Placebos

August 25, 2009 9:08 AM

Placebos Are Getting More Effective. Drugmakers Are Desperate to Know Why.
posted by homunculus (105 comments total)

Utterly fascinating. Thanks.
posted by phrontist at 9:14 AM on August 25, 2009

Also, this epsiode of RadioLab is pertinent, and one of the better examples of an all around great show.
posted by phrontist at 9:17 AM on August 25, 2009 [5 favorites]

Easy. Because in our culture, people are being trained to trust drugs more and more.
posted by rokusan at 9:18 AM on August 25, 2009 [6 favorites]

Because in our culture, people are being trained to trust drugs more and more.

Obecalp.
posted by TheophileEscargot at 9:22 AM on August 25, 2009 [1 favorite]

It turns out that the only thing that's actually infinite is the potential of mankind for credulity.
posted by Copronymus at 9:24 AM on August 25, 2009

Perhaps the need for drug manufacturers to continually discover new drugs in order to keep stock prices up has left them scraping the bottom of the barrel, and these innovative new ways of exploiting brain chemistry via medication aren't any better at it than the brain itself.
posted by waxboy at 9:26 AM on August 25, 2009 [4 favorites]

Can't wait to get my 2010 UHC formulary where it shows the copay for sugar pills is $40.
posted by birdherder at 9:28 AM on August 25, 2009 [1 favorite]

Thanks for the link to my story, homunculus. I really appreciate it.

I agree with rokusan that advertising is training us to trust drugs more, and there's a section of my article that focuses on that aspect of the placebo effect, though there are more elements in play here. But one of the quotes we didn't have room for was from a doctor who defended direct-to-consumer ads by saying that if they were suddenly banned, drugs would become less effective because they would lose their "placebo boost." I find that fascinating.
posted by digaman at 9:28 AM on August 25, 2009 [32 favorites]

rokusan: "Because in our culture, people are being trained to trust drugs more and more."

That explains the curative effects of placebos, but it doesn't explain why placebos would outperform active drugs. I mean, if you're not in the placebo group of a study, you should enjoy the psychological benefit of expecting to see improvement, plus the pharmacological benefit of the drug in question.

If it's a good drug, it should still do better than placebo... except I guess if the drug happens to inhibit the placebo effect itself, but how likely is that to be the case for all of these different drugs mentioned in the article?
posted by Riki tiki at 9:30 AM on August 25, 2009

Huh, weird. But, yeah, I was thinking the same thing as waxboy--seems like a lot of the stuff they're targeting now (depression, anxiety, maybe even IBS) ought to be more responsive to mental state than, say, measles.
posted by equalpants at 9:35 AM on August 25, 2009 [1 favorite]

The part that I find most interesting is that placebo effect is now a widely known effect by complete laymen, and yet this hasn't dampened its effect at all..
posted by shownomercy at 9:35 AM on August 25, 2009 [4 favorites]

Erosion of skepticism about the effectiveness of pills over the last 20 years. Plus, people who volunteer want some form of "praise" - they want to be the exceptional result.
posted by PuppyCat at 9:47 AM on August 25, 2009

That explains the curative effects of placebos, but it doesn't explain why placebos would outperform active drugs.

There's an awful lot of unstated assumptions in the pairing of those two clauses. If "trust" is sufficient to explain the curative effects of placebos then presumably you're saying that we have the intrinsic ability to cure ourselves if suitably persuaded. If you take that as your starting assumption then you haven't stated any limits on this intrinsic power, so you're not in any position to be surprised by how much more or less effective it would be than some unspecified set of introduced "active" chemicals of varying expected efficacy.
posted by George_Spiggott at 9:47 AM on August 25, 2009 [1 favorite]

phrontist, thanks for the link to that great RadioLab episode about the placebo effect. It was one of the inspirations for my article. I love RadioLab!
posted by digaman at 9:50 AM on August 25, 2009

"To remain dominant in the future," he told Forbes, "we need to dominate the central nervous system."

what
posted by jquinby at 9:50 AM on August 25, 2009 [7 favorites]

Thanks for the link to my story, homunculus. I really appreciate it.

My pleasure. It's a great piece, as always.
posted by homunculus at 9:52 AM on August 25, 2009

If only Michael Jackson's doctor had read about this! Michael would only think he was dead!
posted by It's Raining Florence Henderson at 9:54 AM on August 25, 2009

His plan hinged on the success of an experimental antidepressant codenamed MK-869. Still in clinical trials, it looked like every pharma executive's dream: a new kind of medication that exploited brain chemistry in innovative ways to promote feelings of well-being. The drug tested brilliantly early on, with minimal side effects, and Merck touted its game-changing potential at a meeting of 300 securities analysts.

WE BEAT PENICILLIN!
posted by FatherDagon at 9:54 AM on August 25, 2009 [5 favorites]

Placebos are getting better because we're increasingly treating made-up diseases.
posted by GuyZero at 10:01 AM on August 25, 2009 [28 favorites]

Poor ole pharmaceutical industry. They worked so hard to convince that pills would save the day and didn't think to make sure that it was the active components doing the saving rather than just the idea of medication. Now what are they gonna do?
posted by solipsophistocracy at 10:05 AM on August 25, 2009

GuyZero hit on it: What is real and fake here?
posted by kuatto at 10:07 AM on August 25, 2009

That explains the curative effects of placebos, but it doesn't explain why placebos would outperform active drugs.

