Placebos Without Deception

December 23, 2010 1:09 PM

Meet the Ethical Placebo: "A provocative new study called 'Placebos Without Deception,' published on PLoS One today, threatens to make humble sugar pills something they’ve rarely had a chance to be in the history of medicine: a respectable, ethically sound treatment for disease that has been vetted in controlled trials." [Via]
posted by homunculus (76 comments total)

Previously.
posted by homunculus at 1:09 PM on December 23, 2010

I take Cold-Eeze. I'm pretty sure they do nothing, but they seem to reduce the duration and severity of my colds. I'm pretty sure this is a placebo effect.

I don't care.

Generally I am with the doctors that get pissed about people being fraudulently sucked in by untested claims and slick marketing. The whole "but no one is hurt" argument doesn't fly with me since millions are wasted on unregulated dietary supplements (that may not even have in them what they say). And I suppose this makes me a hypocrite since I am contributing to this industry, but at the end of the day, I'll throw $6 at a sucky cough.

Now if only they would come out with a placebo that would help me with my parotid glands.
posted by cjorgensen at 1:18 PM on December 23, 2010

Orac deconstructs this paper.
posted by Rhomboid at 1:18 PM on December 23, 2010 [8 favorites]

What if the placebo effect were explainable because of a actual medical use for sugar. I mean we just assume sugar isn't doing anything.
posted by humanfont at 1:26 PM on December 23, 2010 [6 favorites]

Mrs. nebblesworth works at the hospital and came home discussing this study. One of the reasons for its shortcomings is apparently the authors couldn't/didn't get enough funding to extend it beyond three weeks or put a whole lot of effort into it.


For those who don't know this study addressed IBS, which is a diagnosis of exclusion with no known physical etiology but that is probably caused and/or exacerbated by mental health issues (stress, depression, etc.).

In light of that I think comment #12 on the Orac piece makes a good point:

Would you also describe as "mind-body woo" the belief of many if not most doctors that mental and emotional factors, such as stress, can cause or exacerbate subjective digestive complaints? Surely you cannot argue with a straight face that it is perfectly reasonable for doctors to tell patients that their discomfort may be related to their mental state, but Woo if they then suggest that changing that mental state might reduce the discomfort.

Orac is right about the problems w/r/t design and execution of the study, but I don't understand this point about deception:

Uh, no. The reason I say this is because, all their claims otherwise notwithstanding, the investigators deceived their subjects to induce placebo effects. Here's how they describe what they told their patients:

Patients who gave informed consent and fulfilled the inclusion and exclusion criteria were randomized into two groups: 1) placebo pill twice daily or 2) no-treatment. Before randomization and during the screening, the placebo pills were truthfully described as inert or inactive pills, like sugar pills, without any medication in it. Additionally, patients were told that "placebo pills, something like sugar pills, have been shown in rigorous clinical testing to produce significant mind-body self-healing processes." The patient-provider relationship and contact time was similar in both groups. Study visits occurred at baseline (Day 1), midpoint (Day 11) and completion (Day 21). Assessment questionnaires were completed by patients with the assistance of a blinded assessor at study visits.

I guess I don't get what's deceptive about that? AFAIK everything in those statements is true.
posted by r_nebblesworthII at 1:47 PM on December 23, 2010

nebbles: the deception is what usually happens with use of placebo. Patients are told they're on a new experimental drug when in actual fact they have a 50% chance of just being on placebo. The point of this outcome is that that deception in double-blind trials may not be required - but good luck convincing the world that double-blind is no longer necessary.
posted by polyglot at 1:54 PM on December 23, 2010

What if the placebo effect were explainable because of a actual medical use for sugar.

That's been investigated and they found a few examples where the inert placebo ingredient wasn't.

I guess I don't get what's deceptive about that? AFAIK everything in those statements is true.

The way I see it, it's kind of like a meta- or para-placebo if you suggest beforehand that something can "produce significant mind-body self-healing processes" when the whole point of the experiment was to evaluate the presence of such processes, a sort of begging the question by suggestion. And this effect is all about suggestion and expectation, so framing matters. It's especially dangerous when coupled with the way they patients were self-selected by responding to an ad about a study relating to "mind-body healing" which tends to attract the kind of people that are open to woo in the first place (which was Orac's other point.)
posted by Rhomboid at 1:58 PM on December 23, 2010 [1 favorite]

It seems like the usual deception of the placebo was continued unchanged here. Patients weren't told that they were being given an inert, medically ineffectual substance. They were told they were being given a placebo with remarkable, clinically proven mind-body healing powers.

Good idea for a study, crappy execution.
posted by rusty at 1:58 PM on December 23, 2010 [2 favorites]

Orac deconstructs this paper.

It's a pretty damning read. Wait until you get to the end when you find out who funded the study.
posted by Blazecock Pileon at 2:00 PM on December 23, 2010 [1 favorite]

The placebos weren't actually sugar pills in this case.

The placebo pills were blue and maroon gelatin capsules filled with avicel, a common inert excipient for pharmaceuticals

They did hype the effectiveness of placebos to patients, while explaining that the effects were purely psychological.

"placebo pills made of an inert substance, like sugar pills, that have been shown in clinical studies to produce significant improvement in IBS symptoms through mind-body self-healing processes"

and in more detail

The provider clearly explained ... the following “four discussion points:” 1) the placebo effect is powerful, 2) the body can automatically respond to taking placebo pills like Pavlov's dogs who salivated when they heard a bell, 3) a positive attitude helps but is not necessary, and 4) taking the pills faithfully is critical.

... Our rationale had a positive framing with the aim of optimizing placebo response.

Patients given the placebo were compared with patients given no treatment (but the same doctor visits), not with patients told they might be getting an actual experimental drug with active ingredients.
posted by nangar at 2:07 PM on December 23, 2010

cjorgensen: “I take Cold-Eeze. I'm pretty sure they do nothing, but they seem to reduce the duration and severity of my colds. I'm pretty sure this is a placebo effect. I don't care.”

Just a small wonder: is that possible? I mean, if you're convinced it's a placebo effect, is it still a placebo effect? Maybe this is something I have to rtfa to get, but I've heard people say this, and it never makes sense to me. "I know it's just a placebo, but it helps me anyway." Doesn't that require you to both believe in the efficacy of the drug and not believe in it at the same time? Maybe that's possible, but it's sort of difficult, I would think. And if it is possible, and if that placebo effect is furthermore effective for you, then why not convince yourself that breathing air is the cure?

