Placebos work even when you know they don't contain medicine

boing
41

"The placebo effect is a genuine medical phenomenon that must be accounted for in all medical research, regardless of how inert or active the pharmacology under investigation is. It is also an important part of clinical effectiveness, again regardless of how active or inert the other portions of the therapy might be.

No, No, a thousand times No.

It is a scientific phenomenon, not a medical one. And there is NO clinical effectiveness of a placebo - that is the whole effing point!!! Water does NOT have clinical effectiveness. Sugar does NOT have clinical effectiveness.

What they DO have is clinical confoundedness. They give false positive results in clinical trials - and ALL of these results have absolutely nothing whatsoever to do with “clinical effectiveness” and lots to do with how humans react to being in clinical trials. “Clinical effectiveness” is - in the setting of a clinical trial - EXACTLY the effectiveness of the active agent BEYOND that of the placebo. LOL.

Because, a placebo - by definition - does not have clinical effectiveness.

I am afraid, good sir, that you do not “get it” either.

1 Like
42

It seemed like you did say that when you said:

“I think others may actually receive some actual effects of the Placebo. That is, even though there is not chemical reason they feel better, they feel better. The mind is pretty mysterious still on how we perceive pain some times or how or why we feel how we feel. Even knowing something shouldn’t work might still make a real difference because our mind decided it should. I honestly think that is how it helps half the people who try crap like holistic medicine. “This shouldn’t work, but I will give it a shot.” And then it does even though it shouldn’t.”

Sorry if I misinterpreted you. :slight_smile:

43

I have heard and read about how difficult it is to test medication and get actual data. Telling someone about a test will inherently skew the results, and depending on the nature of the effectiveness of the medication the results of placebo vs the real thing it can be inconclusive. Which must be frustrating for researchers.

44

I appreciate you measured response, but I fundamentally disagree. The “placebo effect” is a constellation of effects– not one. There’s no need to create a mysterious force to explain positive outcomes in the placebo arm. Remember it is the control arm of a study– it is the garbage bin where unaccounted effects are measured against an actual treatment. It could be that some small portion of positive results in the control group are due to some kind of chemical effects of a patient merely expecting to get better. But it is impossible to disentangle those effects from confounding factors such as bias, study design, random chance, error, and deception (self or otherwise). The first step is to try and eliminate those confounding factors before resorting to a novel explanation.

I think it is important to point out why this study was probably conducted: right now you cannot ethically prescribe a placebo to a patient, but if you could show that there is an effect even when the patient knows its a placebo, then you could ethically start prescribing sugar pills. That’s what’s going on here. I would suspect it is because they genuinely think it can help people, but I think they are deceiving themselves and their patients.

I don’t mean to dismiss your point of view out of hand. It could turn out that what you say is true, that part of the placebo effect is down to bizarre pathology we don’t know about, but we have no reason to jump to that conclusion yet. In my view, positive effects in the placebo arm can be attributed to the errors and bias I mentioned, without needing to invent new explanations. Like you said, medicine is hard and these kind of studies really illuminate how challenging it is to run good trials. More evidence may emerge that says I’m wrong– we’ll see.

45

That is not correct, as @Wanderfound says.

Answer me this: if there are no effect of placebos, what is the point of giving them as the control arm of a study, vs not giving anything? By your statement, there should be no difference between those two versions of a control group. So why do gold-standard studies insist on a (well-designed) placebo control group?

The reason, of course, is to eliminate the effects of the placebo alone. It is known that a doctor handing someone a pill will result in a decrease in their symptoms for many ailments, vs not giving them anything.

So, if we are studying to see if Drug X works to treat depression, say, we can’t just compare patients taking Drug X with patients taking nothing, because we already know that handing someone a pill may improve their symptoms. So we baseline it against a placebo, to see whether Drug X works better than can be accounted for by the placebo effect alone.

2 Likes
46

No - wanderfound is deeply misguided on this topic. let me explain:

First of all - there are two basic types of controlled studies - those with inactive controls = placebos, and those with active agents (usually approved drugs that have been on the market for quite some time) as a comparator.

We must use some control(s) so we can see how the results of the active agent have been “confounded” (there is that word again!) by unknown factors. In medical trials, we usually only have one or two control arms, because of cost and ethical factors. This helps to give us a basis for deciding if the effect of the active agent can pass the statistical hurdles needed for it to demonstrate efficacy of “statistical significance”.

I used to do pure biology research. Each experiment I ran might have twenty or more controls - to try to account for and identify different confounders.