I dunno either but I can vouch that they do. Busted my wrist pretty badly a few years back, was laid up in the hospital on a morphine drip. It ran out, a nurse came & fussed with the machine. I thought she swapped in a new bottle but it turned out she hadn't. Eventually another nurse came along & restarted the drip, but until then the morphine my brain was making was actually a good bit stronger then the dosage the hospital was giving me. Now if only I could figure out how to make some on demand...
posted by scalefree at 10:07 AM on August 25, 2009 [2 favorites]

So... What happens if you place placebo vs placebo... FIGHT!

would one placebo perform better than the other? You would think they would be equal within parameters, but... I'd like to see this test performed.
posted by symbioid at 10:10 AM on August 25, 2009

(Symbioid, the article addresses that - the color of the placebo pill matters!)

Thanks for posting this, it's fascinating.
posted by Salamandrous at 10:11 AM on August 25, 2009

The article mentions that different shapes and colors of placebos are better for different conditions, and that patients' reactions to these differences are largely culturally determined, so I bet there's already been a placebo-off or two.
posted by solipsophistocracy at 10:12 AM on August 25, 2009

That's a fascinating article, thanks. The placebo effect is a fascinating phenomenon that needs to be better understood by researchers and the general public. The basic ideas have been common sense for decades; my mother is a primary school teacher and for decades has been using a "magic wave" or "special plasters" (just a different colour) to make kids' scrapes and knocks less painful. It's astonishing to see how much difference these seem to make to the kids' perception of their pain. It reminds me of the position I recently saw a homeopath taking in a debate: if homeopathy does nothing but prime people for a strong placebo effect, surely it's still a good thing?*

The sidebar about the drugs' appearance was interesting too. Some years ago I went to some lectures about developing drugs for psychiatric patients. One of the biggest problems when running placebo-controlled trials is that the placebo group don't experience any of the weird side effects they've learned to expect from drugs, so they don't believe that the drug has any potency, reducing the placebo effects. Research into this area is ethically tricky -- try writing an application to deliberately cause strong side effects in your control group -- but I remember reading that placebos that taste bad were more powerful than ones with no flavour.

*My answer: only if the homeopath has enough medical training to know what can sensibly be treated by placebo alone, and enough persuasive power to make genuinely sick patients go to a real doctor instead of another quack.

Potter tapped an IT geek named David DeBrota to help him comb through the Lilly database of published and unpublished trials—including those that the company had kept secret because of high placebo response.

This is the kind of behaviour that earns big drug companies their terrible reputations. Especially in medical research, it's inexcusable to perform experiments and hide any data that might help patients at the expense of the company's profitability. If evidence suggests that a widely-used drug doesn't perform as well as is widely believed, I can't see any justification for keeping that a secret and thus keeping patients on a treatment of doubtful utility. Even if the drug really is potent and the placebo effect is increasing, that's still medically important information that deserves investigation. The people in management positions at these companies must have hearts of stone, or have evolved deeply weird moral systems.

digaman - But one of the quotes we didn't have room for was from a doctor who defended direct-to-consumer ads by saying that if they were suddenly banned, drugs would become less effective because they would lose their "placebo boost." I find that fascinating.

Presumably that's testable: Has anyone ever compared the "placebo boost" in the USA with the same drugs in the UK where direct-to-patient marketing has never been allowed?
posted by metaBugs at 10:17 AM on August 25, 2009 [2 favorites]

LOLPHARMA!11!@!!
posted by Mental Wimp at 10:17 AM on August 25, 2009

Unstated is how hard the Imperial Sugar Company R&D department is working to maintain that edge.
posted by BrotherCaine at 10:19 AM on August 25, 2009 [7 favorites]

Yeah, something about this article doesn't add up. It repeatedly misstates the nature of the placebo effect, failing to distinguish between "placebo" (a fake medicine) with "placebo effect" (arguably a misleading term, since the effect isn't exclusive to placebos).

In a double-blind trial, the sugar pill and the real medication are both subject to the placebo effect. The placebo effect isn't "people respond to fake meds"; it's "people respond when they believe they're taking meds". That applies equally whether the belief is correct or incorrect.

If the sugar pill starts outperforming the trial medication, it can't mean that the placebo effect is getting stronger—if that were the case, both the sugar pills and the meds would start performing better, since both are subject to the effect.

Instead, here are some of the things it could mean (pick one or both):

• The real meds are actually starting to worsen the condition they're supposed to treat.
• Sugar pills are starting to become effective as medication (i.e., they are developing efficacy beyond the placebo effect).

Neither of these seems a likely answer (though the first one is probably more likely), but they are the obvious starting points for any attempt to explain the observed phenomenon—and the article seems to utterly miss this point.

Fascinating topic; somewhat sloppy coverage.
posted by ixohoxi at 10:21 AM on August 25, 2009 [8 favorites]

digaman, thanks for a great article.

...drugs would become less effective because they would lose their "placebo boost."

Fascinating indeed. Essentially, a large part of the effectiveness of new drugs is due to the placebo effect that, like waxboy said, gets proportionately larger as the actual drugs get more esoteric and subtle. Seems to me that pharmaceutical companies could gain more by increasing their marketing budgets since the increases in placebo really outpace all the new ideas out of R&D.