I suspect this is sort of like another thing I can't really understand: those people who set their clocks ahead ten or fifteen minutes so that they won't be late, even though they're always aware that the clock is ahead; they're basically counting on the fact that part of them will believe the lie they've told themselves. Maybe it's just my scatterbrained, bad-at-keeping-appointments self talking here, but I've always found that problems rarely have actual solutions that involve lying to myself.
posted by koeselitz at 2:21 PM on December 23, 2010 [3 favorites]

Just a small wonder: is that possible? I mean, if you're convinced it's a placebo effect, is it still a placebo effect? Maybe this is something I have to rtfa to get, but I've heard people say this, and it never makes sense to me. "I know it's just a placebo, but it helps me anyway." Doesn't that require you to both believe in the efficacy of the drug and not believe in it at the same time? Maybe that's possible, but it's sort of difficult, I would think. And if it is possible, and if that placebo effect is furthermore effective for you, then why not convince yourself that breathing air is the cure?

Right. I see this same problem in Pascal's wager. If you need to believe to go to heaven, can you really choose to really believe, just because it has better odds?
posted by crayz at 2:30 PM on December 23, 2010 [2 favorites]

Kirsch, in the interview: "We know from our research the things that make a difference: how much time you spend with a patient, how supportive and empathic you are, how well you listen, and how confident you are in being able to help."

Ben Goldacre basically makes the same points at the end of this short video about the placebo effect. It's like a quick summary of what he says on the subject in Bad Science.
posted by The Mouthchew at 2:34 PM on December 23, 2010

Wow, it's super important to read either Orac's take or the original paper, because the story being fed to the press does not even come close to what's being reported in the scientific article. Really, it's nothing less than outright deception.

Reminds me of the arsenic ordeal a little while ago. What a complete embarrassment. I'm losing a bit of respect for PLoS on this debacle.
posted by Llama-Lime at 2:40 PM on December 23, 2010

Er, only meant emphasize "super" there, not the whole comment
posted by Llama-Lime at 2:41 PM on December 23, 2010

If we could figure out how to harness the placebo effect, that would be amazing and wonderful. I heartily support more research in this direction. The more inert things we use, the better we avoid side effects.
posted by stoneweaver at 2:44 PM on December 23, 2010 [1 favorite]

My husband and I were discussing this paper this morning. I haven't read it yet, but unless I'm dramatically misreading the coverage of it, there are several holes large enough to chuck an entire peer review committee through. They chose IBS, which is a diagnosis of exclusion and is frequently exacerbated by stress, plus it has no objectively measurable component. The "informed placebo" was tested against no drug, but not against either a known effective drug OR an uninformed placebo. And lastly, the patients were told that they were being given a clinically effective treatment!

this study COULD have been done in an enlightening and meaningful way. There are measurable conditions that respond unusually well to placebo, like osteoarthritis. They could have simply told patients "this is a placebo, an inert pill that contains no medicine." I'd actually like to see a study where they had five groups: a known effective drug, a placebo that's hyped with all the woo talk, an un-hyped but informed placebo, a blind placebo, and no treatment at all. That might yield useful information! But they didn't, and so it's probably junk.

Gonna go read the original and Orac now.
posted by KathrynT at 2:53 PM on December 23, 2010 [2 favorites]

Am I missing something, or are many of Orac's objections essentially saying that the effectiveness of the placebo might have been due to the placebo effect?
posted by lore at 2:58 PM on December 23, 2010 [4 favorites]

Patients are told they're on a new experimental drug when in actual fact they have a 50% chance of just being on placebo. The point of this outcome is that that deception in double-blind trials may not be required - but good luck convincing the world that double-blind is no longer necessary.

No. Patients in a double blind study would be told that they have a 50% chance of being given a placebo instead of the experimental treatment. Deception arises when doctors give a patient a placebo, usually a drug they think will be harmless to the patient but not actually treat their condition, without telling them that it's a placebo. There's nothing in this study that suggests that double blind studies are unnecessary.

I'll go read Orac's article now.
posted by nangar at 3:05 PM on December 23, 2010 [1 favorite]

Maybe the power is in the symbolic, almost totemic quality of taking a pill? It's well known that we almost unconsciously trust men in white lab coats, maybe we now have the same reverence for pharmaceutical capsules.

Or perhaps we simply do better at recovery when we make some change to our daily routine, even if it's just swallowing a lump of sugar. Maybe it gets our brains primed for radical self-change, and makes us feel positive good-feelies about living up to our end of the sick role bargain.
posted by silentpundit at 3:05 PM on December 23, 2010

The way I see it, it's kind of like a meta- or para-placebo if you suggest beforehand that something can "produce significant mind-body self-healing processes" when the whole point of the experiment was to evaluate the presence of such processes, a sort of begging the question by suggestion. And this effect is all about suggestion and expectation, so framing matters. It's especially dangerous when coupled with the way they patients were self-selected by responding to an ad about a study relating to "mind-body healing" which tends to attract the kind of people that are open to woo in the first place (which was Orac's other point.)
posted by Rhomboid at 1:58 PM on December 23 [+] [!]

But, I still do not see how the researchers "deceived" the study subjects. Ben Goldacre himself uses words like "outrageous", "powerful", "amazing", etc. in describing the placebo effect in Mouthchew's video. He also uses the exact words "the amazing power of the mind over the body" so I don't see the problem with the study's 'significant mind body effect' language.

It after all literally is an effect on the body produced by the mind (in some way that we don't understand). How can this be "woo" if that is in fact is what actually takes place?

Maybe I'm just having a slow day. FWIW I agree that this is a case of bad study design and execution, it's not double-blind, as well as small effect sizes (i.e. what KathrynT said). So the study is probably false/worthless anyways. But I'm stuck on the "deception" language, I really don't get it. please hope me
posted by r_nebblesworthII at 3:08 PM on December 23, 2010

While I know that there's an intense bias toward skeptical "deconstructions" online -- which is worthy of several studies itself -- I did not find Orac's post to be nearly as withering as he apparently believes it is. If you read it carefully, it's so front-loaded with snark -- his harping on the term "mind-body," his use of colorful but irrelevant phrases like "woo-ful," his dissing of PLoS itself -- that by the time Orac insists that the trial volunteers have been "tainted" by the belief that the placebo effect is real, he's no longer deconstructing, but offbase. Since the trial was obviously not a test of the effectiveness of sugar pills against IBS, but a test of the physiological effects of the belief that one is receiving treatment for IBS, "tainting" the volunteers with that belief was crucial to the study design. As was the fact that the trial was open rather than blind, which Orac also raises as a critique. He brings up a valid point about demographics, but the authors of the study are quite clear that this is a small-sample trial of brief duration, and thus preliminary.

