So this is how the healing powers of the saints and gods of the pre-modern age worked.

It also seems like it might be difficult to implement. How would a doctor write a prescription for a placebo? They would have to keep it secret, while having euphimisms or fake names for it widely enough that pharmacies know what to do.

Google "Study Casts Doubt on the Placebo Effect" or read up on it on wikipedia. It's been debunked. It only kinda works for pain. Which ain't nothing. But it's not magic.

Also, that depression study has been debunked. The real result of it was that drugs work about as well as placebo in people who aren't severely depressed. That doesn't mean placebos work, it just means the drugs aren't effective in folks with less severe depression.

What Peter Orszag doesn't understand is that if you're going to give patients a placebo, you don't tell them that you're going to give them a placebo. You don't do that. You don't say that out loud.

Don't placebos work partly because people think they're getting a real treatment? If people know that they're getting a sugar pill, doesn't that diminish the effectiveness? So if it becomes policy to give, say, the 50% of depression patients who are mildly afflicted a placebo, won't the mildly afflicted know or at least suspect that they're getting a placebo and won't the placebo effect then not work? I don't see how you could keep this new placebo-based treatment policy a secret on a system-wide basis.

And aren't placebos cheaper than real medicines? Won't they notice that that $400 pill is now $50 and won't that tip them off that they're getting something bogus? Or are they getting charged the same price? I suppose if you have a single payer picking up all the tabs and the patient never really knows what their treatment cost, it doesn't matter.

I don't get how this is supposed to work.

Exercising and keeping a daily journal are each about as effective in relieving minor depression as Prozac.

Developing ways to measure people's pain sensitivity would help - i.e. don't stint on the good stuff for people who are debilitated by pain, and give hardier folk older, cheaper meds.

This is another way electronic medical records can help. My doctors use them and the history of what works for me is in there. Skin irritation - start at medium. Pain levels - low is fine. If I'm at the clinic seeing a doctor I don't know - what I need is right there. Saves lots of time (and money).

about the depression study

Placebo Effect Is More Myth Than Science, Study Says

Given that phase 3 clinical trials for new treatments have to demonstrate an improvement over placebo and other methods of treatment, this seems like wishful thinking to me.

Alternatively, he may be taking Wennberg too far and arguing that most of modern medicine is a sham.

Doubtful, but possible.

And I agree, the placebo effect should factored into mainstream medical care--but with AGGRESSIVE and rigorous oversight. Which, eh, makes me worried we can't do it at all, because I've lost faith in our ability to do that kind of thing. So sad.

Modern medicine isn't a sham, but several aspects of the ways in which drugs are studied and marketed seem like they are. Here is Malcolm Galdwell's article on the subject from four years ago.

I hate to be pedantic, but...

"...it's far cheaper to give someone a chest incision than an angina..."

maybe should be "it's far cheaper to give someone a chest incision than an ANGIOPLASTY"?

Maybe there is a way to harness the benefits of the placebo effect by educating and coaching patients to use their own power of visualization and expectations to create attitudes that help lead to better health. The placebo effect is essentially convincing someone with problems they will get better, and studies have indicated (perhaps this has been debunked too, I don't know) that visualization exercises and a positive mental attitude, as well as pleasant environments, productive work, and even pets or caring for plants--studies have indicated that all that stuff can have a positive impact on health.

Certainly, in conditions that are heavily dependent on brain chemistry, such as depression, mental exercises, visualization, meditation, social interaction and so on--which we know has an effect on brain chemsitry and brain activity--may have very positive effects.

If modern medicine can accept the roll of attitude, and the benefits of visualization and positive expectations, on real world health outcomes.

Interestingly, all of those examples are used in the chapter about the placebo effect in "Predictably Irrational" by Dan Ariely. Its a good book, generally about the irrationality of people demonstrated via examples with some behavioral econ mixed in. Someone's been reading pop econ books.

Are you really suggesting that we start experimenting with placebo as treatment? Really? Just minor ethical issues.

The important point is that we overestimate the effectiveness of treatments.

When we think about pain management, we know that there are at least 2 aspects. The physical pain and the suffering related to it. They are not one in the same. The same physical pain will cause great suffering one day and significantly less the next. It's influenced by all sorts of internal or external factors.

Placebo probably reduces suffering. Maybe because the patient is taking some action to reduce their pain and suffering, possibly because medication is an indication that another person has taken their pain seriously, or because of their expectations to feel better as a result of taking medication. There are probably other nonpharmacological interventions that could do the same. That might be a lesson worth learning.

if this is any example of how the left thinks we are really screwed. No one is suggesting we give people placebos. There is no way to ovecome the liability issue. What it proves is when people are not responsible for the cost and spending smeone elses money even faux treatments will garner positive results.

A patient paying for these drugs would make more accurate and informed decisions. It's detaching consumption from responsibility that is killing our system.

