on the placebo effect

Some discussion over on Slashdot about the placebo effect. Following are some of my posts on the topic.

If I could explain the placebo effect I'd be a millionaire.

The problem is that there are several different things that get lumped under the label "placebo effect":

  • Patient experiences no difference in their perception of symptoms, but feels compelled by social pressure to report an improvement. I.e., "It still hurts as much as ever, but I don't want to disappoint Dr. Smith, so I'll say it's better."
  • Patient has no difference in symptoms, but perceives them differently. The pain signal arriving at the brain is unchanged, but comes to be processed differently.
  • Patient believes in ability of the healer or treatment, gains confidence that they will recover, stress responses are reduced, and the immune and parasympathetic responses are improved.
  • Patient gains feelings of acceptance into their tribe/social group as a result of being tended to by the healer. Stress responses are reduced, and their relationship to their community is transformed; a new psychological perspective may be adopted that changes their "will to live" and perception of their "quality of life". Humans are social animals, and I think the social aspects of healing have been tremendously underexamined.
  • Patient comes to feel empowered over their own health because they are able to take simple actions, and so are eventually led to make lifestyle changes that lead to improvements.
  • Patient benefits from non-specific aspects of treatment. For example, after placebo surgery, skilled nursing during recovery may well have benefits. (I'd like to take this opportunity to point out that every double-blind placebo-controlled study of a surgical technique, has found the surgery to be no better than a placebo cut. Yet many "skeptics" who demand rigorous double-blind studies of "alternative" treatments will go under the knife without a second thought.)

There are probably more things going on too.

Interesting article on the placebo effect by Ted Kaptchuk here. If you can find it, his book with Michael Croucher, The Healing Arts: Exploring the Medical Ways of the World, is an excellent read.

Tom Swiss | the infamous tms | my blog

You cannot wash away blood with blood

I think he is saying that nobody has done placebo-controlled trial of a useful surgery.

Ah, but without a placebo-controlled trial, how do we determine that a surgery is useful?

It's a tricky question. The placebo-controlled double-blinded trial is supposed to be the "gold standard" for medical research, but its structure is biased towards drug therapies. It's hard to do placebo surgery, placebo bodywork (acupressure, massage, etc.), placebo acupuncture, placebo diets, placebo psychological therapies, or placebo exercise programs. Heck, surgery is the easiest case of those - you get to knock the patient out so they don't remember the actual treatment!

Tom Swiss | the infamous tms | my blog
You cannot wash away blood with blood

Are you saying that, say, a placebo appendectomy works just as well as a real one?

Unknown. No study of appendectomy versus placebo surgery has been done; the hypothesis has not been tested. No scientific statement can be made.

Tom Swiss | the infamous tms | my blog
You cannot wash away blood with blood

Really, I think that placebo surgery would have about the same result as placebo plumbing -- i.e. none.

But in those studies where placebo surgery has been used, many patients receiving the placebo improved.

The first placebo surgery test was for a treatment for angina pectoris called internal mammary artery ligation. This was at one time a popular procedure, but it's not used now because in a head-to-head comparison, 34% of those getting the surgery reported improvement, while 42% of those getting a placebo cut reported improvement.

A 2002 study of arthroscopic knee surgery found that the outcomes for a placebo procedure were as good as those of the "real" surgery.

In a 2004 study of transplantation of embryonic dopamine neurons into the brains of Parkinson's disease patients, "Those who thought they received the transplant at 12 months reported better quality of life than those who thought they received the sham surgery, regardless of which surgery they actually received," according to the researcher.

While double-blind tests have not been performed, the vastly reduced mortality from appendicits since the introduction of the surgical procedure should be enough for anyone. It's all about the statistics.

The reduced mortality could be many factors. We have to consider improvements in nursing care, improved antibiotics, or better diet: "The decline of appendicitis cases in the United States since the 1930s has led some to suggest that dietary fiber or household hygiene is important in the pathogenesis of appendicitis. According to the "fiber hypothesis," fecaliths develop more readily in people who consume a diet deficient in fiber, because their stools are more tenacious. Societies with high fiber intake (Asia, India, Africa) have less than one-tenth the incidence of appendicitis compared with locations where fiber intake is lower (Europe, North America). A high-fiber diet speeds stool transit times, reduces fecal viscosity, and inhibits fecalith formation."

by the way, is a fascinating look at that question from a century ago. The author finds a mortality rate of 6.6% in the period before appendectomy was used, and of 7.8% in the first few decades of its use. Of course the mortality rate is much lower today; but if mortality rates actually climbed after appendectomy was first introduced, then clearly the situation is more complex than "cut here, cut now, cutting good!"

Which is not to say that, under the right circumstances, I'm going to refuse an appendectomy. The surgery is a pretty good gamble.

Maybe that's because surgery gives results that are good enough and consistent enough to be above such suspicion.

Uh huh. You want to quantify that some? "Homeopathy gives results that are good enough and consistent enough to be above such suspicion." "Scientology gives results that are good enough and consistent enough to be above such suspicion." "My magic rock gives results that are good enough and consistent enough to be above such suspicion." And so on.

If we are scientifically investigating what treatments are effective, no modality can be "above suspicion". The guys in scrubs selling their cut-and-sew jobs get no special treatment above the guys selling their special blend of snake oil.

Tom Swiss | the infamous tms | my blog
You cannot wash away blood with blood

Well, okay, but don't get carried away here. I've had cataract surgery, and I am DAMNED SURE it was not a placebo surgery.

As a practical matter, as a patient or a clinician, it really doesn't matter if it's a placebo or not. "Do X and the odds are such-and-such that the patient will improve, and the possible complications are so-and-so", is all you need to know.

But from the point of view of a researcher, as a matter of (trumpet flourish please) *Scientific* *Knowledge*, how strongly can we assert that your cataract surgery was more effective than a placebo treatment would have been? Sure, it seems like common sense that a placebo treatment wouldn't affect cataracts, but it also seems like common sense that it wouldn't affect angina pectoris or knee pain.

I had an acupuncture session on Wednesday. I am DAMNED SURE it was not a placebo treatment. But my subjective DAMNED SURE will not carry any weight with skeptics - nor should it.

The danger of a pervasive distrust of surgery...

I'm not suggesting that people "trust" or "distrust" surgery, or any modality.

Tom Swiss | the infamous tms | my blog
You cannot wash away blood with blood


Add new comment

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.
To prevent automated spam submissions leave this field empty.
This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.
Enter the characters shown in the image.