If your standard for differentiating between placebo and "real" effects is participation in a "real" study, you've by definition eliminated pretty much all introspection as a source of data about ones body.
Although you might find it absurd that folks are feeling better from a dietary modification that hasn't been around for a long while, I find it equally strange to dismiss out of hand the lived experience of "so many people".
People are surprisingly incompetent when it comes self identifying and describe symptoms as well as self monitoring disease progression (people here includes doctors). That's precisely why the placebo effect exists.
For example, what would you bet that most people describing improvement on a glutenfree diet are merely experiencing the benefits of an overall healthier diet that has little to nothing to do with their gluten intake and has everything to do with them not eating half a pizza twice a week.
Also the gluten free diet theory (or at least the paleodiet) is the only diet theory that resorts to evolution to try explaining what´s happening when we eat. It is strange that no other theory is looking in evolution to explain what we have to eat, we do it for our pets and for wild animals, but we consider ourselves out of that possibility (people actually has told me that).
It is true that studies are still scarce, and surely it will change a lot during the next years due to new studies and insights. But if I have to bet to one theory that explains human feeding, I´ll do it to the one that is trying the evolution path. Surely there is also the intelligent design diet out there, for those who need options ;).
I'm okay with both the idea that the mechanism is either purely psychosomatic or the result of some coincidental element being cut out of the diet.
As a mechanism, it seems quite hazily understood.
I'm less okay with the wholesale exclusion of the lived experiences of my fellow humans; at some point, people have only one body that they can reliably poll for information.
The placebo effect is NOT well understood, definitely not to the point that you can make that blanket statement.
> For example, what would you bet that most people describing improvement on a glutenfree diet are merely experiencing the benefits of an overall healthier diet that has little to nothing to do with their gluten intake and has everything to do with them not eating half a pizza twice a week.
Oh, I would take that long bet easily, if we would agree on a way to resolve it, though I guess we wouldn't be able to reach such an agreement.
I have my own n=1 story with respect to wheat (not sure if gluten or another component) - two years ago, I decided to drop it, and lost 30 pounds within a month. My diet changed, of course - but not to the extent that would induce a 1-pound/day weight deficit (that's 2000kcal/day if you believe in the provably wrong caloric theory of weight management).
And it happened to me several times in a restaurant that I asked beforehand if something contained wheat, was told "no", ate it, felt "wheat symptoms", asked again, and it turned out that wheat was a non trivial component (seriously, it's in everything. really avoiding it is freaking hard).
People are surprisingly incompetent when it comes to identifying symptoms, I'll give you that. They are also surprisingly incompetent in the trust they assign to the knowledge they came by. test yourself: [1] is dietary cholesterol harmful? [2] does dietary salt cause hypertension? [3] Does artificially lowering cholesterol (e.g. with statins, rather than a diet change) increase your life expetancy? Medicine got all of these wrong for the last 50-100 years, and is only now (very very slowly) making a U-turn. It doesn't get everything wrong, of course - but it sure as hell doesn't get everything right either.
[1] not really. Dietary cholesterol accounts for 5-15% of serum cholesterol, the rest is made by the body. The recommendation to avoid dietary cholesterol is NOT based on evidence (or logic, for that matter)
[2] for over 80% of the population, the answer is "definitely not". Once you already have hypertension, for 80% of the population, the answer is still "not". For the other 20% (which can be characterized) it is "yes". However, the recommendation for low sodium is made to the entire population. NOT BASED ON EVIDENCE or logic.
[3] No. Statins reduce the probability for a heart attach, but actually increase all-cause mortality.
People have no insight. The man who treats himself has a fool for a doctor. Nobody is a better example of this than myself. I had gastrointestinal issues, I ran through a variety of elimination diets under the care of a doctor, and it was something that I hadn't even considered.
Some doctors are only interested in getting paid and seeing you out the door as quickly as possible. An alarming percentage of them from my experience, actually, if you don't fall within what they perceive as "well defined problem".
Especially if your symptoms do not immediately point at a specific problem, you have to be the primary contractor in caring for your health - no one else will do it for you.
Although you might find it absurd that folks are feeling better from a dietary modification that hasn't been around for a long while, I find it equally strange to dismiss out of hand the lived experience of "so many people".