In this case, the "deconstruction" is as obviously biased as he accuses the study itself of being. But I know that most people often won't note that before linking to Orac's post as if it's definitive, because we all love skeptical deconstructions.
posted by digaman at 3:10 PM on December 23, 2010 [5 favorites]

Whenever people fall back on the catch-all term "woo" instead of articulating exactly what it is they object to, I immediately lose interest in anything else they have to say.
posted by hermitosis at 3:27 PM on December 23, 2010 [6 favorites]

Wait until you get to the end when you find out who funded the study.

Well, not to spoil the suspense or anything, but...

Cue "woo-ful" music...

The study was funded by National Center for Complementary and Alternative Medicine at the NIH and The Bernard Osher Foundation, which funds a number of great programs, including research into alternative medicine.

Who do you think is going to fund research into the placebo effect as a medical treatment -- Big Pharma? Pharmaceutical companies research the placebo effect because it makes their drugs look bad in clinical trials. They're not interested in promoting it as viable treatment.
posted by digaman at 3:32 PM on December 23, 2010

I'd actually like to see a study where they had five groups: a known effective drug, a placebo that's hyped with all the woo talk, an un-hyped but informed placebo, a blind placebo, and no treatment at all. That might yield useful information!

I would too.

Am I missing something, or are many of Orac's objections essentially saying that the effectiveness of the placebo might have been due to the placebo effect?

That's basically my reading. He's arguing that if you take away all the psychological effects there's no more placebo effect, and the authors did a lot to maximize the psychological effects. But placebo effects are understood to be psychological. (You wouldn't expect a placebo to actually kill viruses or anything.) He apparently thinks it deceptive to give people a placebo even if you tell them it's a placebo because it's psychological.
posted by nangar at 3:35 PM on December 23, 2010 [4 favorites]

Why is the fact that NCCAM funded this study a cause for doubt? NCCAM is not some industry shill. Maybe some folks don't realize that. NCCAM is heavily criticized by proponents of CAM, because all of their studies end up negative (duh).

I find it funny when doctors are so opposed to placebo. Every doctor that I've met pays at least lip service to one placebo treatment: a friendly doctor-patient relationship. Does the fact that the only outcomes a good relationship affects are subjective mean that somehow, doctors should refrain from it?

Subjective complaints are a huge part of medicine, and addressing them better is very important. Chronic pain, anyone? Many doctors recognize the role of placebo in treatment of subjective problems. Others try to hide the fact from themselves. Ever see anybody add 1mg dilaudid daily to a patient taking 90mg daily of morphine, plus vicodin prn? Pretending that the dilaudid is doing anything but acting as a placebo is disingenuous.

There may be something of a paradox here: if you tell people that placebos are effective treatment, they function as such; if they don't, maybe they don't function. Even if that's the case, telling people that placebos are effective isn't misleading at all.

It's true that this is a pilot study; it's true that there are a lot of unanswered questions, it's true that there was some sampling bias. But this study is enough for some doctors to feel okay prescribing placebo with full disclosure. That sidesteps the ethical problems that have stopped (really, just slowed, just hidden) placebo prescription in the past.

And I think that's cool. We've been ignoring this very effective treatment for too long.
posted by nathan v at 3:37 PM on December 23, 2010

Just a small wonder: is that possible? I mean, if you're convinced it's a placebo effect, is it still a placebo effect? Maybe this is something I have to rtfa to get, but I've heard people say this, and it never makes sense to me. "I know it's just a placebo, but it helps me anyway." Doesn't that require you to both believe in the efficacy of the drug and not believe in it at the same time? Maybe that's possible, but it's sort of difficult, I would think. And if it is possible, and if that placebo effect is furthermore effective for you, then why not convince yourself that breathing air is the cure?

Right. I see this same problem in Pascal's wager. If you need to believe to go to heaven, can you really choose to really believe, just because it has better odds?

Niels Bohr supposedly had a lucky horseshoe above his door. Someone said to him, "Niels, you don't believe in that mumbo-jumbo, do you." He replied, "Of course not, but I've been told it works, even if you don't believe in it."
posted by outlandishmarxist at 3:39 PM on December 23, 2010 [4 favorites]

NCCAM is a joke because they throw away massive amounts of public money studying horseshit like homeopathy in a futile effort to appease cranks. That's why you get the eye rolling from Orac at the end.
posted by Rhomboid at 3:40 PM on December 23, 2010 [11 favorites]

There is actually no reason to be all that skeptical about what the study is studying. The placebo effect itself is real enough that it is the very reason for double-blind studies. The question "can it be triggered with conscious knowledge that you're taking a placebo" is only crazy if you assume humans are rational actors who carefully and logically evaluate what is going on around us.

Both rituals like pill-taking and lying to yourself can be very powerful, because our minds are not at all as simple as we like to think they are. Rituals trigger powerful associations, which is why so many religions make use of them. There is a similarity there, too, in the fact that many of those religions (Christianity obviously being an exception) state that belief is not necessary and the ritual will be effective if you simply make a habit of doing it. If the ritual of taking a pill is powerfully associated with feeling better later, it could very well trigger the placebo effect even though we consciously know the pill is fake, because whatever mechanism makes rituals work isn't part of our conscious perception.

Lying to ourselves is also very powerful because some parts of the mind (and parts with particularly powerful control over the physical body as it happens) are quite primitive and gullible and unsophisticated, and if you (ritually, heh) repeat a phrase like "the doctor knows what he's doing" or "I trust the engineer who built this thing" enough, you will begin reacting as if those things are true even if your higher consciousness knows better.
posted by localroger at 3:48 PM on December 23, 2010 [2 favorites]

With all this new placebo effect research (I'm thinking about the "rising effectiveness of the placebo effect," in particular), I'm curious about one thing: are researchers adequately looking at lifestyle components that may give the placebo effect a leg up? One would think that if someone goes to a doctor to resolve a problem, then they're also trying things on their own. With IBS, for example, reducing dairy or meat might have a beneficial effect over a long period of time. If the Internet were enabling people to understand their illnesses better and thus "treat" them with lifestyle changes, it seems that that might help account for the rising placebo effect (assuming the study isn't just nonsense).
posted by outlandishmarxist at 3:49 PM on December 23, 2010

Finally! Evidence for the comforts of church.
posted by john wilkins at 3:58 PM on December 23, 2010 [1 favorite]

Rhomboid, do you think it makes your arguments more solid if you use outraged words like "horseshit" and refer to stuff like eye-rolling? I don't.

