The Placebo Response in the Treatment of Chronic Fatigue Syndrome: A Systematic Review and Meta-Analysis, Cho et al (2005)

This paper is of interest.

https://sci-hub.se/10.1097/01.psy.0000156969.76986.e0

 

Subgroup analysis showed that in clinical trials of psychological-psychiatric interventions (n=8), the mean pooled placebo response was 14%.

 

2005.
Sir Simon Wesley, MD.

Honestly, I would rather prefer to see someone else replicating this, with more trustworthy inclusion criteria.
I would hate to refer to the MD who in 1992 explained symptoms of aluminium poisoning in Cornwall with "significant psychological factors."

 

What are they measuring exactly?

It is important not to confuse the various reporting biases with a true placebo effect. I daresay, across ME patients as a whole, these reporting biases are likely to be similar to any other patient group.

 

Perhaps the psychologic interventions have lower placebo response because the patients are unblinded. For example, Prins 2001 had one of the lower placebo responses at 8.5%. Their methods demonstrate how it's obvious who is getting treatment.

However the authors designed the paper, or retroactively wrote it, so that the either outcome would satisfy their conclusions about CFS. For the first part they say

so that if placebo response is high, it can be concluded that the condition is varying on patients thoughts, behaviors and expected symptoms [non identifiable physiologic correlates]. But then they found placebo response was low, and so added

O.K., but, he's neglected to mention that in the CBT/GET treatments patients know, and those treating them know, who is getting the treatment. So, since the placebo response carries more than just the placebo effect, it seems unreasonable to compare it to blinded "physical" treatments unless both physical and cbt/get were both unblinded or both blinded.

I hope that makes sense, I even got a little confused writing this. For example of one bias difference, if the CBT treatments are unblinded, a patient would be less inclined to "please" the researcher by reporting a better outcome. Whereas a patient who thinks they got a real treatment would still have this desire to please. So it may be that the lower placebo responses don't have anything to do with belief in physical illness just poorly conducted studies.

 

And it's crucial to understand what outcome measures are being influenced by placebo response and which not, and which of them are the most important. If the most important outcomes are those unaffected, then that fact has a certain relevance .

ETA: Edited for clarity. My last sentence was intended as being a tad sarcastic (definitely not at you @Snow Leopard, but toward the usual suspects), and I've edited to help avoid ambiguity. I basically mean that if the outcomes that matter are not influenced by placebo effect, then placebo effect is irrelevant anyway. i.e. As is the case with ME and objective outcomes.

 

The only proven "placebo effect" is the short term reduction of pain through conditioned activation of the endorphin system, everything else is speculative aka pure bullshit.

 

Just to clarify, I realised my post was somewhat ambiguous, so I've edited in case people may have been taking it the wrong way.