New blog: Placebo effects are weak and reflect regression to the mean

[Woolie]

Placebo effects are weak: regression to the mean is the main reason ineffective treatments appear to work
by David Colquhoun

This blog sets out some of the reasons that we should be suspicious of people telling us the placebo effect demonstrates "the power of the mind over the body". It argues that most of the placebo effect is artefactual. That is, there's no real underlying change to the illness being treated, it just looks as though there is.

The blog focuses particularly on what they call the "get-better-anyway effect", which can make it look as though people have improved from a placebo treatment, when really the placebo had nothing to do with it. The "get-better-anyway effect is defined like this:

You tend to go for treatment when your condition is bad, and when you are at your worst, then a bit later you’re likely to be better

The get-better-anyway effect doesn't mean the person's cured or even recovered. It just means that most illnesses have fluctuating symptoms and that people are most likely to seek treatment when their symptoms are at their worst. Chances are, they'll soon be heading back into a better period, if thy just wait long enough.

In the comments section (also by David Colquhoun)

I’m not doubting that genuine placebo responses exist, and can be measured in experiments that are designed to find them. But the empirical evidence suggests that they provide little or no benefit to real patients. That being the case, their possible mechanisms (endorphins etc) are not only speculative, but are also of little interest.

The main reason why people appear to respond to dummy pills (or to no treatment at all) is nothing to do with placebo responses: it is a statistical artefact. The principle has been understood for over 100 years, so it isn’t obvious to me why it isn’t better understood. I guess one reason is that understanding it may well reduce the income of both big pharma, and of the legions of quacks, who are eager to sell us things that don’t work.

Okay, so he only mentions big pharma and what he calls "quacks". But psychotherapy and other behaviorual therapies are also benefiting massively from this effect.

 

[alex3619]

He does mention this applies elsewhere, such as the efficacy of traffic cameras. Regression to the mean is a general phenomenon where (typically) weak statistical data is used to conclude something that is only really vaguely implied and may be wrong. An artefact is a good way to put it.

Its another bit of language we can use regarding the PACE trial outcomes.

PS In case I did not make it clear, this occurs in large part because the baseline is made at a peak in the data, and over time it will tend to show more like the average outcome. I think in psychiatric medicine based on subjective outcomes we have an additional bias built in because patients are essentially conditioned to give the responses the doctor or researcher want.

 

[SOC]

I guess one reason is that understanding it may well reduce the income of both big pharma, and of the legions of quacks, who are eager to sell us things that don’t work.

Any idea how big pharma is supposed to benefit from belief in placebo effect? I would have thought pharmaceutical companies would prefer to sell expensive "real" medicines for real illnesses. Surely they're not getting rich off placebo pills. Do pharmaceutical companies even market dummy pills? Do doctors really give dummy pills in this day and age when anyone can look up the medication they were prescribed on the internet to see what it's really for?

 

[Woolie]

Any idea how big pharma is supposed to benefit from belief in placebo effect? I would have thought pharmaceutical companies would prefer to sell expensive "real" medicines for real illnesses. Surely they're not getting rich off placebo pills. Do pharmaceutical companies even market dummy pills? Do doctors really give dummy pills in this day and age when anyone can look up the medication they were prescribed on the internet to see what it's really for?

You're right, @SOC. The placebo effect in general is most beneficial to "open label" research - where people know whether they're getting the dummy or the real thing. And of course, virtually all psychotherapy/behavioural interventions are "open label". They say tht's okay because:

1. You can't properly "blind" a behavioural study (people have to be aware of the treatment)
2. The placebo effect is "real" and therefore is a legitimate component of any behavioural intervention.​

I don't have to tell you how weak the second argument is. Given the recent evidence suggesting the placebo effect is largely artefactual, this argument just doesn't fly. Plus, it puts behavioural studies on the same level as even the most questionable alternative therapies.

The first argument is not good either, you can control for many of the factors that induce a placebo response in various easy ways, even in a behavioural study. For example, create a "dummy" treatment, like relaxation therapy, and promote it as a treatment in the same way as the one of interest. Controls for participants' expectations, their degree of investment in the treatment, as well as the "get-better-anyway" effect.

 

[Sean]

Good find, Woolie.

I have been arguing for years that the placebo effect, while genuine, is vastly over rated and of little real world use. It must be controlled for in trials, of course, but beyond that, a big fat meh.

It is one of the main reasons I am not a fan of Ben Goldacre, he is besotted with the placebo effect.

"Within-patient change in outcome might tell us how much an individual’s condition improved, but it does not tell us how much of this improvement was due to treatment."

In order to eliminate this effect it’s essential to do a proper RCT with control and treatment groups tested in parallel. When that’s done the control group shows the same regression to the mean as the treatment group, and any additional response in the latter can confidently attributed to the treatment. Anything short of that is whistling in the wind.

Sound familiar?

 

[A.B.]

I don't think it's safe to assume that the difference between a placebo and no treatment is the placebo effect (in the sense of a mind body effect).

 

[Skippa]

I think the psyche effect can be explained, at least partially, because it takes so damned long to get your first appointment that you are already "regressing towards the mean" when you begin.

Big Pharma get their benefits because A) maybe their treatments aren't as universally effective as they would have us believe in every case, and, B) their "soft options" like OTC painkillers have been successfully been portrayed as magical cure-alls that we reach for whenever we so much as stub our toe.

 

[Hip]

I read an article around 15 years ago saying that the placebo effect was mainly due to many illnesses or symptoms having a natural tendency to get better over time. So if you go to your doctor with some vague pains, even if he gives you no treatment for it, in many cases the pains will just clear up on their own.

 

[barbc56]

I haven't had a chance to read the above, probably after the holidays, but I think there are a lot of misconceptions about the placebo effect, even among scientists. It's a fascinating subject.

Interesting thread. Thanks @Woolie.

Barb