This, combined with the recent article on why test for prostate cancer cause more harm than good, explains why we think that our own bodies have so much perceived capacity for self-healing: In this system, the medicine is more harmful than the disease.
posted by infinitefloatingbrains at 10:21 AM on August 25, 2009 [2 favorites]

Mostly placebos are good for psychotropic purposes. Pain relief, minor depressive symptoms, etc. Placebos for stage IV lung cancer are pretty worthless. The beauty of finally forcing pharma to do rigorous trials against placebo is that we keep most useless but expensive drugs off the market. (But hey, if I test 20 useless drugs with a type I error rate of 5%, I expect one of them to succeed, so they still have that going for them.) As the article points out, some of the more lucrative older anti-depressants never underwent rigorous controlled trials (Said with wide-eyed concern trolling: "We can't refuse them treatment; that would be unethical!")

The example of improved placebo effect is in anti-depressants, a condition notoriously susceptible to placebo. Why, even just talking about their condition improves symptoms in some folks.
posted by Mental Wimp at 10:25 AM on August 25, 2009 [1 favorite]

If it's a good drug, it should still do better than placebo...
posted by Riki tiki

Nope. Placebos don't fool you into thinking you're better. Placebos fool the brain/body into producing its own drugs, which in many cases are better drugs.
That is why Naloxone not only prevents opiate drugs from working, but also prevents placebos from working.
posted by weapons-grade pandemonium at 10:27 AM on August 25, 2009 [2 favorites]

Ixohohi, you seemed to skim over these crucial sentences, which precisely address your point:

[Beecher] demonstrated that trial volunteers who got real medication were also subject to placebo effects; the act of taking a pill was itself somehow therapeutic, boosting the curative power of the medicine. Only by subtracting the improvement in a placebo control group could the actual value of the drug be calculated.

And, in fact, metaanalyses of antidepressant trials indicate that, in many cases, the response in both drug groups and placebo groups is rising. That's just what one would expect, as Beecher pointed out. But the response in placebo groups is rising higher and outstripping the drug performance in many trials, as I described.

Also, the section of my article on "therapeutic ritual" -- and, in fact, many other statements in the piece -- make it clear that it's the medical context that invests sugar pills with apparently therapeutic potency. I'm certainly not saying that sugar pills are getting "stronger" on their own.
posted by digaman at 10:32 AM on August 25, 2009 [5 favorites]

If it's a good drug, it should still do better than placebo...

Nope. Placebos don't fool you into thinking you're better. Placebos fool the brain/body into producing its own drugs, which in many cases are better drugs.

False dichotomy. They can do both.

...combined with the recent article on why test for prostate cancer cause more harm than good...

Hey, I'm a coauthor on the US paper!
posted by Mental Wimp at 10:34 AM on August 25, 2009 [1 favorite]

Problem: Trials show that your treatment doesn't work any better than the placebo.
Solution: "Trials show that our treatment works! And hey, the placebo works too!"
posted by shammack at 10:35 AM on August 25, 2009 [1 favorite]

Georg Groddeck, in his 1926 Berlin lecture on "Treatment" quoted his mentor, Ernst Schweninger, as follows: "The majority of all illnesses cure themselves, no matter how they are treated or whether they are treated at all; if I (Groddeck) am not mistaken, he said 75 per cent, a figure I consider too low ... A further number, perhaps 15 per cent will never improve, no matter what treatment is given. Then there is the remaining 10 per cent for whom the kind of treatment they get really matters. -- It is of no importance whether these figures are right or wrong, the fact remains that treatment rarely decides whether the patient recovers or remains ill." From "The Meaning of Illness" page 223.
posted by RichardS at 10:43 AM on August 25, 2009 [3 favorites]

17% more fairy dust in today's placebo, compared with the formulations of yore.
posted by Mister_A at 10:45 AM on August 25, 2009

ixohoxi - If the sugar pill starts outperforming the trial medication, it can't mean that the placebo effect is getting stronger—if that were the case, both the sugar pills and the meds would start performing better, since both are subject to the effect.

That's true if placebos are actually outperforming the medication, but not if they're merely closing the gap. If a drug has a fairly small effect relative to the (potential) placebo effect, the drugs efficacy could easily be masked.

For example, a table of some made-up numbers showing percentage improvement of symptoms in response to (1) placebo alone or (2) drug+placebo.

Trial..............................Placebo only.................Placebo + Drug
First Trial...........................10%...............................20%
Second Trial.......................80%...............................90%

In the first trial, placebo+drug is twice as good as placebo alone, so the drug seems potent. In the later trial the drug is just as potent, but the strong placebo effect masks it. You'd think this would be very easy to spot by looking at the increased efficacy of both treatment conditions. However, it's hard to do without comparing the groups to a third group (completely untreated) which is very hard to justify ethically unless you're specifically looking for the effect. Also, it's different to compare studies from decades ago with modern studies thanks to changing diagnosis and assessment criteria or techniques, differences in other aspects of medical care, diet, etc.

Prozac was mentioned as an example where placebos were rapidly catching up with the drug, at least in the US. While depression is a disease that often has physiological roots (or so I've read; I'm not an expert by any means), it's one that I'd expect to be highly malleable by a person's moods and expectations. Diagnosis and assessment of depression have also changed radically over the past few years/decades. So I'd expect it to be a perfect candidate for the effect I've suggested above.
posted by metaBugs at 10:45 AM on August 25, 2009 [2 favorites]

weapons-grade pandemonium: "Nope. Placebos don't fool you into thinking you're better. Placebos fool the brain/body into producing its own drugs, which in many cases are better drugs."