I also don't happen to believe in the explanations from homeopaths of how homeopathy works. But here's a recent controlled study that suggests that the benefits of homeopathic treatment are not derived from the pills at all, but from the lengthy consultations, which involve the same kinds of listening and caring attention that have been found to elicit placebo effects. Big Pharma certainly knows about that, which is why they go to such lengths to try to eliminate that kind of empathic behavior from trial staff.

That kind of analysis is much more interesting to me than more snark about homeopathic horseshit, even if many of the traditional rationales offered by homeopaths are, indeed, horseshit.
posted by digaman at 4:04 PM on December 23, 2010 [1 favorite]

Just a small wonder: is that possible? I mean, if you're convinced it's a placebo effect, is it still a placebo effect? Maybe this is something I have to rtfa to get, but I've heard people say this, and it never makes sense to me. "I know it's just a placebo, but it helps me anyway." Doesn't that require you to both believe in the efficacy of the drug and not believe in it at the same time?

Not necessarily. The placebo effect does "help anyway" -- it causes real, measurable improvement, not just the mistaken belief that one is improving. Despite the ouroboros-like nature of believing in something because it works because you believe in it, believing in the efficacy of something which is in fact effective isn't lying to yourself; it's just common sense.
posted by vorfeed at 4:38 PM on December 23, 2010

Great links and great discussion! Digaman's response to Orac seems good to me. I expect that future studies will bear out this research.

koeselitz: "I know it's just a placebo, but it helps me anyway." Doesn't that require you to both believe in the efficacy of the drug and not believe in it at the same time? Maybe that's possible, but it's sort of difficult, I would think.

I don't really see the paradox. When you take a drug marked 'placebo', you shouldn't believe in the efficacy of the drug qua pharmaceutical, but you should believe in the efficacy of the drug qua thing that will make you get better. And because of the placebo effect, believing that the drug will make you better will itself make you get better. You don't even have any false beliefs! There's something a little strange, on the face of it, that having a belief with a certain content causes the content of the sentence to be true, but this likely happens all the time. For example, it's probably true that in some cases, my belief that I want something causally sustains that desire.

If you think that the placebo effect exists, the result of this study is fact what you should expect. Which is why the phenomenon it purports to have found isn't that surprising to me.
posted by painquale at 4:50 PM on December 23, 2010

Doesn't that require you to both believe in the efficacy of the drug and not believe in it at the same time?

That's precisely the last question I ask researcher Irving Kirsch in the post behind the link.
posted by digaman at 5:13 PM on December 23, 2010

I completely agree that the world would be a much better place if physicians were encouraged to spend more time with patients, and to work on their bedside manner. But sadly I don't think that we will get there until we get single payer universal coverage so that we can eliminate the mountains of forms and coverage/claims jujitsu that a physician's practice has to deal with currently. I wouldn't be surprised if most physicians agreed that they'd rather be talking to patients at length instead of dealing with nonsense. But to lash that perfectly sensible idea on to the hobbled wagon of snake oil is something I have a problem with. A quack that convinces a patient suffering from a form of malignancy which is very receptive to treatment with surgery and adjuvants that they need only nutritional supplements, brown rice and organic vegetables, and maybe some reiki instead of the standard of care has blood on their hands no matter how compassionate and caring they are and no matter how many hours they spend bonding with the patient. And wasting precious public grant money on attempting to appease these people (and failing) is almost as bad.

By the way, 'Big Pharma' is just as much a loaded word as 'horeshit woo'.
posted by Rhomboid at 5:26 PM on December 23, 2010 [1 favorite]

Right, it's just that you're clearly not familiar with the growing body of peer-reviewed literature that demonstrates that the placebo response is not a "hobbled wagon of snake oil," but as real and testable in the lab as the immune response. I'd suggest starting with Fabrizio Benedetti's overview of the work in this area, Placebo Effects. It's excellent.
posted by digaman at 5:38 PM on December 23, 2010

Doesn't that require you to both believe in the efficacy of the drug and not believe in it at the same time?

That's precisely the last question I ask researcher Irving Kirsch in the post behind the link.

I thought that exchange was interesting. Kirsch goes on to say that you can believe in both the efficacy and the non-efficacy of the drug without contradicting yourself, and gives the sort of response that vorfeed and I do right above. But your claim that there might be "two layers of belief in the brain" was most interesting to me. It suggests that the placebo effect is not caused by an explicit belief in the efficacy of the drug, but something that happens at a "different level."

There are a few things this could mean. For one, it could be that there are subsystems in the brain that each have their own beliefs, like dual process theory posits, and so while the subsystem responsible for reasoning (system 2) leads me to consciously disbelieve that the doctor is giving me medicine, the subsystem responsible for automatic intuition (system 1) believes that he's giving me medicine. Perhaps the beliefs of system 1 generate the placebo effect, no matter what the beliefs of system 2 are.

Alternately, if you don't want to talk about the literal beliefs of subsystems: there are various types of belief-like states in the brain, and it could be that the placebo effect is caused not by believing that the therapy will heal you, but by entertaining some other sort of mental attitude towards the proposition that the therapy will heal you. For example, maybe imagining that the medicine will heal you is what causes the placebo effect. Or maybe it's a mental state for which we don't have a commonly used term. Maybe we alieve that we're going to get better (to use Gendler's term), and that's what causes the placebo effect.

So I think Kirsch is right to say that believing that taking a sugar pill will cure you does not entail believing a contradiction. But this is not to say that that the belief is what's necessarily causing the healing effect.
posted by painquale at 5:54 PM on December 23, 2010 [1 favorite]

I never said that the placebo effect wasn't very real and very testable, and can be responsible for real healing. However, if your modality of treatment is nothing more than kabuki theater and owes the entirety of its effect to the placebo effect (such as sham acupuncture being just as effective as 'real' acupuncture) then that is most definitely a hobbled wagon in my book.
posted by Rhomboid at 5:55 PM on December 23, 2010 [3 favorites]

I'm excited about straightforwardly using psychology to treat physical ailments.

