If it works, it seems very humane to prescribe placebos rather than prescribing things that do permanent damage or have high risk of doing so. For instance, it seems reasonable to try a placebo before a SSRI, since placebos have less risk of triggering an avoidable suicide. If they don’t work you can move on to Prozac afterwards. Why would that bother you? What’s wrong with “first, do no harm”?
Oh, I get it now… we must not consort with witches.
EDIT: re-reading the above it seems harsh. Sorry about that, but I haven’t time to frame it better - gotta go fix some stuff.
Yes, but that’s a big “if” in my opinion. I think research like this wrongly assumes an answer to the question of whether there even is a “placebo effect” to begin with. I don’t think there is enough good evidence right now that can’t be attributed to poor study design, bias, and statistical noise. So many of these studies are done on small samples with self-reporting, non objective measures. It is true that there are some non-physical interventions that can have a physiological response, especially in the perception of pain and I definitely support that effort. If just the activity of seeing a doctor is an effective treatment, I’m all for it, but I don’t think deception should be on the menu.
My comment about alt med was a bit flippant, but I think the placebo is the last bastion of hope for the alternative medicine cause. Over and over their treatments are no better that controls (placebo), so now they claim that placebos are actually effective, so alt treatment must be too! This is a deep logical hole to fall into and really throws all standards of evidence out the window…
I don’t think your comments are harsh at all, so no worries! (hopefully you don’t think mine are either)
I was too quick to assume that you had not looked at the researcher’s work but were just voicing theories about other people’s motives that you didn’t know about. My apologies.
Entirely true. But it’s a huge step from self-deceiving scientists who twist facts to fit their pet theory, to real manipulation along the lines of “actively attempting to justify” something. Hanlon’s Razor.
Interesting. I haven’t heard that argument myself yet, I’ve only heard people attacking alt med by pointing out how effective placebos are. And after all, how can the alt med people profit if sugar pills without D20-Diluted poisons will work just as well, and it really doesn’t matter where they stick their needles? The alt med fans I know are still trying to market their superior ancient knowledge and understanding of the body’s “energy flows”.
Cheers, I appreciate that. I was worried for a second that I really had come off like a crazy conspiracy theorist! I know what you mean though, I definitely have been guilty of putting an opinion out there pre-maturely.
Yes, absolutely. Everything I mentioned I gleaned from listening to “The Skeptics Guide to the Universe” podcast and “Skeptics with a K”– they have covered a few cases where people reverse the standard of evidence, but you’re right that in most cases it goes the other way and the control group is treated like it should be. That’s what I mean about it being a last bastion; they really have nothing else, so are clinging to supposed placebo magic.
That’s just a small sub-set of alt-med, mostly from earnest researchers who are searching the cracks of studies for a negligible effect that may just be an artifact of research. It’s too bad. The " ancient knowledge energy flows" people are doing just fine; they don’t need any standards of evidence.
[quote=“tgarretteaton, post:64, topic:101769, full:true”]
Yes, absolutely. Everything I mentioned I gleaned from listening to “The Skeptics Guide to the Universe” podcast and “Skeptics with a K”– they have covered a few cases where people reverse the standard of evidence, but you’re right that in most cases it goes the other way and the control group is treated like it should be. That’s what I mean about it being a last bastion; they really have nothing else, so are clinging to supposed placebo magic.[/quote]
Okay…
I can see that you mean well here, but you’re basing your arguments on a partial understanding of reality, and you’ve gone a bit off base.
My knowledge of this subject comes from my time as a medical research scientist. You can see my doctoral thesis here:
The placebo effect is simultaneously
The reason why physiologically inactive bullshit alt-med quackery nevertheless often has some minor therapeutic effect, and
A genuine and significant factor in the effectiveness of scientific, evidence-based medicine.
The placebo effect is not just an explanation for the otherwise mysterious limited effectiveness of sugar pills; it is the manifestation of the hugely complicated way in which the physiological repair mechanisms of the body may be influenced by cultural and psychological means.
We account for it in research because we know that the psychological influence of the ritual of medical consultation itself has a consistent, measurable impact on health outcomes. This effect is not small; very often it outweighs the impact of whatever pharmaceutical or physiological treatment is being tested or delivered.
While it is true that a minority of quacks have retreated to a misuse of placebo terminology as a last-ditch defence of their unethical profession, this is no justification for discarding the baby with the bathwater. The purpose of modern scientific investigation of the placebo effect (overall, I’m not talking about one questionable researcher here) is to discover the ways in which the benefits of psychology may be best applied to maximise the therapeutic benefits of already effective treatments.
As an example: the evil cousin of the placebo effect is the nocebo effect. It’s the phenomena of people experiencing “side-effects” from drugs, even when the “drugs” are inert.
If you give people a sugar pill but tell them that it might make them nauseous, a statistically significant number of them are going to puke. And nocebo affects real drugs as well: people tend to get the side-effects that they expect to get, and the more effort that you put into warning people about possible side-effects, the more likely those side-effects are to occur. So how can we best inform patients without making them sick in the process?
We also know that medical treatments are more effective when patients are happy and confident about their medical care. White-coated doctors in fancy offices who can afford time to establish rapport with their patients get better results than hurried and scruffy medics in grungy clinics. So what is the most cost-effective means of achieving that benefit?
Etc. Placebo research isn’t just about sugar pills.
Wow. This is the pretty arrogant. So you are basing your arguments on a full understanding of reality? Sorry, I didn’t realize there was a timelord in this thread.
You’re not even responding to any points I made, and nothing you’ve said contradicts any of them, so it just looks like dick wagging. Why even do that?
