The first time I experienced the phenomenon known as the “placebo effect” was in a Star Trek: The Next Generation episode “Realm of Fear.” In this episode Lt. Barclay (Dwight Schultz) believes that he is coming down with a Star Trek universe disease known as “transporter psychosis.” The disease, by the in-universe time of the episode, has been eradicated. The ship’s counsellor, Deanna Troi (who is also a bridge member for some reason) believes that Barclay’s problem is anxiety and she recommends that he tap a spot on his neck to release endorphins. Perhaps, in-universe this is a legitimate treatment, but I recognised this treatment as a way of giving Barclay something to distract his character from his transporter anxiety.
This is a placebo in that there is no mechanism of action, but it is effective because while Barclay is tapping on his neck, he’s not worrying about the space ghosts in the transporter (of course it turns out that he’s right, there is something in the transporter stream, but that is irrelevant to this discussion.)
The intervention works because it helps Barclay calm down, but it is not a treatment—it is a placebo. These are the kinds of placebos we do not usually blink at: tell someone that a sprained ankle will feel better if they find a smooth stone works because the sensation of pain is subjective (not pain itself mind you, but the sensation (the “qualia”) of it) and it can be distracted. A person suffering from minor depression can “forget” about it temporarily with the introduction of a new hobby.
In both cases we can forget about our conditions because they are minor. Barclay is not suffering from a long-eradicated condition, and I am a high functioning depressive that can be “nudged” out of a minor slump. However, there are those that seek to argue that the placebo is the cure itself, and their most academic voice is Dr Kathryn Hall.
I first came across Dr Kathryn Hall on episode 45 of the television show “Adam Ruins Everything.” Up until this episode I rather enjoyed the iconoclastic nature of the show as he debunked myths surrounding hygiene and the death industry. I thought it was very clever that the show would put up bubbles as sources, which saved me time in researching some of them—a thing I do sometimes as both a skeptic and an academic.
This episode has been covered before in this magazine, so I’ll give a summary of the general problem. In this episode, Hall points out a few studies which showed that a medically-accepted intervention (to avoid alternative between surgery or pill I am just going to use the word “intervention”) performed just as well as a placebo (which is to say—nothing) so; therefore, thinking something happened is just as good as something happening. As Mike Hall (no relation) has pointed out many times on the Skeptics with a K podcast, this is technically true but it implies the wrong thing; the correct interpretation is to say that if the placebo is effective then the intervention we have been testing has no effect.
“Essential” Knowledge?
Last month Dr. Hall published “Placebos” in the MIT Press’s Essential Knowledge Series. It is very unfortunate that this book is characterised as essential knowledge, because the idea Hall promotes throughout the book is the same as the one promoted on the Conover’s show. Over and over again she comes to the same conclusion: that placebos represent a mind over matter when it comes to treatment. She writes in the conclusion:
Page 15In focusing on why the drug failed, we are missing that other mechanism, the one that didn’t fail. This other mechanism that gives rise to the placebo effect .
Hall is discussing the failure of a drug in a clinical trial, and that we should focus on patients who did receive the active treatment to investigate why they reported improvement. This is not a bad idea, provided it was a significant portion that improved, but we are going to be left in the same place where we started. The only thing that we found out is that the intervention we were testing should be tossed. In the case of a pill, it makes sense: if one half of the experiment improved but received nothing more than a sugar pill, we should reconsider whether any treatment at all is worthwhile – provided the methodology of the experiment met acceptable criteria.
With regard to surgery, Dr. Hall uses the example of shoulder athroscopy for sub-acromial shoulder impingement, and her conclusion that the sham therapy and athroscopy resulted in the same benefit fails to account for post-operative physical therapy. This was covered in the article above, yet she commits the same oversight in the book (pg. 128), without considering that everyone suffering from sub-acromial shoulder impingement could just be sent to a physical therapist instead of either surgery or sham therapy.
I’m not a medical doctor, so take this claim with the appropriate skepticism, but let’s frame Hall’s claim a different way: shoulder athroscopy is an invasive surgery in which bone spurs are scraped out of the shoulder to improve movement. If Hall is correct in that belief is as effective as this intervention then Hall is claiming that belief has removed the bone spurs. When phrased like this it seems less believable that the placebo effect has treated the condition.
Hall goes through a list of similar examples, portraying each one as a case of when a surgery is done the patient is healed. The list completely ignores any post-op therapy, yet she concludes that:
Page 136some surgeries steeped as they are in ritual, are potent inducers of placebo effects.
Hall’s framing of the placebo effect is one in which the effect is a positive phenomenon that would not have happened otherwise. This could be true in that there is an effect that ritual has on an outlook, but those are only for those effects and conditions which are both subjective and minor. The placebo effect is not going to swoop in to erase those bone spurs (which apparently have nothing to do with the condition) or cure someone suffering from severe depression.
Dismissal of alternate explanations
The most frustrating portion of this book is that Dr. Hall gets so close to the correct explanation only to scuff wide of an open goal. She spends the entirety of chapter 2 discussing expectations, where a person with a condition really wants the intervention to succeed, and later explains the Hawthorne effect, where people behave differently when they know they are in an experiment (i.e. the inmates and guards in the Stanford Prison Experiment). Hall discusses the natural course of the disease, such as the common cold which eventually just goes away, and she discusses regression to the mean—where a person fluctuates between feeling better and worse. Yet Hall is apparently blinkered from seeing these phenomenon as the likely explanations for the “placebo response.”