I thought the problem with alternative medicine is that people are tricked into taking dangerously useless treatments that have no scientific backing. So if there are public research entities out there actually performing studies and introducing scientific rigor to the claims of alternative medicine then that is a good thing, right?
posted by Danila at 6:02 PM on December 23, 2010 [1 favorite]

If placebo is the active ingredient in homeopathy and acupuncture, and is just as powerful on its own, then that actually should discredit those approaches rather than strengthening them. Rigorous scientific testing of the effectiveness and limits of placebo treatment will give better guidance for when to use it, in contrast with the broad and dangerous effectiveness claims for "alternative" treatment.
posted by parudox at 6:02 PM on December 23, 2010

An aside, but there's some interesting speculation that many cases of IBS (at least the diarrheal version) may be explained by a dysfunctional gallbladder, and easily treated... I know, no double-blind study, no RCT, but maybe there's something to it.
posted by greatgefilte at 6:15 PM on December 23, 2010

NCCAM is a joke because they throw away massive amounts of public money studying horseshit like homeopathy in a futile effort to appease cranks. That's why you get the eye rolling from Orac at the end.

NCCAM has a specific mandate, that involves the testing of a lot of things without any plausible mechanism. I agree that much NCCAM funding would be better spent elsewhere, although I believe that there are likewise much worse ways to spend that money.

Saying that NCCAM is a joke is a little misleading, because despite the questionable subject, NCCAM has proven that they are capable of rigorous, negative studies.

In this case, NCCAM has found a way, within their mandate, of investigating a plausible and potentially useful hypothesis. So, given that, exactly how is this source of funding relevant to this study?

However, if your modality of treatment is nothing more than kabuki theater and owes the entirety of its effect to the placebo effect (such as sham acupuncture being just as effective as 'real' acupuncture) then that is most definitely a hobbled wagon in my book.

A great deal of medicine relies on theater. White coats, speaking with confidence, and even, some would argue, pharmacological treatment of many ailments.

It's true that this theater does very little for, say, mortality rates.

But it's not an either/or proposition. People with diseases don't just suffer from the threat of death or disability; they suffer pain, nausea, and anxiety, and in many cases, pharmacological treatment has proven inadequate at managing these problems.

Placebo and non-placebo treatment are not inconsistent. In the cases of many diseases, placebo is about as good as it gets, or is at least responsible for the largest part of response to the few/only medications that have been shown (sometimes spuriously) to be of any use.

This isn't about denying people life-saving treatment. This is about offering treatment to people who suffer chronic debilitating illnesses that we are unable to treat effectively in other ways: IBS, fibromyalgia, depression, anxiety, chronic back pain, chronic abdominal pain, chronic anything pain.
posted by nathan v at 6:35 PM on December 23, 2010

Rhaomi: if your modality of treatment is nothing more than kabuki theater and owes the entirety of its effect to the placebo effect (such as sham acupuncture being just as effective as 'real' acupuncture) then that is most definitely a hobbled wagon in my book.

What if the treatment strengthens the placebo effect or makes one especially susceptible to it? Hypnotherapy arguably works this way. Many talk therapies could be considered to work this way.

My own favored pet theory (which very might well be discredited by people in this thread; if so, then great, I can move on) is that antidepressants work predominantly by the placebo effect. This is not to say that SSRIs have no neurochemical effects... they obviously do. And the chemical effects are therapeutic. But the role the chemical effects play is in aiding the placebo effect. After taking SSRIs, you end up with bodily changes that you're forced to interpret in some way. Because these bodily changes are the result of a medicine, you interpret them as salutary, and the placebo effect is strengthened. Plain old sugar pills don't provide as convincing an illusion of healing. (And it should be said that SSRIs aren't really that much more effective than placebos.) Moreover, perhaps by altering attention, etc., the specific chemical changes it puts one in make one especially susceptible to suggestion. Whatever mechanism leads to the placebo effect is aided by the chemical effects.

Maybe this isn't the only therapeutic effect of antidepressants, but it could be one of them. If people have evidence against the hypothesis, I'd really like to know. Given that I've never really seen it touted, there's probably something wrong with it.

Anyway, just suppose that's how antidepressants work. Would that be reason to think that they're a "hobbled wagon"? I'd think not. Now, I really doubt that acupuncture and homeopathy aid the placebo effect more so than any arbitrary treatment. And that's certainly a point against them. But I don't think that a therapy should be discredited if "the active ingredient" is placebo.
posted by painquale at 7:03 PM on December 23, 2010 [2 favorites]

Thank you rhomboid for the Orac link, I was waiting for his take on this. Also to greatgefilte -- from what I've seen most people who report IBS also fit the "4F" profile for gallstones.
posted by ltracey at 7:31 PM on December 23, 2010

In this case, the "deconstruction" is as obviously biased as he accuses the study itself of being. But I know that most people often won't note that before linking to Orac's post as if it's definitive, because we all love skeptical deconstructions.

Orac's post may be a bit snarky in places, and he may well be biased, but that doesn't diminish the strength of his critique.

I think the most devastating point is the one that notes that study participants "were recruited from advertisements for "a novel mind-body management study of IBS" in newspapers and fliers and from referrals from healthcare professionals. During the telephone screening, potential enrollees were told that participants would receive "either placebo (inert) pills, which were like sugar pills which had been shown to have self-healing properties" or no-treatment.""

The importance of this detail is that the study selected for people who might be prone to wanting to believe that 'mind-body' medicine is effective - which would mean that they would be more prone to report that the placebo 'worked' even if their subjective experience was no different. I can't understand why the researchers would advertise in this way except in order to produce this bias - why not just advertise for "a study for a novel management of IBS."

The point here is that the researchers contaminated the placebo effect (the impact of expectations on individuals' experience of symptoms) with a biased study design (the impact of expectations on individuals' reporting of symptoms). If the research selects for people who want to participate in a "novel mind-body" treatment, and then tells them that the placebos "produce significant mind-body self-healing processes," then it runs a serious risk of encouraging people to report more symptom relief with the placebos.

Again, the problem here is not that the study produces actual placebo effects - it is that the study design encourages people to report placebo effects, and has no mechanism for determining whether or not they are doing so because of the study itself.
posted by googly at 8:29 PM on December 23, 2010 [2 favorites]

My cousin was recently a participant in a second-stage clinical trial for an immunological therapy. "Second stage" in this case meant the treatment was already deemed efficatious, and the purpose of the trial was to determine optimum dosage. The structure of control and experimental groups was explained to her in detail, and she knew exactly what her chances of receiving placebo were. Her participation ended up providing an ambiguous, perhaps negligible result.