Please don’t come around to talk down to people; it hurts good points you make. I acknowledge being a layperson, but I understand more than you’ve given me credit for.
Your citation of SGU as an authority suggested that you had some appreciation of expertise.
Since you want point-by-point, though:
You don’t need to be “primed” to believe in the placebo effect; it’s an innate feature of psychology.
Even in animal studies, we control for placebo. When I’m giving my treatment rats a drug injection, the control rats get an equal-volume saline injection at the same time. Otherwise, there is no way to know if any neurological, physiological or behavioural differences are the result of the drug or the result of the distress and discomfort of the injection.
The psychological impact of that distress and discomfort is a part of the placebo effect. That is what I am trying to get across to you: the placebo effect is not just an excuse for quackery, it is the entirety of psychology’s influence on the body.
The primary focus of placebo research is on enhancing otherwise effective treatments and exposing ineffective therapies. Nobody with serious influence within the evidence-based medicine community is proposing abandoning effective therapies in favour of purely placebo treatments. What they are advocating is using every tool available in order to make the most of what we have. The Goldacre video I linked upthread addresses this point.
That isn’t a very plausible hypothesis.
The reality of the placebo effect is supported by a century’s worth of medical research. It’s about as well-established as the existence of gravity.
Yes, pain is subjective: it is a psychological phenomena. When people are psychologically influenced to “trick themselves” into feeling less pain, they actually are feeling less pain. We have a fuckton of neuroscience research to back that up these days.
It is very well established that placebo effects influence physiological measures.
Pacemakers improve heart function after they’ve been implanted but before they’re switched on. Two sugar pills a day heal gastric ulcers (a physical, observable symptom) faster than one sugar pill a day. Altering the verbal instructions given to subjects in a drug trial can measurably influence the rate at which drug molecules are eliminated from their bloodstream.
Part of the control arm.
More typically, you’ll have at least three groups. For something with no effective treatment, that’s likely to be “no treatment”, “placebo control” and “active drug” groups. If there’s an existing effective treatment, you’d normally replace the no treatment group with a standard-of-care group.
For even better control, you want an active placebo group: something that doesn’t directly address the health issue in question, but has enough of a physiological effect to convince the subject that they’re getting an active treatment.
In animal research, we often also use a negative control: something where you already know what result you’re going to get, conducted just to prove that your experimental technique is competent. For example, you might have a group of rats that are dosed with a well-researched drug and run through the same behavioural tests as your experimental critters. If the negative control responds unexpectedly, you know you’ve fucked up somewhere.
If a placebo treatment group shows significant improvement over the no-treatment group, this is evidence that the placebo was therapeutically effective. That doesn’t mean that you abandon all effort at finding physiologically effective treatments; it just means that the influence of placebo is a factor to be accounted for.
From a medical point of view, we don’t care if the patient’s healing is “falsely” psychologically driven. A healthier patient is a healthier patient, however they got there.
The placebo effect is neither a mysterious force nor a novel explanation. It is an extremely well-established medical phenomena which is the subject of active and productive research.
It is not impossible to disentangle placebo from other confounding factors; that’s the whole reason why we have placebo control groups, in order to disentangle that factor. No treatment vs placebo control: the difference between those two groups is the placebo effect, with all other confounds equally affecting both groups. This is basic experimental design.
Again, see the Goldacre vid upthread. Nobody with any influence is seriously advocating substituting sugar pills for modern pharmacology. Concern about that is getting into conspiracy territory.
There is no bizarre pathology involved. The placebo effect is a well-documented medical phenomenon, and current research is focused upon exploring its subtleties rather than questioning its thoroughly-established existence.
There are some real issues surrounding the use of placebos in research, but they aren’t what you appear to think they are. Some are hinted at upthread, in my discussion of nocebo: what to do when patient health and patient autonomy are in direct conflict? There’s a good introduction to some others in this video:
Start at the five minute mark if you don’t want to watch the whole thing.
Woah, no need to spill so much ink! I’m afraid you’ve willfully misconstrued most of my points, and are really reaching to find any faults in what I’ve said. In fact, I stand by all of it. I’m on your side, and generally agree with much you’ve said– you really don’t contradict me, and there’s no extra points won for being ‘most right.’
But… Ben Goldacre has some of this stuff really wrong. I like his work overall, but many of the studies he cites are poorly conducted or just mis-represented. I really encourage you to listen to the Skeptics with a K podcasts from the last few months– they specifically break down a lot of the Goldacre stuff– even that pacemaker study. Lots of the studies in “Bad Science” are poorly controlled, fail to be replicated, are based on self-reported measures, and use small sample sizes.
This is very declarative, where I think there is room for nuance. The placebo effect is many things. It does encompass phycological effects on the body, but when so many studies have poor blinding and are based on self-reporting, it is difficult to separate those effects from other factors. Regression to the mean mixed with expectation bias and clinician bias can skew results pretty well. This isn’t my own pet theory, check out Science-Based Medicine on the topic.
Lastly, you are not an authority on this, and neither am I. I respect that you have science training and are proud of that, but even people trained in science can be wrong about areas outside their expertise. I trust you know what you’re talking about in general, but I don’t consider a credential as sacrosanct and I doubt you do either. It is always better to treat the people you are trying to convince of something with a modicum of respect; I really think you need to re-read your own posts in this thread and reconsider the tone you’ve taken with me and others. Besides the content, the way you’ve gone about sharing your opinions does nothing to make me want to change any of my positions and I think others would feel the same.
Thank you for the extensive, well written and nuanced post. It works as a great complimentary piece to the original topic post. Although I’m not a big fan of Goldacre, personally.