But unless you're saying that's an intrinsic property of milk sugar, there's no reason why drugs with active ingredients wouldn't also do that unless they inhibit opiate production or something, right?

I'm not, you know, "smart", but I feel like I'm missing something here. The placebo effect should, at the maximum, give the placebo group similar benefits when compared to the test group. It shouldn't matter if (for sociological reasons or whatever) the placebo effect has become more effective over time other than to increase/decrease the margin between the two groups.
posted by Riki tiki at 10:50 AM on August 25, 2009

My table would be clearer with extra columns. Sorry:

Trial......................Drug Only........Placebo only..........Drug + Placebo........Efficacy of drug over placebo
First Trial....................10%...............10%......................20%.............................100%
Second Trial................10%...............80%......................90%.............................12.5%
posted by metaBugs at 10:51 AM on August 25, 2009

Placebos are getting better because we're increasingly treating made-up diseases.

I don't think that's true in the way you meant it, but I favorited it anyway.
posted by Sova at 10:52 AM on August 25, 2009

It is of no importance whether these figures are right or wrong, the fact remains that treatment rarely decides whether the patient recovers or remains ill." From "The Meaning of Illness" page 223.

Interesting but put that in perspective -- it was written in 1926. Read Lewis Thomas' The Youngest Science, which is pretty much all about the fact that until well into the 20th century, doctors couldn't actually treat diseases worth a damn because they didn't have any drugs that worked. (Pardon the over-paraphrasing.)
posted by George_Spiggott at 10:54 AM on August 25, 2009 [1 favorite]

Placebos are getting better because we're increasingly treating made-up diseases.

I'm curious to find out what these made-up diseases people talk about.
posted by P.o.B. at 10:55 AM on August 25, 2009

But unless you're saying that's an intrinsic property of milk sugar, there's no reason why drugs with active ingredients wouldn't also do that unless they inhibit opiate production or something, right?

That's correct -- and drugs with active ingredients are also subject to the placebo effect, which was one of the points of Beecher's paper, referenced in the article.

metaBugs, this is right:

If a drug has a fairly small effect relative to the (potential) placebo effect, the drug's efficacy could easily be masked.

In fact, drug developers call the problem I describe in the article "the crisis of assay sensitivity" -- i.e., how to design trials so they're more sensitive to the "signal" of the active drug's efficacy beyond the placebo effect?
posted by digaman at 10:57 AM on August 25, 2009

RLS.
posted by Mister_A at 10:58 AM on August 25, 2009

My guess is that the studies where placebos outpace actual drugs are predominately psychological disorders which of course are real problems but not exactly diseases in the same way, say, MRSAs are. Without any proof or research I feel fairly confident to state that placebos don't outpace real drugs in the treatment of MRSA staph infections. yes, that was redundant, I know
posted by GuyZero at 11:00 AM on August 25, 2009

GuyZero, while mood disorders are easy to dismiss that way (except for people who have them or treat them), this problem is also coming up in trials of drugs for chronic pain, Parkinson's disease, high blood pressure, and other diseases. It's one thing to say that a disorder is merely "psychological," and another to say that there is frontal cortex involvement, as I say in the piece.
posted by digaman at 11:08 AM on August 25, 2009 [1 favorite]

Nice piece there, digaman. I will be sharing it with my colleagues.
posted by Mister_A at 11:13 AM on August 25, 2009

To follow on from from shammack's half-joking point, the efficacy of placebos is indeed being used to tout treatments that would otherwise be dismissed by medical scientists, particularly in the fuzzy end of "alternative medicine". Recent studies of acupuncture, for example — not double-blinded, but comparing professionals acupuncture needle placement for back pain vs. random placement vs. pressing toothpick points against the patient's skin in random locations as a placebo (all on the lower back, so patients were not aware which treatment they were receiving) — showed no difference between placebo and traditional acupuncture. The institute doing the study then turned the results around, essentially claiming that acupuncture as placebo or with real needles worked, and therefore acupuncture worked.

To broaden the context somewhat, this is essentially the age-old moral problem of religious adherence, psychic readings, homeopathy, etc — usually summarised as if it makes them feel better, what's the harm?.

There is no question that horoscopes, aura cleansing and the rest can make many people feel better: it is the definition of the placebo effect, along with the salutary effect of being supported by a community of like-minded people. Medical science, at least for the last 100 years or so, has not been about "feeling better", per se, but being medically effective in a measurable, repeatable way. As humans, we can be fooled into (or fool ourselves) into "feeling better" any number of ways. But actually curing a disease is a different problem.
posted by Bora Horza Gobuchul at 11:15 AM on August 25, 2009 [2 favorites]

Hopefully you're still here, digaman, as I wanted to ask you if you were familiar with Pristiq, which recently won FDA approval. The higher tested dose of the drug was less effective than the lower dose, and failed to beat placebo; despite that, the lower (50 mg) dose was approved for marketing in the US by FDA. That trial, and that approval, have bugged the hell out of me; just wondering if you had come across any nuggets you could share on the topic.
posted by Mister_A at 11:18 AM on August 25, 2009

digaman, I really don't mean any disrespect except perhaps to drug companies who seem to be trying to fix everything with a drug. I'm perfectly willing to go with the view that NSAIDs are a miracle and are the exception in that a single drug works to cure a variety of problems rather than the rule. And it's hardly disrespectful to state that depression and a staph infection are vastly different types of issues where one is addressed better by drugs than the other.
posted by GuyZero at 11:24 AM on August 25, 2009

I figured someone would throw that out there. If you have ever talked with someone who suffers from severe Restless Leg Syndrome, you wouldn't scoff it off so easily. Trust me, it's not something that is pleasant or mild.
posted by P.o.B. at 11:26 AM on August 25, 2009 [1 favorite]

ixohoxi: If the sugar pill starts outperforming the trial medication...

