I don't think this is right. Placebo effect definitely exists for conditions that are largely influenced by mental perception. The common example is pain. You can reduce people's perception of pain by deploying the placebo effect, e.g. giving them sugar pills that you convince them will reduce their pain. It extends to other similar conditions which are not generally (or possible to be) measured directly, but rather based on a patient's self-reported scoring. Like "on on a scale of 1-10 how would you rate your experience of this condition". Placebo effect can work for that. But not for other more tangible conditions.

Not only does a placebo reduce pain, naloxone will reverse the pain reduction just like it would if you'd given them morphone instead. Placebo effect isn't simple psychosomatic, rather something real and physical is going on inside the human body.

In my mid 20s I tried antidepressants for the first time. To me it was a big step because like many, I had a false perception of it having an unnatural effect on my personality, but I was finally ready to try them. The doctor said they will take at least 2 weeks to have any effect, and despite knowing that AND knowing about the placebo effect, I still "felt better" for several days after I started the regimen. To me, that was absolutely proof of the placebo effect, especially because after the 2 week window the effect was a backfire where I was in bed for a day and couldn't do anything. The pills backfired on me.

A single data point like that should never be considered anything close to "absolute proof" of anything - because you have absolutely no way of knowing that, for whatever reason (random chance, or the food you were eating at the time, or a compliment somebody paid you on the day you started taking them, or....), you might have felt better on those first days even if you hadn't started taking antidepressants at all.

Correlation is not causation, as they say.

(Hope your depression is gone or at least not too bad now days, regardless of what drugs or placebos may have played a part!)

Did you read any of the articles I linked?

> The most important study on the placebo effect is Hróbjartsson and Gøtzsche’s Is The Placebo Powerless?, updated three years later by a systematic review and seven years later with a Cochrane review. All three looked at studies comparing a real drug, a placebo drug, and no drug (by the third, over 200 such studies) – and, in general, found little benefit of the placebo drug over no drug at all. There were some possible minor placebo effects in a few isolated conditions – mostly pain – but overall H&G concluded that the placebo effect was clinically insignificant. Despite a few half-hearted tries, no one has been able to produce much evidence they’re wrong. This is kind of surprising, since everyone has been obsessing over placebos and saying they’re super-important for the past fifty years.

The words "clinically significant" and "benefit" are not the same thing as the effect being real. To me it reads as if they are testing the hypothesis that a patient comes into ER with a sprained ankle and the doctor gives them this "new powerful prescription pain pill that just came out" and instead tricks them with a sugar pill. If this worked, I'm sure it would be used as much as possible. And the study you linked is simply confirming that PE is not an effective treatment for anything.

That's not the topic at hand here, which is "is the PE real?". For me it absolutely is.

This is what you were replying to when you said "The placebo effect itself isn't real":

> while a sugar pill doesn't make a tumor disappear, it can be very good at pain management.

And now you go to this version:

> There were some possible minor placebo effects in a few isolated conditions – mostly pain

So is it real or not? We're just saying the same thing, what's the point of saying it doesn't exist and then revert back to exactly what was said originally?

Using science to attempt to measure qualia sounds like a good way to produce whatever results you want.