Interventions that manipulate how patients report symptoms as a separate form of bias

Choosing the right term for this is a bit tricky because there are a number of closely related concepts (such as the placebo effect, and others which I'm not sure have names), and if a term actually comes into common use people will probably end up mixing this concept with a number of other concepts (like how psychosomatic, health anxiety, factitious disorder, malingering, self or mis-diagnosis, and exaggerating symptoms are all mixed up).

I personally like "treatment-inherent cognitive bias". Like Jonathan Edwards said, I think "treatment-induced cognitive bias" could probably be applied more broadly, such as in cases where people get side effects from the actual drug in a placebo controlled trial, and hence are more likely to think the drug is working. So it's a cognitive bias which is treatment induced, but it isn't treatment inherent in the same way that telling someone to lie about or ignore their symptoms is.

I think "treatment-inherent cognitive bias" also covers the problem with only using subjective outcomes as it's assumed a cognitive bias would apply to subjective but not objective outcomes. There are other issues like being told the treatment is or isn't evidence based, or feeling pressured to please the therapist or seem like a good patient, but I don't think they would be covered under that term, because they aren't really inherent to the treatment or necessarily a cognitive bias. Those could possibly be covered by a term like "practitioner-influenced response bias" (similar to "courtesy bias" or "social desirability bias" but more situation specific).

 

I think the roots of the kind of concept of bias we're grappling with here go back even further.

Ever since I've been sick the information publicly available in all the places a patient will look first - NHS, MEA, AfME - all give the idea that most patients will recover and pacing, or somehow expertly managing the minutiae of daily living while keeping one eye on a clock & the other on a symptom diary will, eventually, lead to results.

So even before they participate in a study patients are primed to believe if only they do it right, have faith, recovery will happen.

It's a bit like expecting a fair trial with an unbiased jury when members of the jury have been avidly following events in the newspapers.

 

I was talking about this thread with my husband and he said that the way they use these questionnaires is the way a magician forces a card on the member of the audience.

 

Exactly. Both will result in skewed answers, but how much is attributable to each of those two factors will be indeterminable.

 

Actually, the decision by the ethics committee in Norway against the proposed LP study there basically referenced this sort of bias as a reason for rejecting the study. This describes the bias in a nutshell.

"NEM believes that the method poses a risk that the intervention may affect the participants' responses, for reasons other than real improvement."

 

Which may explain the shock of researchers in the field and the ridiculous level of media coverage they are pushing. This bias is everything, without it there is nothing. Not surprisingly, as this is exactly what they think is the problem and therefore seek to influence, they just know they don't have to admit to it. Hence the whole "we may never be able to do one of those excessively biased studies again", which isn't true but should be and may be on its way with small decisions like this, if more are to follow, however unlikely.

IMO this is what people think the placebo effect is. Accurately measure something and it disappears entirely. Allow for the imprecision of a guesstimate/rating and you can change it subtly, in a similar way someone's rating of a song or a meal can be influenced by other factors, because there is no good, real answer here, the numbers mean nothing, it's just a relative scale that is very unlikely to be properly graduated, such as the difference between 1-2 is not the same as the difference between 5-6, whatever the scale.

Without this, the whole of psychosomatic ideology, functional this and that, and in general psychobehavioral manipulation falls apart entirely, in the end Wessely's "the placebo is one of the most powerful interventions we have" depends on this Big Lie. And that's not even what's happening, no doubt many more terrible experiments will be funded in the future, they just went too hard at it this time, thought they could maximize all the biases and get away with it. It nearly worked.

But any scrutiny at all is bad news for this industry, it exists by people not looking at it, just imagining what it could do if it were real. They tell it loud by saying how this could mean that standards could be raised and how they are obviously terrified of it.

 

Interesting article highlighted by Nina Steinkopf in her blog:

https://twitter.com/user/status/1419240086231736323


The Dangerous Side of Placebo Research: Is Hard Science Boosting Pseudoscience? - Benedetti - 2019 - Clinical Pharmacology & Therapeutics - Wiley Online Library

 

https://twitter.com/user/status/1419240089910120452


https://twitter.com/user/status/1419240092825268232

 

https://twitter.com/user/status/1419240095887003655


https://twitter.com/user/status/1419240102384087041

 

The Benedetti article is interesting in that it is clear that Benedetti himself would like to make use of placebos. One might say that he can hardly complain if the quacks want to as well!

I think the parallel with the BPS people, who also feel that it is OK for them to make use of the placebo effect, as long as it is 'official' therapy that does it, may cover Benedetti too.

 

Fascinating.

I spent a lot of my working life trying get account managers not to raise client expectations beyond what was reasonable to deliver within their budgets.

In every other walk of life setting someone's expectations too high will result in complaints and maybe even legal action.

Perhaps if the feedback loop between patients and healthcare professional existed the placebo effect would be seen in the same light. The customer/patient doesn't necessarily complain initially but waits for the assumed benefit to materialise and eventually realizes the emperor is naked.

Contrary to the opinion of some patients that disappear off the books, never to be heard of again, are not necessarily happy customers.

 

I had the same impression. Haven't looked at it closely but I have a suspicion that the "hard science" Benedetti refers to (his own work presumably) is rather weak.

I haven't seen anything yet that shows the placebo effect isn't simply due to response bias.

 

Surprised it took this long to become noticeable. When you lower standards all the way down to nothing just to make it easier for your pseudoscience to get pushed through, you allow all other quacks to do the same. The BPS ideology has provided a very easy to repeat formula, and it's super cheap as well. It's even being promoted by many "serious" institutions like entire national health systems and publishers like Cochrane, who are clearly endlessly just as gullible about everything "mind-body-blah-blah" as a Goop convention is over quantum healing energy, or whatever is a fad these days.

Fitting that it's the people pushing for quackery who are mad about it, aside from us anyway. I don't know if they genuinely expected to keep control of this but that was always a fantasy, probably bolstered by how long non-official charlatans took to clue in on the gravy train. The fact that cheap apps basically have the same "outcome", i.e. different responses on generic biased questionnaires of little relevance to the problem, opens a free-for-all like never before, a true golden age of pseudoscience in medicine.

Neuroscience plays roughly the same role here as "quantum" is used in various other forms of woo, Deepak Chopra opened the door wide on this. Relying on science too hard for anyone to understand is time-honored tradition of quacks everywhere. An entire discipline of pseudomedicine, all the power with non of the accountability, was built entirely on this. They have no one else to blame for themselves, especially as their only grievance is clearly that others are abusing the weakness they have built for themselves. They're not even concerned at pseudoscience, it's their thing after all, they're only mad they aren't the only ones allowed to.

In hindsight it all looks very stupid but to be fair, it looked just as bad the day this started, well over a century ago. Endlessly running the same idea in loops is some people's idea of "science", but it sure as hell isn't.

 

Would “treatment-induced reporting bias” be a good phrase?

 

There is evidence of an acute (short term) effect on pain through conditioning of endorphins (which are an evolutionarily derived mechanism for us to escape danger when suffering from pain) and a similar effect for acute nausea.

There are no other scientific effects. The claim that there is an placebo effect on increasing dopamine in the brains of Parkinson's Disease patients is trivial, because it doesn't actually help with their symptoms and a myriad of other interventions would also have a similar effect. (Such as my soon to be patented, eat-cake therapy).

 

Doesn't sound very appetising, but maybe I'm missing something.