I wonder if there isn't a reverse placebo effect: suspecting you might be receiving one reduces the efficacy of real medication .
posted by clarknova at 8:40 PM on December 23, 2010

My own favored pet theory is that antidepressants work predominantly by ... the role the chemical effects play is in aiding the placebo effect. After taking SSRIs, you end up with bodily changes that you're forced to interpret in some way. Because these bodily changes are the result of a medicine, you interpret them as salutary, and the placebo effect is strengthened.

A great way to test this would be with the corollary: administer them under the rubric of "unhappiness pills". The recipients are informed that they're being given depression-inducers as a punishment.

The ability to conduct experiments like this is just another bullet point in a long list of good arguments for political gulags.
posted by clarknova at 9:00 PM on December 23, 2010 [1 favorite]

My own favored pet theory (which very might well be discredited by people in this thread; if so, then great, I can move on) is that antidepressants work predominantly by the placebo effect.

That's also the pet theory of Irving Kirsch, the subject of the interview behind the link. In fact, we talk about that in the interview, and he writes about it in his book The Emperor's New Drugs.
posted by digaman at 9:06 PM on December 23, 2010

However, if your modality of treatment is nothing more than kabuki theater and owes the entirety of its effect to the placebo effect (such as sham acupuncture being just as effective as 'real' acupuncture) then that is most definitely a hobbled wagon in my book.

Rhomboid, not to put too fine a point on it, but you should really read Benedetti's book -- or a study like this, this, or this. You'll understand the nature of placebo effects a lot better than you do now.
posted by digaman at 9:16 PM on December 23, 2010

What is this . . . I don't even . . .
posted by IvoShandor at 9:26 PM on December 23, 2010

Googly's critique seems fair.
posted by digaman at 9:27 PM on December 23, 2010

That's also the pet theory of Irving Kirsch, the subject of the interview behind the link. In fact, we talk about that in the interview, and he writes about it in his book The Emperor's New Drugs.

Does he also think that the chemical action of antidepressants makes one especially susceptible to placebos? I haven't read his book, but was under the impression from your interview that he thinks that the chemical features of antidepressants don't do anything of clinical worth. (He says that nearly all drugs for depression are of equal benefit.) I was trying to suggest that the drug action of antidepressants is clinically important insofar as it can strengthen the placebo effect.
posted by painquale at 9:56 PM on December 23, 2010

Googly's critique seems fair.

To be fair, its my extrapolation on Orac's observation.

I should add that I have seen Kirsch speak and read his work and been very impressed with his critiques of the role of the placebo effect, especially in SSRI clinical trials. That's why I'm somewhat surprised that his name is on this study - it seems to suffer from design flaws that he would (rightly) eviscerate in trials conducted by pharmaceutical companies. I was really hoping to see a well-designed study that demonstrated how effective the placebo effect can be, and I believe that it will eventually happen, but this isn't quite it yet.
posted by googly at 10:17 PM on December 23, 2010

Googly, there are lots of studies on the placebo effect. Get Benedetti's book, or search for him, Wager, or Zubieta in PubMed. Really interesting work, done on a shoestring because very few people want to fund this. It wasn't until pharmaceutical companies started losing a lot of money to inexplicably high placebo effects in RCTs that they started sponsoring trials by these guys.

But this trial wasn't a test of general efficacy. It was specifically a test of "open-label" placebo effectiveness, which creates the challenges you describe. The authors don't claim to be doing more than offering a template to be tampered with.
posted by digaman at 10:46 PM on December 23, 2010

Well, having read the article now, and the study, I have some problems concluding that the study really demonstrates anything at all. The main difficulty is in the choice of effect studied: IBS. IBS is well-known to be a shifting and difficult-to-pin-down malady. I'm aware that this is a subject that is sometimes controversial, but many respected and thoughtful physicians would probably judge that IBS is largely a psychological problem. I take haste to point out that it's nonetheless a very real problem, but the indications that it's largely psychological are strong. painquale mentions SSRIs above; ironically, from what I understand SSRIs are often found to be the most effective treatment for IBS.

This is interesting in itself; I think IBS provides an intriguing example of how psychology can affect the body. But that fact in itself makes IBS sort of a horrid choice for a study on the placebo effect, given that IBS is taken by many to itself have a unique relationship with placebos and with psychological maladies and cures. The abstract makes this sound as though it was measured in a very rigorous and careful way that was clear-cut and direct; but anybody who's ever suffered from IBS will tell you that there is really no easy way to diagnose and measure it, and each case is very different.

But thinking about it now, I guess maybe that was the point; they wanted to choose a malady that was highly susceptible to the placebo effect to make it easier to isolate. However, I don't think it's that easy with IBS. It's too fraught with psychological implications already.
posted by koeselitz at 12:20 AM on December 24, 2010 [1 favorite]

My own favored pet theory (which very might well be discredited by people in this thread; if so, then great, I can move on) is that antidepressants work predominantly by the placebo effect.

If this were so, different people would not respond differently to different antidepressants.

(Is this all part of the effort to pretend the brain isn't a biochemical consciousness machine and that consciousness is the product of some vaguely-defined spirit instead of chemistry? Because the whole "putting psychoactive chems in your brain doesn't really do anything, it's all theater!" doesn't make any sense unless you're one of those stupid assholes.)
posted by Pope Guilty at 1:20 AM on December 24, 2010 [1 favorite]

But thinking about it now, I guess maybe that was the point; they wanted to choose a malady that was highly susceptible to the placebo effect to make it easier to isolate. However, I don't think it's that easy with IBS. It's too fraught with psychological implications already.

I think they chose a malady which is extremely susceptible to placebo to make it more likely that they'd get the results they wanted all along, just like how they wrote their ad to attract people who would tell them what they wanted to hear and told people what effect they expected to hear reported before dividing the subjects into the group of people who knew they were getting nothing and the group of people who were selected and primed to report improvement.

This is, when you look at it, a textbook case of how shitty the quality of alt med research is.
posted by Pope Guilty at 1:22 AM on December 24, 2010 [1 favorite]

"I know it's just a placebo, but it helps me anyway." Doesn't that require you to both believe in the efficacy of the drug and not believe in it at the same time?

I strongly "believe" in the effectiveness of placebos because their effectiveness has been demonstrated very rigorously and robustly. There is no reasonable scientific doubt that placebos are effective for some medical problems.

So if a doctor were to give me a pill and tell me that it's a placebo and that placebos are an effective treatment for my sickness, I would expect that to trigger the placebo effect for me (whatever that is and however it works) just as effectively as if he gave me a sugar pill and told me it was a powerful antibiotic or a powerful antidepressent. Because I believe in placebos just as much as and for the same reasons as I believe in antibiotics and antidepressents.
posted by straight at 1:47 AM on December 24, 2010

There is no reasonable scientific doubt that placebos are effective for some medical problems.