I do not believe that it is significantly cheaper to give a placebo than say generic fluoxetine Prozac (off patent). The cost of producing the active ingredient in the pill is tiny. On the other hand, fluoxitine causes an improvement on the Hamilton scale which is strongly statistically significantly greater than the placebo.

There is no reason to look at the ratio of the treatment effect to the placebo effect. The comparison should be between the lowest plausible estimate of the difference in the effects
and the costs of the treatment (not just in dollars but also in side effects such as, gasp, weight loss which are, I'd guess a main reason so many people take Prozac).

Also, the bar labled placebo is *not* an estimate of the placebo effect. It is a measurement of the change in symptoms of people who received the placebo. For all the data which you present show, it could just be the normal course of depression, knee pain or angina. To measure the placebo effect, you have to compare treatment with a placebo to no treatment. That means you have to get people to participate (which means return for the second measurement of their symptoms) even though they know that nothing is being done for them.

Consider the placebo effect on the common cold. If you have a cold and I give you a
sugar pill, in a week you will feel better. This is not an argument for giving people with colds sugar pills. That should be obvious. the argument you make in this post for syndromes other than the common cold is just as obviously invalid.

The evidence that prozac is better than the placebo is statistically significant at standard significance levels. The lower end of the 95% interval of the benefit from prozac over the placebo is definitely positive. Comparing it to the cost of prozac (tiny *including* measured side effects) is the reasonable thing to do.

We are talking about less than a gram per year per patient of a chemical which can be easily synthesized (plus the side effects which are a on average a benefit).

see http://tinyurl.com/5gdytl for more of my thoughts on the topic.

chris,

I never studied much about the placebo effect in college, but I did study a lot of neuroscience and psychology and one thing that was always, always the case was that studies were badly misrepresented in the mainstream media. So I'm a little skeptical of the idea that *one* study written up in the New York Times is sufficient to say that the placebo effect is "debunked".

Now, maybe you've got a phD in this stuff and I'm walking into a pwn, but it seems a little quick to just dismiss the placebo effect as "debunked".

The data on depression leaves out two entire classes of antidepressants: Tricyclic and MAOIs. People who don't respond to any of the modern SSRIs-of-the-month will often respond dramatically well to these older drugs, although the side-effects can be more serious.

Also, Serzone was removed from the market years ago because it was linked to liver damage.

My back is still fucked-up after a gay-bashing I received in 2005. And I take pain killers (not chronically, but PRN).

I RESENT that a placebo would lessen my damaged spine pain.

What's with the scaling on the first chart, anti-depressants v. placebo? 2-4-6-8-30-32-34.

I am not a statistician so I generally like my bar charts with equally spaced divisions like the others in your series.

Listen to Waldman, y'all. Chris' "about the depression study" link (to Delong's blog, and a pretty accessible graph done by Mark Liberman, is also a good place to start.

Another problem with placebos that hasn't been mentioned above is their potential to undermine trust -- not only between health care workers and patients, but even between patients and family members. It's essentially impossible to keep a secret like that indefinitely, and when it comes out someone's bound to feel manipulated.

You do understand that an internal mammary ligation is not an angioplasty, don't you, Ezra? Internal mammary ligation is a long-discredited treatment for angina. I couldn't figure out the history from a quick perusal of Medline, but I would suspect the 1960s or even 1950s. I know that bypass was in its infancy in the early 1970s.

There are multiple M.D.s who read this blog. We're actually liberals (Obama is running 2 to 1 in physician donations this year -- that's a rough, if inexact measure of liberals in medicine.) If you aren't clear about a medical term, procedure/treatment, or what-have-you, just ask. We'll be happy to give you an answer, supply or reference, or at least shower you with opinions.

I have long believed that a person could make a fortune selling placebos under the "Cure All" brand. It is after all "the standard" against which all other medicine/drugs are tested, it is effective (a little bit) and as Ezra notes, rather cheap.

The NYT article isn't about a single study. What the researchers did was to find every study they could find that compared treatment, placebo, and no treatment. They found 114, and found another 42 a few years later. They found, except for pain, no difference between placebo and no treatment. If I remember correctly they even added fields to the MEDLINE database to aid data collection going forward.

http://tinyurl.com/66ttxf -- Is the placebo powerless?

The whole placebo story started out as a myth, and was never tested, and now that it's been tested it's not supported by data. There's no reason to buy into it. It's just folk science.

BTW, the authors are leaders in the field of evidence-based medicine, which ought to be a big part of health care reform. Ezra needs to get on the ball here.

The NYT article isn't about a single study. What the researchers did was to find every study they could find that compared treatment, placebo, and no treatment. They found 114, and found another 42 a few years later. They found, except for pain, no difference between placebo and no treatment. If I remember correctly they even added fields to the MEDLINE database to aid data collection going forward.

http://tinyurl.com/66ttxf -- Is the placebo powerless?