See, that's a very big if. Nowhere in the article does it say that this happened. What it does say, though, is that in up to as much as 70% of cases for some drugs, they failed to outperform the placebos. So yeah, if they're both merely subject to the same placebo effect, they'll come out the same. And that's a loss for the real drug, since it ought to pull out ahead if it were actually effective. Or that's the logic anyway, the article (implicitly) poses some interesting questions to that...
posted by Dysk at 11:29 AM on August 25, 2009

digaman/metaBugs: thanks for the clarifying remarks. Perhaps I misunderstood the article's claims; I'm running on insufficient sleep today. I'll reread the article (and your comments) later—I'm interested in the topic.
posted by ixohoxi at 11:29 AM on August 25, 2009

> But one of the quotes we didn't have room for was from a doctor who defended direct-to-consumer ads by saying that if they were suddenly banned, drugs would become less effective because they would lose their "placebo boost." I find that fascinating.

Yeah, extremely so.

Too bad it got cut-- the unconscious reliance on the placebo effect, even for Tested and Proven Methods, could bear much more investigation. The question of How does Product X compare against Nothing is in some ways less interesting than the question of how, exactly, does "Nothing" work so well, and so often.

Obviously, there is some sort of underlying mechanism for the Nothing of belief-based healing... and it would be nice to be able to systematically exploit that mechanism.

Also, where are the two competing, celebrity-endorsed, heavily-publicized brands, Placebex and Placebor, for which I've been waiting so long?
posted by darth_tedious at 11:29 AM on August 25, 2009 [1 favorite]

Nice post, Bora. Am not familiar with Pristiq, Mister_A, but that's utterly fascinating, thanks.

For sure, GuyZero. My post may have sounded more critical than I meant it. I totally appreciate what you're saying.
posted by digaman at 11:30 AM on August 25, 2009

I've often wondered about the relationship between side effects and placebo.

Instead of giving placebo, they might give something with some mild but real effect (e.g. dry mouth). The "nearcebo" does nothing else, but would the presence of some noticeable physical change increase the power of the placebo effect? If the real medication has a dry mouth side effect, how do they know how much of the effect is due to what they really think the drug is doing? Would reducing a drug's side effects reduce the "nearcebo effect" making it seem to do worse against normal placebo?

But can it cut the other way? If your test group is more cynical or fussy, perhaps they might think things like, "I hate this stupid dry mouth this drug sucks it's not working." If they have the belief that not only should a drug cure their condition but it shouldn't make their legs cramp up either, a pure placebo effect might do better than effective drug + placebo-cancelling side effect.

For that matter, placebo vs. X where they tell the X group, "This is just a placebo," how does placebo do vs. the drug as X?
posted by fleacircus at 11:33 AM on August 25, 2009

"Milk sugar"? So, uh, lactose? Hm. If I was given lactose pills, I'd feel definite "effects," which may lead me to believe the pills are "working," and kick the placebo effect into, erm, effect. Lactose intolerance is not exactly rare.

One reason placebo might seem to be doing "better" now than in years past is that the pharmaceutical industry has recently received some flak for, it turns out, publishing only the favourable trial results and binning the others; maybe the new results are the same as the old results, but we're now getting to see all of them?
posted by Sys Rq at 11:35 AM on August 25, 2009 [3 favorites]

(Not "sure" if I "used" enough "quotes" in that "comment.")
posted by Sys Rq at 11:37 AM on August 25, 2009 [1 favorite]

The "nearcebo" does nothing else, but would the presence of some noticeable physical change increase the power of the placebo effect?

Yes, fleacircus. These are known as "active placebos." Antidepressants, for example, fail more often in trials when tested against active placebos -- which is very interesting.

SysRq, that's an interesting notion, but note that Potter in my story had access to all the data, including the never-published trials, years ago, and was disturbed by the apparent rise even then.
posted by digaman at 11:43 AM on August 25, 2009

... the morphine my brain was making was actually a good bit stronger then the dosage the hospital was giving me. Now if only I could figure out how to make some on demand...

I've always thought that faith healing was bunk, but if you believe something strongly enough, I don't think it would matter if that belief was in the Magical All-Better Pill or the Magical Unseen Diety's Blessing.
posted by filthy light thief at 11:47 AM on August 25, 2009

Hey, I'm a coauthor on the US paper!
posted by Mental Wimp

Time for a new user name.

(And congrats.)
posted by rokusan at 12:11 PM on August 25, 2009

Yeah, there really are some smart bastards here.
posted by Mister_A at 12:12 PM on August 25, 2009

"Ten years and billions of R&D dollars after William Potter first sounded the alarm about the placebo effect, his message has finally gotten through. In the spring, Potter, who is now a VP at Merck, helped rev up a massive data-gathering effort called the Placebo Response Drug Trials Survey."
While it is certainly depressing that it took so long this makes me very happy as it would certainly have been easier and likely more profitable for Big Pharma to simply downplay the placebo response while enjoying its benefits. It is also a perfect illustration of the collision between a desire for profit and good medicine.

digaman: "... But one of the quotes we didn't have room for ...."
Would someone please fix this? Compensate the advertisers somehow so that online articles don't suffer dead tree limitations?