The whole point is that it is important to understand how drug trials work, so that we will not be confused into thinking that:

  1. placebos have any useful effect

  2. drugs that do not outperform extremely confounding placebo effects have no real value

Example - antidepressant drugs. There has been a huge and misguided uproar about antidepressants not outperforming placebos and how this is all a conspiracy by Big Pharma to foist useless drugs onto depressed patients. This is dangerous nonsense which stems from the fact that in antidepressant trials, one sees a VERY large placebo efficacy effect. The placebo arm in these short trials may show 65% of patients showing improvement, with only, say, 70% improvement in the active drug arm. This may well not rise to statistical significance.

People with an understandable skepticism against Big Pharma have rushed in and concluded (well, shrieked!) that antidepressants are therefore a farce. Well, I can tell you that they are NOT a farce, and that without them there would be millions more successful suicides every year. Looking back at the theoretical study results I described to you, what we should conclude is that the active drug showed efficacy in 70% of patients. Unfortunately, the placebo did almost as well in that category. But antidepressant studies have always been known to have a very high rate of being confounded by placebos - patients are forced to get out of bed, groom themselves, go outside, interact with caring people, be the center of attention, etc. This affects their self-evaluations.

The experimental situation is doing that - not the placebo! Remember - the placebo is water or sugar! Yet, that IS the placebo effect, the true “power” of the placebo - to give ->false<- positive results. (Yes - this means antidepressant clinical trials (and other types) are difficult to perform and interpret!))

Also, remember - these false results in clinical trial situations are not a real clinical medical effect. In real life, if you are given sugar pills to treat major depression, any self-deception caused by you taking what you think is a real medicine will have a very minimal and very short-lived effect on your depression. And this is why placebos are unethical to use as real therapy in real patients outside clinical trials. And why I, and many others, believe quack medicine like homeopathy, reiki, acupuncture, etc - which are nothing more than glorified placebos - should be illegal.

1 Like
47

But your description just seems to be describing a poor study design. If the working drug does not perform statistically-significantly better than the placebo, how does the study show it’s a working drug? In other words, how would the study’s results have been different if it were a non-working drug?

Would you agree that the study didn’t prove the effectiveness of the drug (which may very well work, as you say)? If the drug really does work, and yet the study did not show that, it sounds like a better study should have been done.

So it sounds like you’re saying there is no value in placebo-controlled studies. So why do you think they remain? Just historical red-tape that everyone has to jump through?

1 Like
48

No, most studies are not so highly confounded and the results are more clear-cut. Antidepressants are tough to study - they require very careful protocols, lots of repeat studies, and different types of studies as well. Sometimes just longer-term studies to tease out the differences. Sometimes longitudinal studies, where large populations of patients are studied for very long times vs cohort populations which have different drugs or protocols.

Psychiatry is a bit of an art as a consequence. Many patients require combinations of approved drugs, tailoring of therapies to symptoms, etc. But psychiatrists see real-world patients getting real drugs over time. They see how their patients respond. We also know what happens when depressed patients stop taking their drugs if they become noncompliant, or are put on a drug challenge - where they are carefully observed after the doctor takes them off the drugs. The results are compelling: patients with major depression who quit their drugs relapse and commit suicide with increasing frequency after each challenge. By the time you have a patient who has relapsed five times, if they are left untreated as many as 95% of them will die by suicide. (!)

So, it is obviously true that 1) most antidepressants do help about 70% of patients they are tried in, at least for while and 2) antidepressant studies are highly confounded by placebos. So, medicine progresses, just not easily.

49

And can intervene if the patient reacts badly, which is a significant risk with all those drugs. In France (to speak of what I know), the conventional protocol is to set a first appointment ten days into a new treatment, for control.

50

Superb post, I hope everyone reads it.

If you haven’t read Jennifer Wright’s Get Well Soon I highly recommend it. It was almost entirely information I already knew (surely it is unsurprising that a medievalist would have studied the great plagues?) but it was so well written and engaging that I didn’t mind getting such a small amount of new info out of it.

It’s not the main thrust of the book, but the author uses several ancient and modern examples to suggest that humans are more likely to be healed by their own bodies if their psyches aren’t being undermined by angry judgemental people telling them to harden up or that their illnesses are hoaxes. The inference is that the psychological solace of having a compassionate caregiver pay attention to you, even if it’s just to give you a known placebo, can reduce psychological stresses that are manifesting physically in ways that are preventing normal healing.

