatic improvement in your symptoms you can ask yourself:
How much of an expectation did I have that it was going to help me?
Is this something that would naturally get better over time and if not, how did this actually help my body heal?
Is it possible that my symptoms were going to get better or worse anyway as I tend to ebb and flow, or did this intervention actually do something?
Without a good understanding of the human body and health it’s hard to know the answers to these questions. That’s why it’s good to find an up-to-date therapist that can make sure you get interventions that are really going to help you, minus the fluffy stuff.
When in doubt, sometimes it’s best just to keep calm, and keep moving!
Espay, A. J., Norris, M. M., Eliassen, J. C., Dwivedi, A., Smith, M. S., Banks, C., . . . Szaflarski, J. P. (2015). Placebo effect of medication cost in parkinson disease: A randomized double-blind study.Neurology, 84(8), 794-802. doi:10.1212/WNL.0000000000001282
Mayoral, O., Salvat, I., Martín, M. T., Martín, S., Santiago, J., Cotarelo, J., & Rodríguez, C. (2013). Efficacy of myofascial trigger point dry needling in the prevention of pain after total knee arthroplasty: A randomized, double-blinded, placebo-controlled trial. Evidence-Based Complementary and Alternative Medicine, 2013, 694941-8. doi:10.1155/2013/694941
Kahneman, D. (2011). Thinking, fast and slow. Toronto: Doubleday Canada
Ulus, Y., Tander, B., Akyol, Y., Durmus, D., Buyukakıncak, O., Gul, U., . . . Kuru, O. (2012). Therapeutic ultrasound versus sham ultrasound for the management of patients with knee osteoarthritis: A randomized double‐blind controlled clinical study. International Journal of Rheumatic Diseases, 15(2), 197-206. doi:10.1111/j.1756-185X.2012.01709.x