Clinical studies have shown reading Metafilter increases libido as well as sexual satisfaction in 100% of respondent's surveyed over a 5 year period.
posted by vapidave at 12:13 PM on August 25, 2009

Why does every odd person seem to be lactose intolerant lately? What, is it all caused by vaccines, or plastic water bottles, or microwaves from the CIA, or what?

I mean, seriously... half the people I meet lately. Where was this problem 30 years ago?

(And what's the deal with airplane food, yeah....)
posted by rokusan at 12:14 PM on August 25, 2009

So you have a problem with odd people???
posted by Mister_A at 12:15 PM on August 25, 2009

placebo:  control for psychosomatic responses
nearcebo: placebo with dry mouth
arecibo:  searches heavens for alien communication

posted by ...possums at 12:15 PM on August 25, 2009 [2 favorites]

I think a lot of people are being convinced they have depression when what they really have is a case of the blues or perhaps they are not taking very good care of themselves. Perhaps the placebo effect is enough to lift these folks out of the dumps. But according to the aggregate study of SSRIs, they do have a noticable response over placebo for those who are suffering from severe clinical depression.

Also, I work in a Neuro clinic where Neurontin is prescribed to nearly every patient who walks through the door. I'm convinced it is the world's most expensive sugar pill.
posted by Acromion at 12:21 PM on August 25, 2009

This is a bit off topic rambling, but articles like the on in the FPP, and this one on a Stanford Med School study on tai chi makes me think that there's a whole realm of internal body chemistry that we have the potential voluntarily control. Some day my quest for real prana-bind training + internal organic-chemical control will come true and I will dominate this world. How can you stop a man who feels no pain?
posted by Mister Cheese at 12:44 PM on August 25, 2009 [2 favorites]

I feel like I'm missing something here. The placebo effect should, at the maximum, give the placebo group similar benefits when compared to the test group. It shouldn't matter if (for sociological reasons or whatever) the placebo effect has become more effective over time other than to increase/decrease the margin between the two groups.
posted by Riki tiki

I see what you're saying, Riki. IANAD, but I know the body has ways of shutting down if presented with too much of a good thing. Hyperventilation, for example, starves the body of oxygen because your oxygen regulators say, "Whoa, there's too much oxygen here, let's shut things down." That's why the antidote is to breathe into a paper bag. So maybe real opiates prevent the body's own endorphins from kicking in, whereas placebos don't.
posted by weapons-grade pandemonium at 12:56 PM on August 25, 2009

I see many of the points I was trying to get at were actually already covered, or strongly implied, in the article itself.

I consider myself reminded to RTFA.

Good article, digaman.

> How can you stop a man who feels no pain?

An insufficiently lucrative promotional offer might do the trick.
posted by darth_tedious at 12:59 PM on August 25, 2009

Why does every odd person seem to be lactose intolerant lately? What, is it all caused by vaccines, or plastic water bottles, or microwaves from the CIA, or what?

IDK about everyone else, but mine started out mild (and presumably genetic), and was exacerbated to its current terrifyingly apocalyptic level by repeated bouts of amoebic dysentery. <>
posted by elizardbits at 1:01 PM on August 25, 2009

Fascinating article.

It makes sense to me (as a layperson) that if stress can cause or aggravate as many conditions as it's purported to (such as high blood pressure, incontinence, impotence, decreased immune response, insomnia, any condition caused by over- or under-weight, etc.), that neurochemical processes ought to be able to be manipulated to ameliorate or alleviate the condition without resorting to pharmaceuticals. Placebos strike me as one piece of that puzzle. Maybe something has shifted culturally (possibly related to advertising, as digaman posits) to increase our collective faith in pharmaceuticals as a means to wellness.
posted by notashroom at 1:13 PM on August 25, 2009

"Milk sugar"? So, uh, lactose?

I'm not sure why the term "sugar pill" stuck as the term for placebo. Most pills contain a lot of inactives so I think the strategy there is usually more of the same.

When you're trying to do a clinical study the big deal is that you don't want the patients or the clinicians to be able to tell the two apart other than the lot number on the container (that whole double blind thing.)

This is easy when your product is a colorless solution - just use your formulation buffer. When your product is slightly colored, or when the doctor is reconstitution a lyophilized cake it can be a real bitch since you don't want to just stick some random extra crap in there to make up the difference or give it a tint. There's apparently quite an art to it.
posted by Kid Charlemagne at 1:16 PM on August 25, 2009

One thing I'm curious about is that drug ads are pretty much just a US thing - is this happening in other countries as well where there are no drug ads?
posted by GuyZero at 1:18 PM on August 25, 2009

By definition, inert pills have no effect, but under the right conditions they can act as a catalyst for what he calls the body's "endogenous health care system."

Does this imply that the body has reserves of substances used for immunity and healing which it deploys in larger volume via the placebo effect? Wouldn't this mean in some cases that patients who've recently received placebo are more vulnerable to subsequent infection or trauma than those who've received an equivalent dose of real drugs?
posted by Anything at 1:26 PM on August 25, 2009

Damn you Wired for your unfailing production of fascinating articles!