3 Likes
51

Bravado aside, you didn’t read I qualified my remarks by saying it was my experience that MOST people reporting pain are softkoks?. And I knew it wasn’t the sum of human experience?. I would repeat that to your face, grin, and give you a lesson in proper behaviour if you got out of order.

Sorry to hear your wife is unwell, console yourself with philosophy.

52

Time to pull out the Goldacre again.

Here’s one that y’all might not have seen before:

The last 90 seconds are particularly on point.

1 Like
53

I’m not getting it. First, it’s a matter of definitions. Hopefully we all agree that “the placebo effect” refers to the effect that patients are more likely to show improvement if they believe they are undergoing a treatment from which they expect positive results, not the non-existant effects caused by the sugar and water in the pill.

This effect confounds studies because the patients who are getting the real drug that is being tested most likely also believe that they are undergoing treatment from which they expect positive results.

But why do you declare it “not real”? If the experimental situation by itself causes 65% of patients to show improvement, then that is the placebo effect, but still 65% of patients show improvement. You’d need to compare that with a control group where patients are sent home to know for sure whether that’s worth it, though.

Next, why are you dragging statistical significance into it? If you can’t show with statistical significance that a drug works better than a placebo, then you still know nothing about whether the drug works. If you are looking for a small effect, such as having 70% of patients get better instead of just 65%, then you need a big study to know if the effect exists.

You can’t just declare the 65% who improved to be “not real” even if they only got better because they were given the feeling that they’re well cared for and about to get better.

They might have value if they outperform the placebo effects for some people. However, to even know if the drug that has any value at all, you have to show that it performs, and that means performing better than nothing, and you have to show that with statistical significance. And you have to do it on the first try - no jelly bean science, please.

I’m sure there are also real drugs that have short-term effects only. You need to do trials or at least analyze data to know that, of course. I assume the relevant studies have already been done and have shown that in the long-term, antidepressants significantly outperform the placebo effect, by a larger margin than in the short-term studies.

Whether the same is true for irritable bowel syndrome, or for gastric ulcers, I don’t know.
So again, this has nothing to do with the reality of the pacebo effect. If the effects are to short-term to be useful, how does that make them “false”?

No. They’re doing it to better understand the placebo effect, which will, if nothing else, help in better designing studies for real drugs. No conspiracy here. Move along.

4 Likes
54

Yah; when distressed people recieve close attention from an empathetic, skilled caregiver, they unsurprisingly feel better. And when people feel less stressed, their muscles relax, their heartbeat smooths out, their digestion improves, their immune function picks up, etc. etc.

Healing requires energy; your body will send as much as it can spare, but if it’s locked in fight/flight/hide mode that won’t be enough. Getting your mind chilled and happy lets your body concentrate on fixing the problem.

There are plenty of things that the body can’t fix on its own; pharmacology and surgery are also vital tools. But it never hurts to get the innate healing abilities helping things along.

5 Likes
55

People have been known to heal themselves through anger, too; I’ve harnessed that effect myself. And I’ve watched people die of despair, which can happen remarkably quickly.

It is very strange to see people who claim to be evidence-driven arguing that taking actions that provably alter patient outcomes (such as using placebos wisely) are “not part of medicine” or “not real medicine”, isn’t it? Great healers have written about the value of holistic mind/body treatments ever since we’ve had writing.

56

I wonder if it isn’t anger, per se, but purpose. Anger often leads to a purpose. I know a guy who thinks his father lived several extra years because he was at war with a heron eating his gold fish out of his pond.

2 Likes
57
58

My friend wrote this article published recently in the Smithsonian Magazine about placebos he took to help him write.

2 Likes
59

Please save your “move along” condescension for the Last Jedi forum– I don’t appreciate it and I doubt you would either. This researcher and others like him are actively attempting to justify prescribing placebos, or to create ammunition for alternative medicine claims… It could be that he genuinely wants to get to the bottom of why control arms can show positive results, but you have to look at the history of a researcher’s work and who they choose to collaborate with for clues about their motives. Believe it or not, sometimes people don’t follow the facts, but instead frame the facts to fit their hypothesis. Many times studies have been conducted with rigor, but results are conveniently spun to confirm a pre-existing narrative. I’d like to think that all science is done with a clear and open mind, but it’s not that simple; people have biases and might not even know it. Studies confirm: self deception is a hell of a drug.

60

I find I require extra-strength placebos myself.

2 Likes