NZ Skeptics Articles

Stop crediting the placebo effect, it doesn't make you better!

Brad MacClure - 8 July 2024

Over a year ago my wife and I were on a cavalcade. Present on the same cavalcade was a woman who was an equine vet. While we were discussing various ‘alternative’ treatments and such she happened to say “but I’m a real believer in the power of placebo, though…”. I found that statement a little surprising, coming from (I assume) a qualified medical person, but perhaps in a way this is telling.

Perhaps this is the difference between a vet and a vet who is also an epidemiologist, or an academic who reads scientific papers a lot? I honestly don’t know, and I’m surely less qualified than the person who made that statement.

However in my opinion (and I’m just a schmuck from Invercargill, so please don’t take my word for it) statements like “I believe in the power of placebo” demonstrate a fundamental lack of understanding about what the placebo effect really is, what its place is, and why we refer to it.

Thinking of the placebo effect as “mind over matter”, or equating it with the “power of positive thinking” is wrong in my opinion. Placebo is a subjective effect. Saying you “believe in its power” can be an excuse for letting someone continue to use an unproven or ineffective modality. It can assuage the cognitive dissonance you experience when perhaps you know there’s likely no useful therapeutic effect happening, but you want to let the person making a dubious claim off the hook.

The place where placebos are truly useful is in clinical trials, and the usefulness has nothing to do with any therapeutic effect on the patient. What placebo is about in that setting is to control for its effect, so that you know how significant the effect is of the real therapy or drug. Placebos are literally just one of the tools in a scientist’s kit for finding out what’s really happening, so that they don’t unduly credit an effect to the treatment being tested. I think we have to stop saying things like “yes, but the placebo effect is real” or “well, at least there’s an effect there”.

From Wikipedia:

A Cochrane review in 2010 found that placebos do not appear to affect the actual diseases, or outcomes that are not dependent on a patient’s perception. The authors, Asbjørn Hróbjartsson and Peter C. Gøtzsche, concluded that their study “did not find that placebo interventions have important clinical effects in general”.

This is the study referenced by the Wikipedia article:

Hróbjartsson A, Gøtzsche PC (January 2010). Hróbjartsson A (ed.). “Placebo interventions for all clinical conditions” (PDF). The Cochrane Database of Systematic Reviews. 106 (1): CD003974. doi:10.1002/14651858.CD003974.pub3. PMC 7156905. PMID 20091554. Archived from the original (PDF) on 2019-04-02. Retrieved 2018-06-25.

If you’re wondering what is meant by “outcomes that are not dependent on a patient’s perception”, it’s pretty much anything except for pain. Because pain is subjective, it’s dependent on perception.

Confounding factors such as regression to the mean (I reckon the best way to think of this is to take an example of a patient getting better on their own, or who would have gotten better on their own, but instead the intervention is being credited with having an effect) can often be wrongly called a placebo effect. Also, just like regression to the mean, other interventions which you didn’t account for, or changes made at the same time as the intervention, can also then have their effects credited to the intervention, or falsely be called part of a placebo effect. It would solve a lot of issues if we just stopped referring to ‘the placebo effect’, except when discussing a clinical trial.

One more thing: you sometimes see dodgy modalities (acupuncture springs to mind) that refer to studies that show “no effect beyond placebo” as somehow a success story, because now “we can say that our treatment had an effect”. However, the “no effect” part of that statement is the take-home message there.

I have a great example that I hope will demonstrate what I mean. This article was originally going to be about horse-based pseudoscience, but as you can see I ended up getting into the weeds with this topic instead! While I was reading up about different horse therapies, I ended up on the website of a person offering many different horse therapies. There’s a “case studies” section on that site which I found interesting, and it may be a good exercise in how to spot flaws in, err… logic? Advertising? Sorry, I don’t know what you call it really.

If you want to look the case study, it’s here:

Now, I’m sure this woman is very sincere. Also, I’m not claiming her therapies, some or all of them, don’t work. What I am claiming, though, is that I can’t find good evidence that they do work. I’ll let you get deep into the details if you want, and I might even do so in a future article. But, for now, here’s the guts of the case study:

  1. A horse was lame in the paddock (walking on three legs)

  2. She took him into the barn, and commenced acuscope therapy

  3. After 5 days of daily acuscope therapy, the horse was a bit better

  4. After 3 more treatments, over 6 days (now 11 days in total) the horse was trotting a bit

  5. At day 21, it was walking a bit better

  6. Over time he fully recovered, and in 2009 (2 years later) he won some national showjumping titles

I don’t think you’d call this a scientifically conducted case study, it’s more of a testimonial. It’s also the only ‘case study’ on the case studies page… a bit like me calling my one fan my “fan base”, but I digress. In any case, I presume you can spot some issues with this.

This ‘case study’ is itself a case study in why we need proper trials for these modalities, and it’s also a case study in why testimonials are very weak evidence.

Here’s a list of what springs to my mind:

  1. They took the horse out of the paddock and into the barn. This is not a bad thing to do, but how do we know that the change of environment isn’t the thing that helped? He may have been eating something bad out in the paddock, or may have been kicked, etc.

  2. After 5 days a bit better, and after 11 days better still, etc. Sounds a lot like regression to the mean to me.

  3. What was the diet while in the barn? Presumably not grass, like it would have been in the paddock. Did he receive anti-inflammatories? Or medicine? We don’t know. Did the vet visit during this time? We don’t know. We assume not of course, but we just don’t know.

  4. Yada yada… we won the show jumping. But that was 2 years later, so we’re expected to believe it’s all down to the therapy?

I know that was a bit of a left turn away from placebos, with the horse stuff, but if you’ll allow me I think I can bring it back: placebos should only be used and discussed in the context of clinical trials. Bandying them around as if they have a physiological effect, and implying that that makes it okay to endorse ineffective or unproven modalities, is BS - and we should fight against it. We should also educate people about it whenever we can. It’s dangerous, and it’s dishonest. The above study, while it’s about a horse, it’s a good example of how these so-called alternative practitioners work and think. A good example of the sort of argument they use to persuade gullible people to part with their cash. People who perhaps don’t know the difference between a testimonial - a single case study with zero controls that is likely missing important confounding factors - and a properly placebo-controlled clinical trial.