Seriously, fascinating post.
posted by Lutoslawski at 1:31 PM on August 25, 2009

"I'm curious to find out what these made-up diseases people talk about."

Stuff like Blivv and Snerdds.

I blame the Jedi Mind trick.
posted by Smedleyman at 1:54 PM on August 25, 2009

It's like this P.o.B., infectious disease and anything that would respond to your doctor's advice about loosing weight and getting more exercise (which you then cheerfully ignore) is real. Everything else is a made-up disease that only the weak complain about.

Unless you belong to a religion that says reality isn't real. Then you're on your own.
posted by Kid Charlemagne at 2:05 PM on August 25, 2009 [1 favorite]

Interesting question, Anything - I have no idea!
posted by digaman at 3:05 PM on August 25, 2009

It works in reverse too: Pointing the Bone.

We might in time develop a coherent theory of placebo, psychosomatic illness and cure, anosognosia, "psychic surgery", hypnotism, and related matters. It's clear that the mind is capable of exerting influence over the body, the question is exactly how much influence and under what conditions.
posted by aeschenkarnos at 3:05 PM on August 25, 2009

But if these same drugs were vetted now, the FDA might not approve some of them.

That sounds ludicrous. If a drug that was approved because of its efficacy no longer demonstrates that efficacy, shouldn't its FDA certification be removed/adjusted?
posted by mrgrimm at 3:16 PM on August 25, 2009

Trial..............................Placebo only.................Placebo + Drug
First Trial...........................10%...............................20%
Second Trial.......................80%...............................90%

In the first trial, placebo+drug is twice as good as placebo alone, so the drug seems potent. In the later trial the drug is just as potent, but the strong placebo effect masks it.

Actually, metaBugs, the simplest statistical test for this comparison is the binomial test, and it would give exactly the same result for both comparisons, because it is symmetric. Think of it this way: in the first trial you test the difference in the success rate; in the second, the difference in the failure rate. In both cases it is then comparing 10% to 20%.
posted by Mental Wimp at 3:23 PM on August 25, 2009 [1 favorite]

Yes, it does sound ludicrous, mrgrimm. Perhaps you could send a letter to the major pharmaceutical companies, or the FDA, suggesting that every major drug on the market be put through more multimillion-dollar clinical trials every couple of years to ensure that they're still effective. :)
posted by digaman at 3:31 PM on August 25, 2009

A fascinating article, but the thing that struck me most was that (according to the author) the placebo response, as a unique and verifiable effect, went virtually unstudied for nearly sixty years. Beyond its capability to moderate and test the efficacy of drugs, it's an amazing phenomenon.
posted by lekvar at 3:34 PM on August 25, 2009

Why does every odd person seem to be lactose intolerant lately?

That's just what happens when you join the Internet Dairy Fart Support Forum.
posted by Sys Rq at 3:44 PM on August 25, 2009

Oh, and digaman? Nice work.
posted by Mental Wimp at 3:46 PM on August 25, 2009

Perhaps you could send a letter to the major pharmaceutical companies, or the FDA, suggesting that every major drug on the market be put through more multimillion-dollar clinical trials every couple of years to ensure that they're still effective.

Make the drug companies pay for it. Who's paying to test Prozac vs. placebos now?
posted by mrgrimm at 3:47 PM on August 25, 2009

Nthing all those who said "fascinating!"
posted by futureisunwritten at 4:12 PM on August 25, 2009

"Why does every odd person seem to be lactose intolerant lately?"

Yeah, that's weird. I drank a big glass of milk every night before bed. Ate it with my Cherrios, etc. One day, boom, milk makes me sick. I've tried other kinds of milk. Amish, organic, whatever, still makes me queasy. Like someone flipped a switch.
posted by Smedleyman at 4:18 PM on August 25, 2009

"Having trouble concentrating? Can't understand the placebo effect? Challenged by an uncertain future in the pharmaceuticals industry? Try new Evifectium! Ask your doctor if Evifectium is right for you."
posted by chairface at 5:01 PM on August 25, 2009

Make the drug companies pay for it. Who's paying to test Prozac vs. placebos now?

Nobody.
posted by Mister_A at 6:29 PM on August 25, 2009

I can never read this stuff without immediately thinking of the equally fascinating "nocebo" effect. The flip side of the coin.

Most fascinating of all, however, is the other powerful force, mentioned at the end of that link. To wit, people tend to get/feel better themselves, without any (real or imagined) medication at all. Yet another variable to worry about!
posted by smoke at 6:52 PM on August 25, 2009

Oh and also, goddamn how much does this stuff remind me of that opening scene from Do Androids Dream where they are arguing about what emotions to get the machine to give them that day.
posted by smoke at 6:53 PM on August 25, 2009

"The quality of care that placebo patients get in trials is far superior to the best insurance you get in America," says psychiatrist Arif Khan, principal investigator in hundreds of trials for companies like Pfizer and Bristol-Myers Squibb. "It's basically luxury care."

Really interesting, and probably the most compelling explanation: with managed care, health care for the average patient who is not in a clinical trial has been getting worse and worse in the last decade. Add to this the dismissive attitude, demonstrated upthread, toward poorly understood conditions like restless leg syndrome, and it's not hard to see why getting individual care from a doctor who takes the patient seriously could lessen stress that might have exacerbated his or her symptoms.
posted by transona5 at 7:10 PM on August 25, 2009

I read a book about this a while back, "The Placebo Response" by Brody. It discusses a theme along the same lines as this article: that variations in the response of the control group play a big role in the success or failure of a medical trial. Then the second half of the book was "practical advice" about "unleashing this power in yourself," which I think seemed too lame to read. But first half was great.

