Placebos Are Getting More Effective

Wired is reporting that the well-known "placebo effect" seems to be increasing as time goes on. Fewer and fewer medications are actually making it past drug trials since they are unable to show benefits above and beyond a placebo. "It's not only trials of new drugs that are crossing the futility boundary. Some products that have been on the market for decades, like Prozac, are faltering in more recent follow-up tests. In many cases, these are the compounds that, in the late '90s, made Big Pharma more profitable than Big Oil. But if these same drugs were vetted now, the FDA might not approve some of them. Two comprehensive analyses of antidepressant trials have uncovered a dramatic increase in placebo response since the 1980s. One estimated that the so-called effect size (a measure of statistical significance) in placebo groups had nearly doubled over that time."

Comments

WTF

[Xeriar]

You keep saying that word, I do not think it means what you think it means.

There are plenty of other reasons for this to be occurring. Better testing procedures among them.

Re:WTF

[digaman23]

I'm the author of the Wired article, and I would encourage people to read the article itself before taking Peter's post on Science-Based Medicine as the final word on the subject. Peter's blog runs on two sites, and if you visit the other thread here -- http://scienceblogs.com/whitecoatunderground/2009/09/placebo_is_not_what_you_think.php -- you'll see that Peter's well-informed readers offered up many citations supporting my central thesis that he seemed unaware of, many of which were contained in my article. I know that words like "crappy" and "smackdown" feel really bracing to post or read on a blog, but they're no substitute for science-based medicine. Thanks for the link, ScuttleMonkey.

Re:WTF

[ChameleonDave]

You keep saying that word, I do not think it means what you think it means.

No, the guy who wrote that article is wrong. He is using "placebo" where he should be saying "control". A control is what you use to measure the difference between normality and the thing you are testing. In medicine, this may or may not involve a placebo (which means a "pleaser"). For example, I can give 1000 people my new drug, and put another 1000 people in a control group, with no drug. However, I may worry that some of the improvement in my patients is due to the psychological effect of popping a pill; I therefore may give the control group a fake pill to take, called a placebo. If I have enough funding, I may even have three groups: one with the real drug, one control group with the placebo, and one true control group with absolutely nothing. This will often produce three levels of improvement.

A control cannot be described as strong or weak, but a placebo given as part of a control certainly can be. Although it is something designed to have no real effect, the fact is that every aspect of the treatment situation (the colour of the pills, frequency of treatment, the crispness of the white coats...) alters the strength of the pleasing effect, which can have major consequences for health and well-being.

Re:WTF

[Imrik]

The 'original' (as in, the ones used at the time the placebo effect was becoming known) placebos were sugar pills and so sugar has become associated with placebos as a result. Modern placebos are generally inert in the context of the study.

Re:WTF

[anotherhappycamper]

The FDA contradicts your assertion that sugar pills are no longer used. I just did a google search "placebo sugar pill site:fda.gov", and I found plenty of drugs approved this year that list a sugar pill as the placebo.

Or not.

[cloudnin]

Here's an excellent rebuttal to this article by Peter Lipson on the Science Based Medicine blog: Science Based Medicine: Placebo Is Not What You Think It Is

Re:Or not.

[Anonymous Coward]

If by excellent, you mean whiny and with little effect, then yes, it is an excellent rebuttal. Basically the linked article is merely a hair-splitting complaint about the phraseology used in the original article. Interesting only to a pedant, and even so I'm not sure that the pedant would agree with it.

Re:Human race evolving?

[MightyMartian]

1. Of course we're still evolving and always will.

2. It's very likely nothing to do with our brains, and a lot to do with more rigorous testing.

Re:Human race evolving?

[Anonymous Coward]

Speak for yourself; I've got Wolverine's powers.

Re:Patients entering trials are different

[ColdWetDog]

That, surprisingly enough, was in the FA. He points out that the standard rating scale for depression, the Hamilton-D, was developed and validated among depressed individuals who were institutionalized - a very different population from the ones that the drug companies are studying now.

Re:Grunt

[Anonymous Coward]

Statistics are like bikinis.

What they reveal is suggestive, but what they conceal is critical.

Nice to meet you, Aaron Levenstein.

Re:Shooting themselves in the foot

[ceoyoyo]

It does. Read the article.

Re:Believing

[ceoyoyo]

Yes. This is why we have a procedure we call "science" that attempts to take our subjective biases out of the equation. There are no shortage of examples where people have absolutely convinced themselves of things that aren't true.

Actual evidence

[ceoyoyo]

Note that the only actual evidence for a more robust placebo effect referred to in the article is two studies looking at antidepressants. There are also a couple of anecdotes (from companies looking for a scapegoat for their failure) about Parkinson's and Crohn's, but that's hardly evidence.

It would be interesting if there was data for conditions that can be assessed objectively.

The article needed to be about two paragraphs and could certainly have stood to lose all the gushing about how powerful and neglected the placebo effect is. On the bright side, I see Wired is hiring people with no photography or design experience to generate their figures.

Re:It could be

[Mo Bedda]

The earliest humans were, in fact, scavengers, were not contemporary with saber toothed tigers

I do not think that is correct. Smilodon went extinct only 10,000 years ago. I believe humans driving them to extinction is still one of the popular theories.

Re:Placebos future

[rrohbeck]

(Is that enough !!!!s to ensure that nobody thinks this is a serious comment?)

No. Please use !!!1!!one!eleven!! next time.

Re:Placebos future

[DavidTC]

Here it is.

Also, simply inhaling warm vapors when you have a cold, and drinking warm things, especially stuff that 'sticks' to your throat thanks to the fat in the broth, has known medical benefits. That is, in fact, the entire point of cough drops and vapor-rub.

Oh, and don't underestimate the value of just eating something you're sick. Chicken soup provides proteins and carbohydrates in a form that even someone with the worst throat irritation can eat. While they would not, for example, want to eat a cheeseburger, which would have nearly the same nutritional content.

So at the very least, it is a) something warm to drink that will help clear nasal passages, that b) people can actually eat easily while sick and even coughing, and we know both those things already for a fact. Any additional chemical medical benefit is still hypothetical and being tested.

Re:I've read the article

[digaman23]

> But drug trials are not designed to measure the placebo response; they are designed to measure the drug against the placebo Absolutely. That's why the drug companies are pooling their data and doing the Placebo Response Drug Trials survey that I write about - to find out what's going on.