So if a doctor were to give me a pill and tell me that it's a placebo and that placebos are an effective treatment for my sickness, I would expect that to trigger the placebo effect for me (whatever that is and however it works) just as effectively as if he gave me a sugar pill and told me it was a powerful antibiotic or a powerful antidepressent. Because I believe in placebos just as much as and for the same reasons as I believe in antibiotics and antidepressents.

Statements like this are why some days I think everybody who isn't a scientist should be kept far, far away from information about science.
posted by Pope Guilty at 2:24 AM on December 24, 2010 [1 favorite]

painquale, the theory you're describing for SSRIs has a name - "active placebo." For example, patients in a trial might be given niacin, which causes flushing, but not necessarily other bodily effects. In that situation, they feel somatic effects and it makes them more likely to look for relief of their symptoms, because "something is happening."

One of the problems with the placebo effect is that it is temporary. It fades out after a couple of months, and has to be re-energized by some sort of novel approach. Have you ever met someone who used not one, but many forms of alternative medicine in succession? Or who cycles through them over the course of a year? I know people like this.

The placebo effect is also limited in scope. The types of mental and physical problems which can be (temporarily) relieved by the placebo effect do not represent that wide a spectrum of ailments. If you have a disorder consisting solely of problems with pain, pain perception, or stress, great! But CAM advocates consistently overreach in their claims about what their treatments can physically affect, and patients can suffer because of that.

Can a placebo reduce swelling, measured by external criteria and not just how it feels? Can placebos thin mucus? Can they increase white blood cells? Give us SOMETHING that isn't just based on patient self-reporting!
posted by overeducated_alligator at 5:22 AM on December 24, 2010 [1 favorite]

koeselitz, what would you say if the study authors had chosen depression? "Well, it's a psychological malady, really subjective..." Or anxiety? Or chronic pain?

I'm afraid placebo researchers are doomed to disappoint you, because these are the maladies for which the placebo response provides relief. All these disorders are right at that nexus of subjectivity and biology. There is not, and never will be, a placebo antibiotic or placebo chemotherapy. So IBS is actually an ideal test of the placebo effect.
posted by digaman at 7:12 AM on December 24, 2010 [1 favorite]

But CAM advocates consistently overreach in their claims about what their treatments can physically affect, and patients can suffer because of that.

Yes, that's true -- as bad as what the pharmaceutical companies do with mood drugs like SSRIs, except that the side effects of antidepressants are much more severe than the side effects of acupuncture. But that doesn't mean that the placebo effect shouldn't be studied. In fact, once enough studies are done that people can start distinguishing between different types of placebo effects, we'll be in a better position to know what various treatments really can and can't do.
posted by digaman at 7:16 AM on December 24, 2010

Can a placebo reduce swelling, measured by external criteria and not just how it feels? Can placebos thin mucus? Can they increase white blood cells? Give us SOMETHING that isn't just based on patient self-reporting!

Well, you could always try reading a few studies. For example, the first study I linked to in one of my replies to Rhomboid above measured this via positron emissions topography:

Placebo treatment affected endogenous opioid activity in a number of predicted μ-opioid receptor-rich regions that play central roles in pain and affect, including periaqueductal gray and nearby dorsal raphe and nucleus cuneiformis, amygdala, orbitofrontal cortex, insula, rostral anterior cingulate, and lateral prefrontal cortex.

Doesn't sound like a questionnaire to me.
posted by digaman at 7:22 AM on December 24, 2010

And, what do you know, a completely independent group of researchers in Germany today just announced a new study that found that the placebo effect changes signaling activity in the dorsal horn region of the spine. So, that "something" beyond patient self-reporting is just a couple of clicks away.

How these phenomena play out in the real world of disease and treatment is what we need to know. That's one of the things that the Kirsch/Kaptchuk study is a first step towards.
posted by digaman at 7:27 AM on December 24, 2010

Or would be, if it weren't terribly designed and biased at every step.
posted by Pope Guilty at 8:03 AM on December 24, 2010

Specific critiques, like googly's = productive. Vague snark = zzzzzzz.
posted by digaman at 8:44 AM on December 24, 2010

One of the problems with the placebo effect is that it is temporary. It fades out after a couple of months, and has to be re-energized by some sort of novel approach.

That doesn't jive with anything I've ever read.

One of my favoritest studies was the 2001 Moseley study where they compared arthroscopic knee surgery with sham surgery. For the sham, they put the patient under, made a small incision in the knee, and placed a band-aid. Two years later, people from the surgery and sham group both reported increased level of function and decreased pain. (Surgery is an especially powerful placebo for modern US Americans, it appears.)

If this were so, different people would not respond differently to different antidepressants.

That's like saying that if neither homeopathy nor accupuncture was an effective physiological treatment for pain that people wouldn't respond differently to those treatments.

People respond differently to blue sugar pills than they do to red sugar pills. People respond differently to drugs that make them nauseated than to drugs that give them a headache. People get better and attribute their improvement to whatever interventions were performed recently.

(Is this all part of the effort to pretend the brain isn't a biochemical consciousness machine and that consciousness is the product of some vaguely-defined spirit instead of chemistry? Because the whole "putting psychoactive chems in your brain doesn't really do anything, it's all theater!" doesn't make any sense unless you're one of those stupid assholes.)

No, because the whole "theater doesn't really do anything" idea isn't consistent with claims that the brain is a biochemical consciousness machine. (And c'mon, that's needlessly aggressive.)

I think they chose a malady which is extremely susceptible to placebo to make it more likely that they'd get the results they wanted all along, just like how they wrote their ad to attract people who would tell them what they wanted to hear and told people what effect they expected to hear reported before dividing the subjects into the group of people who knew they were getting nothing and the group of people who were selected and primed to report improvement.

That is, at the very least, uncharitable.

Many people have long believed that the placebo effect cannot be elicited without deception-- that's the entire basis of ethical arguments against placebo use. This study provides one counterexample to that. (If you want to argue that their description of the placebo was deceptive, that might be interesting, but it's really just about definitions, and the words they used have more than one each.) Sampling bias is a valid concern, but even so, it suggests that for at least one population, placebo can be evoked while making placebo status explicit. That is an interesting finding.