The whole placebo story started out as a myth, and was never tested, and now that it's been tested it's not supported by data. There's no reason to buy into it. It's just folk science. It's not so much a matter of disproving the placebo effect, but of establishing it in the first place!

BTW, the authors are leaders in the field of evidence-based medicine, which ought to be a big part of health care reform. Ezra needs to get on the ball here.

That's called metanalysis and it's subject to the same pitfalls and problems as any other study -- there are lots of ways mess with the parameters to get the result you want. The same initial skepticism should apply.

With the medications the shocking number is not that 6-8% of the people who improved with a placebo (the placebo effect is well known). What's shocking is that only 8-12% of the people who took the medications improved. Remember, there are also side effects with taking drugs, on some cases very serious side effects.

You want a comparison between placebo and no intervention at all. Your graphs here may reflect no more than that most diseases improve on their own, either by their natural history or by regression to the mean.

"The implication of course is that it's far cheaper to give someone a chest incision than an angina..." is just dizzying. We haven't been doing IMA incisions for angina in decades - because they do no better than placebos. You're impressed that placebos fare well against discredited techniques? Not sure what "an angina" is.

Ezra,

If you are going to hold yourself out to be the resident blogospheric expert on health care, do everyone a favor and take a class in research design and biostatistics. Some of your stuff is just as bad or worse than mainstream media on biomedical research. You do realize that the "skin incision vs. mammary artery ligation" study is from 1959, don't you?

Sheesh.

Don't placebos work partly because people think they're getting a real treatment? If people know that they're getting a sugar pill, doesn't that diminish the effectiveness?

Many insurance companies already cover placebo treatments. It's called alternative medicine.

One issue with any large studies of drugs is that they ignore potential individual differences in patients - i.e., some patients respond better to one type of medicine than another within the same kind of malady.

I've seen an example of a family member who responded much better to a "me too" drug than a very similar version of the drug. Everyone has slightly different bodies.

This is one reason why "me too" drugs should not be ignored, and why when we look at things like drug response or placebo response, we should also try to dig into individual patient differences rather than just look at large populations generically.

Be sure not to over-emphasize the cost advantage of the placebo. Cheap placebos are less effective than expensive placebos.

"Commercial Features of Placebo and Therapeutic Efficacy"
http://jama.ama-assn.org/cgi/content/full/299/9/1016

Winner of the 2008 Ig Nobel Prize in Medicine.

Ah yes, the "financial plebotomy" as placebo.

As for how to write a placebo Rx, this quote from my book The Throwing Madonna: Essays on the Brain:

Placebos raise a problem in these days of the pharmacist labeling pill bottles with their contents. One cannot admit that the pill is nothing but sugar if it is to work, so a fancy brand name is needed. Among the proposals made in the scientist's humor magazine The Journal of Irreproducible Results for what to name a brand-name placebo are Confabulase, Gratifycin, Deludium, Hoaxacillin, Dammitol, Placebic Acid, and my favorite, Panacease.

Mr. Klein,

It does not appear that you need another response to this post, but as a cardiologist I feel I must point out a few problems with the theory. I will first admit my own ignorance in that I have not seen Peter Orszag’s full presentation. I do, however, chafe at the implication that physicians should be using the placebo effect or any other sort of chicanery to treat patients.

First, a pedantic correction – I offer it respectfully, so please don’t take offense:

“The implication of course is that it's far cheaper to give someone a chest incision than an angina….”

I assume you meant angioplasty and not angina. Angina is a symptom of chest pain that a person has, not a treatment that you give to someone.

As for the example of placebo vs. internal mammary artery (IMA) ligation: IMA ligation is an ancient (1950s-1960s) and no longer used treatment for angina. It is not fair to compare an ineffective treatment to placebo. The standard treatment in modern times for angina and certain related conditions (heart failure caused by blocked arteries) is a bypass surgery – a totally different and much more effective treatment.

Placebo always looks good as long as you compare it with marginally effective treatments. Diseases such as depression and chronic pain (knee pain or otherwise) are notoriously difficult to treat. Hence, placebo looks good in comparison.

Now let’s see what things look like if we compare a disease that has an effective treatment with placebo. Bacterial meningitis is universally fatal if untreated (or if treated with placebo). Appropriate antibiotic therapy is an effective cure about 90% of the time. So if we were to show this slide, the placebo column would show a 100% death rate and the antibiotic column would show a

The same difference would be seen in diseases such as severe aortic stenosis, CML, left main coronary artery disease, bowel rupture and many others.

The cornerstone of western medicine is that a treatment must be proven effective in order to gain acceptance. If placebo is as effective as a proposed treatment, that treatment is rejected. The search then continues for a better treatment. To propose to opposite - that we should treat people with a known ineffective treatment (placebo) - is to undermine the very mechanism by which the practice of medicine is improved. If Mr. Orszag is in fact suggesting this, I must vehemently disagree.

Sincerely,

Carter Hemphill

Oops-

Should read, "the antibiotic column would show a less than 10% death rate."