The pervasive problem with placebos in psychology: Why active control groups are not sufficient..., 2013, Boot et al.

The true believers in the "powerful placebo effect" have no problem with magical healing crystals, though some feel that mainstream medicine should have a monopoly over placebo effects (rather than complementary/alternative medicine).

 

I am not sure what you mean. The psychological research that is being done needs controls with accepted physical diseases. Take the PACE trial. They found that there was a large number of drop outs in the GET arm but assumed this was related to social things (moving home and so on) not the treatment. They also found a slightly significant effect of this group at the end of the trial.

If they had had a control group with a physical disease, graded exercise may well have made them fitter and there may well have been less drop outs. That is, it would have put the results into context.

The other groups may have had similar results.

The BPS have made psychological interventions the only treatments researched but it all leaves out any effect of long term chronic illness and that has to be controlled for.

 

I still hold by my argument that these folks defending the status quo for psyc trials would make a huge ruckus if anyone tried to publish a non-blinded study showing support for magic healing crystals.

Supporting evidence here:

Fiedorowicz, J. G., Levenson, J. L., & Leentjens, A. F. (2021). When is lack of scientific integrity a reason for retracting a paper? A case study. Journal of Psychosomatic Research, 144, 110412

 

This sounds akin to the problem that used to occur with older style voltmeters, Avo voltmeters etc, where the very act of connecting the meter could change the very voltage you were trying to measure. Much less of a problem in that case however, because the characteristics of the meter were clearly defined and any reading compensated for appropriately. But if the measuring instrument's characteristics can shift indeterminately under your feet, then there is no way to correct the readings.

 

A good exercise for this would be to submit a BPS-ME/MUS paper with the fewest changes possible, swapping everything related to whatever behavioral crap they are doing with healing crystals. It would likely not be accepted and no doubt the methodological aspects would be admonished, but only because of the use of healing crystals, otherwise they are basically standard practice for clinical psychology. The very flaws they dismiss as irrelevant would be all that matters, with insistance that it's not because healing crystals are involved, even though of course it would be.

I'm pretty sure of the outcome of this experiment. It's entirely because belief in the magical powers of the mind overrule everything, turn critical thinking off entirely.

 

Ben Goldacre for example.

 

I wanted to draw people's attention to some interesting bits in this paper.

They include a section where they address what are common counterarguments against controlling for expectancy effects in psychological intervention designs. Here are the four arguments they consider. The final two are ones we hear a lot from the BPS brigade (I've shown them in bigger font).

 

Brilliant, @rvallee!

 

''

'The requirement to control for placebo problems will make it too difficult to “get an effect”'

We actually do see this (or something close to it) being used. Considering that blinding is a form of controlling for unwanted effects in trials, the BPS crew have been known to say a good few times, that because it is impossible to fully blind psych trials, then they have to go with what they can manage. But as noted plenty of times before (@Jonathan Edwards first highlighted I believe), if the only kind of trial you can do as a cr@p one, then that is no excuse to treat it as if it were a good one ... it is just plain wrong.

 

EBM folk everywhere: hold my beer and excuses.