Now, there's lots of room for exploring the boundaries. What about other populations? What about different forms of placebo? What about placebo in the absence of a nurturing therapeutic environment? What about placebo sans any statement of efficacy, or when saying, "This pill is worthless, here you go"? Those are all interesting questions that I hope to find explored in future studies.

Statements like this are why some days I think everybody who isn't a scientist should be kept far, far away from information about science.

I'm not sure what you're referring to-- straight's claim that placebos are doubtlessly effective for some medical problems? Do you not consider pain to be a medical problem?

So, that "something" beyond patient self-reporting is just a couple of clicks away.

Diga, I assume you're familiar with the study where they found secretly administered naloxone blocked placebo pain relief? Can't find the link now. Cool stuff.
posted by nathan v at 9:49 AM on December 24, 2010 [3 favorites]

Absolutely, nathan. And thanks for your excellent post.

The naloxone studies really launched the field of placebo research, outside of drug trials. Really interesting stuff.
posted by digaman at 10:05 AM on December 24, 2010

PG, I'm puzzled by some of the positions you've taken here.

> Is this all part of the effort to pretend the brain isn't a biochemical consciousness machine and that consciousness is the product of some vaguely-defined spirit instead of chemistry? Because the whole "putting psychoactive chems in your brain doesn't really do anything, it's all theater!"

I don't think claiming that antidepressants are ineffective requires a belief in dualism, or a belief that psychoactive drugs have no effects on mental states. painquale's speculation was that antidepressants are psychoactive, and a person taking them can feel their effects, but that they might not directly affect depression. These effects might just help convince the patient that the drug is doing something, and enhance it's placebo effect if the drug is presented as an antidepressant.

I don't think that either painquale or Kirsch doubt that caffeine is a stimulant or that alcohol gets you drunk. I doubt if Kirsch as skeptical of the effectiveness other psychoactive drugs that have more solid empirical support as he is about the effectiveness of SSRI's as antidepressants.

I don't have a position on the effectiveness of antidepressants because I haven't read the research on them. It's just that your response to this suggestion seemed really odd. It's odd to insist that it's scientific to believe that claims for the effectiveness of a psychoactive drug must be true just because it's claimed to be psychoactive, and that it's religious or superstitious to be skeptical about a drug's psychoactive effect just because the claimed effects are psychological.

> I think they chose a malady which is extremely susceptible to placebo to make it more likely that they'd get the results they wanted all along ...

Would this procedure be invalid if this was a test of pharmacological treatment rather than a psychological one? Is wrong to test a drug's effectiveness for a condition you believe it's likely to effective for? Is it wrong to test it in a population that you believe believe may respond well to the treatment? It seems like your problem is more with the fact that the researchers were studying the placebo effect than because of the study design. I agree with what Kathryn T said above, that it would have been better if the study included more conditions. And as the authors themselves pointed out, it's unknown how effective the placebo would have been if they had recruited more skeptical volunteers.

What I'd like to see follow a study like this are studies asking, if people with IBS can get some relief of symptoms with just a chance to talk about their symptoms with a sympathetic doctor a few times, a daily pill taking ritual, and a pep talk about the pill taking ritual, how much better would they do with a more aggressive psychological intervention, say, daily or frequent relaxation or stress reduction exercises - not a rejection of any research into psychological treatment of psychosomatic (or psychological) disorders.
posted by nangar at 11:27 AM on December 24, 2010

One other thing.

> If this were so, different people would not respond differently to different antidepressants.

Given the nature of depression, this isn't really a valid argument in favor of the effectiveness of antidepressants. Depression is not usually a chronic condition. It usually happens in response to life events that lead to hopelessness and despair. People usually get over depression eventually (assuming they don't kill themselves before then), especially if the situation causing the depression improves or as it recedes further in time. If antidepressants are in fact a placebo treatment for depression, some people might find them ineffective, go back to their doctor who then tries a different medication, and repeat the process until they eventually recover on their own, in which case the last drug they tried would be credited with "curing" the depression.

(Once again, I'm not arguing that SSRI's are placebo treatments for depression. Since there have been a lot of studies of antidepressants, and the results have been mixed, it would take a lot of work to come with an idea of what's going on with them - or even for me just to have an opinion. I'm just pointing out that this argument in favor of their effectiveness isn't valid on its face.)
posted by nangar at 12:02 PM on December 24, 2010

Good points, nangar.

> there have been a lot of studies of antidepressants, and the results have been mixed

Particularly the ones we never saw, because they were buried by pharmaceutical companies after the control-group volunteers exhibited an "unexpectedly high placebo response." (Once you start noticing that phrase in pharma-news stories, you'll see it everywhere, making one wonder when it starts to become expected). Those were the studies that Irving Kirsch saw, which led him to write his book.
posted by digaman at 12:08 PM on December 24, 2010 [1 favorite]

Thanks for the enlightening discussion, guys. And thanks to overeducated_alligator for cluing me in to active placebos.

(Is this all part of the effort to pretend the brain isn't a biochemical consciousness machine and that consciousness is the product of some vaguely-defined spirit instead of chemistry? Because the whole "putting psychoactive chems in your brain doesn't really do anything, it's all theater!" doesn't make any sense unless you're one of those stupid assholes.)

No, I am a physicalist. Don't be rude.
posted by painquale at 2:10 PM on December 24, 2010

Pope Guilty, I'm hoping you just misunderstood my point because I wrote it so sloppily.

I wasn't trying to make such sweeping assertions about the scientific consensus on placebos and I didn't mean to imply that, for example, placebos were effective replacements for antibiotics.

What I meant was that I'm convinced that there is strong scientific evidence that "placebo-type" treatments (the word "placebo" should probably technically be reserved for the total statistical noise you get in the control arm of a double-blind research study) are effective for certain things, most notably pain and depression.

And my point was that, since I'm convinced that sugar pills are often effective for pain treatment, I would hypothesize that someone like me would be just as likely to get pain relief from a sugar pill knowing it was a sugar pill as from a sugar pill thinking it might be a new experimental pain drug.

So I was arguing that whatever the "placebo effect" is could be triggered by a belief in the effecacy of placebos in the same way that it is triggered by a belief in medicines.
posted by straight at 3:50 PM on December 24, 2010

A helpful new overview of placebo/brain science by researcher Fabrizio Benedetti:

http://www.nature.com/npp/journal/v36/n1/full/npp201081a.html
posted by digaman at 8:37 AM on December 28, 2010

Cheating Death - Placebo Effect, Immortality & Wild Lynx
posted by homunculus at 12:37 PM on January 7, 2011