Then I read a sheaf of blank paper, which was even more great.
posted by fantabulous timewaster at 7:31 PM on August 25, 2009

Smoke: Here's the original draft of my "nocebo" paragraph, which has more evil goodness:

"Benedetti has also probed the power of the placebo's evil twin, the nocebo. (The word placebo comes from the Latin for 'I shall please,' and nocebo means 'I shall harm.') Mild nocebo effects can be triggered by something as trivial as a doctor saying, 'Just try to breathe normally.' But there's nothing trivial about nocebo. Parkinson's patients receiving deep-brain stimulation for motor disorders suffer significant losses of movement when they believe the device has been switched off, even if it remains on. Men given a popular drug for prostate cancer who are informed that the drug may cause sexual dysfunction are twice as likely to become impotent. Chinese-American men born in a year traditionally associated with a disease like cancer die faster than whites if they get the disease -- but only if they believe in the virtues of Chinese astrology."
posted by digaman at 8:06 PM on August 25, 2009

I've always thought that faith healing was bunk, but if you believe something strongly enough, I don't think it would matter if that belief was in the Magical All-Better Pill or the Magical Unseen Diety's Blessing.

Uh, well, that might be true for things like pain because pain is a signal from your brain mediated by chemical activity. It sure wouldn't work so well for a lot of things though. All the belief in the world isn't going to fix your compound fracture, your sucking chest wound, or your stage IV pancreatic cancer. (Okay, all the medicine in the world probably won't fix the last one either but one could substitute other conditions as preferred).

the fact remains that treatment rarely decides whether the patient recovers or remains ill." From "The Meaning of Illness" page 223.

Utterly false in the modern world for many illnesses and diseases. And the more serious a disease, the more likely medical intervention is to make a difference. Up to a point, of course, since some things remain incurable. But saying that most people get better on their own from a cold or the flu is not a criticism of modern medicine.

I drank a big glass of milk every night before bed. Ate it with my Cherrios, etc. One day, boom, milk makes me sick. I've tried other kinds of milk. Amish, organic, whatever, still makes me queasy. Like someone flipped a switch.

Most people are lactose intolerant their whole lives. It just so happens that the Metafilter demographic is skewed towards that segment of the world population that tends to not start out lactose intolerant. But even among those of us who start out able to produce lactase age can lead to an increasing intolerance. It's common.

I'm like you: I could consume all the lactose in the world with no problem. Then one day I ate most of a pint of Haagen Dazs and a couple hours later was curled up on the bathroom floor with the worst cramps of my life. It wasn't some sort of contaminant either because the same thing happened next time I tried eating a ton of ice cream. So now I eat much less ice cream at a time and no problem.
posted by Justinian at 8:15 PM on August 25, 2009

holy flying shitballs Digaman, that's even worse than I thought!
posted by smoke at 8:26 PM on August 25, 2009

Very interesting. One thing I wonder about is how pharmacogenomics and other forms of personalized medicine will change the placebo/active drug response rate. If you can find certain biological markers (genetic or otherwise) that can predict who will respond to a certain medication and who won't, there's the potential for greatly increasing the the success rate of the drugs in question. We're still some way from getting to that point, but it's neat to think about.

As an aside, there's someone interesting work out there trying to correlate EEGs with antidepressant efficacy. The studies so far have been pretty small and published in some crummy journals (The Journal of American Physicians and Surgeons hardly deserves the appelation), and a neurologist I spoke to thought it was "quackery," but there seem to be kernels of good stuff in there that could potentially make a huge difference if their present clinical trial is positive.
posted by greatgefilte at 9:27 PM on August 25, 2009

I also think one of the goals of the secret pharma project I describe in the piece, the Placebo Response Drug Trials Survey, is to search for biomarkers for high placebo response, so those people can be excluded from trials.
posted by digaman at 9:42 PM on August 25, 2009

Great topic, great article, great post, great commentary, great thread. The best of Metafilter. Sidebar-worthy. I lost count of the number of fascinating facts I just learned.

For that matter, placebo vs. X where they tell the X group, "This is just a placebo," how does placebo do vs. the drug as X?

A neat and bizarre fact is that even if you tell a placebo group that they're just taking placebo sugar pills, they still show the placebo effect! So you couldn't just tell the X group that they're taking placebos; you'd need to secretly dose them. (I always thought that this would never fly past review boards, but Benedetti's open/hidden trials discussed in the penultimate paragraph of digaman's article seem designed to get around this problem.)

But why would the placebo effect seem to be getting stronger worldwide? Part of the answer may be found in the drug industry's own success in marketing its products.

Another possible factor (and a possible reason that people show the placebo effect even when told that they're taking placebos) is that people are perhaps now more aware of the power of the placebo effect. The more that you believe in the placebo effect, the higher your expectations of being cured, and hence, perhaps, the stronger the effect. The effect bootstraps itself.
posted by painquale at 1:12 AM on August 26, 2009

Indeed, painquale, thanks for that!
posted by digaman at 7:22 AM on August 26, 2009

More new data about the physiological basis of placebo pain relief.
posted by digaman at 9:44 AM on August 26, 2009

Placebo and the Situation of Healing
posted by homunculus at 8:41 AM on September 19, 2009