Dr. Hall also mischaracterises the history of medicine as well. Hall begins the book with a historical overview of the placebo effect, and two early attempts at clinical trials to test Mesmer’s Galvanic baths and the Perkin’s tractors (two sticks which were supposed to alleviate pain). The former trial was conducted by an A-Team of the day’s natural philosophers: Benjamin Franklin, Anton Lavossier, Jean-Sylvain Bailly, and Joseph Ignace-Guillton.
The dream team of natural philosophy determined that the effect of Mesmer’s treatments was the same as when they faked all of the pomp and circumstance as the authentic treatment. It did not matter, for instance, if they used mesmerised rods or rods that they claimed were mesmerised. The commission declared, according to Hall (pg. 26) that something was going on but a sham mesmerised bath and a “real” mesmerised bath produced the same results. Hall concludes from this that,
The key finding was not that these strange and bewildering remedies didn’t work; it was that the sham devices worked just as well as the real ones.
Page 22
Which is saying the same thing twice. Yes, the sham devices worked as well as the authentic ones—because neither of them worked.
There can be many reasons that the patients told that they were going to be treated with Mesmer’s techniques would react positively. One striking reason is that they sunk a high cost of not only money but their own personal endorsement thus it had to work. No one wants to be considered a fool. While I am hypothesising, the evidence in support of my claim comes from Charles Mackay who writes (which I took from the Wikipedia page on the investigation) in his 1841 “Extraordinary Popular Delusions and the Madness of Crowds”,
…that those effects could be produced without passes or other magnetic manipulations—that all these manipulations, and passes, and ceremonies never produce any effect at all if employed without the patient’s knowledge; and that therefore imagination did, and animal magnetism did not, account for the phenomena.
That Hall mentions Mesmer’s treatment, then the investigation, only to come to the conclusion that they are both equal in a good way is incredible. It’s further incredible that she endorses homeopathy as treatment because of its placebo effectiveness (pg. 26). This is accomplished by maligning medical science in comparison as “Western Medicine”… as if the German-created homeopathy isn’t of “the West”.
Her lauding of integrative medicine is difficult to understand. If homeopathy or hydrotherapy or so effective at eliciting placebo responses in patients thus making them effective, then why do we need to integrate them?
She also uses the 18th century patients’ reluctance to submit to “regular medicine” and favour “irregular medicine” like the aforementioned homeopathy. This ignores the problems of 18th century regular medicine’s reliance on a treatment modality that can be summed up as “feed them whiskey and then bleed them until they feel better.” If the same person that cuts my hair was responsible for removing the weird growth on my leg, I’d roll the dice on homeopathy, too. The preference does not prove anything, and as an American, I’d be less likely to undergo the treatment which killed the previous unkillable George Washington (the attending doctors drained five pints of his blood (2.36 litres) in a misguided effort to treat him).
Beecher’s Powerful Placebo
Hall relies on a 1953 study “The Powerful Placebo” by Henry Beecher published by the Journal of the American Medical Association. She reports that Beecher’s conclusion justified reliance on the placebo effect as a viable intervention. However, all of the necessity of the placebo that Beecher discusses is within the framework of an experiment. The paper itself opens with:
Beecher 1953…it is only recently that recognition of an enquiring kind has been given the clinical circumstance where the use of this tool is essential…
Before leading into an earlier definition of the term “placebo.”
It is used to eliminate bias and its effectiveness is practical in this regard, but relying on the effect as a treatment qua treatment and not as a comparison is an error unless it is being used to please the patient (i.e. make them, like Lt. Barclay, go away).
Beecher does admit that placebo response is measurable in pain relief but that proper definitions need to be considered:
Beecher 1953Certainly, in these and other studies shown in table 2 the validity of the thesis here (namely, that the placebo can have powerful therapeutic effect) hinges largely on the definitions of “satisfactory relief”
The further problem Hall has with relying on Beecher is that not only does she mistake Beecher’s intent, according to Kienle and Kiene’s (Kienle and Kiene 1997) analysis of the Beecher’s article:
The overall result was that for none of these trials (the ones used by Beecher) was there any reason to assume the existence of the slightest placebo effect…However, on the basis of the published data, in all of these trials the reported outcome in the placebo groups can be fully, plausibly, and easily explained without presuming any therapeutic placebo effect. [emphasis authors’]
Hall cites the Beecher paper because she’s almost obligated to do so. Anyone writing about the placebo effect is going to have to cite this paper—that is how important it is to the world of placebo enthusiasts. However, for someone of her stature to get it this wrong, or, at least offer up some defense of Beecher’s conclusions instead of betting that no one would be familiar enough to read it is hard to understand. Of the many questions that we should be asking is whether or not she’s aware of Kienle and Kiene paper (I, admittedly was not until I was between drafts on this article), but they offer a damning condemnation of the placebo effect:
There can be no doubt that the extent and frequency of placebo effects as published in most literature are gross exaggerations.
In conclusion
Much has changed in medical science since the 18th century and even again since the 1950s – but what hasn’t changed is the need for things like the placebo effect which offer the necessary comparison to determine an effective novel treatment or at least rule out something that doesn’t work. The latter is whether the placebo is most effective: if acupuncture does no better than placebo it means that acupuncture is not an effective intervention in that study – not that placebo is as effective.
Hall’s book is exactly this. It is illustrative in its portrayal of the placebo effect for how the public hears it. However, they only hear it that way because of books like this and shoddy reporting on medical science in the English-speaking world. Hall is trying to claim that there is something extra-normal going on; that belief somehow shapes reality, but reality always wins. The placebo effect cannot make anyone better, just as Hall does not make the case here.
