Blog: The Science Bit by Brian Hughes

[Kalliope]

Good intentions regularly lead to terrible outcomes. While researchers’ hearts may be pure, that fact alone does not guarantee that their works will be blessed.

In psychology, we sometimes talk about the “just-world fallacy”, the belief that in the end our collective actions will average out to produce fair and fitting consequences. This, of course, is a fallacy because our efforts are free-standing. They stand or fall on their merits, not on the basis of karma.

Putting all hands to the pump in a time of crisis certainly feels appropriate, but there are dangers in rushing headlong into the COVID-19 research frontline.

The belief that good intentions lead inevitably to good outcomes may prove to be one of our world’s most dangerous delusions.

https://thesciencebit.net/2020/10/1...ovid-19-science-promote-substandard-research/

 

[Sean]

The belief that good intentions lead inevitably to good outcomes may prove to be one of our world’s most dangerous delusions.

Good intentions are not only not enough, they can even be your worst enemy by blinding you to any harm you are doing.

The person you must be most careful to avoid fooling is yourself.

In many fields — for example, studies of behavioural healthcare interventions — condemning a study’s research methodology can result in quite a bitter backlash. In some cliques it is considered extremely poor etiquette (if not bullying or harassment) to complain about a healthcare intervention study because, after all, the researchers were just trying to help people.

It’s as though mistakes made in the milieu should be written off because of the nobility of the researchers’ aspirations.

But it is perfectly acceptable, of course, for those Very Important People to ruthlessly smear and dismiss their critics, especially if those critics are ungrateful bastard peasants patients who just don't want to get better and lose the wonderful secondary benefits of 'playing the sick role'.

 

[chrisb]

Are the secondary benefits of writing trash papers ever studied? Knighthoods, for example.

 

[Snowdrop]

The cartoonist Gary Larson wrote a book (satire) on the subject of good intentions.

It's called 'There's a hair in my dirt'.

https://www.harpercollins.com/products/theres-a-hair-in-my-dirt-gary-larson?variant=32131125805090

Also, with regards to critics. In the new documentary 'The Social Dilemma' computer scientist Jaron Lanier opines that it is really the ciritics who are the optimists in society. Interesting idea. Then on thinking further it makes sense. Critics hope that things can be made better. Those who smear their critics simply want to be left alone to get on with fulfilling their own lives and not be bothered by how it affects the world around them.

Kudos (again) to BH for taking on a necessary critique from within his own ranks.

 

[rvallee]

The problem is evidence-based medicine. It encourages a race to the bottom where quality of research and reliability of outcomes are irrelevant. Everyone simply tries to chirp out the tiniest slice of "better than nothing" then fight to the bitter end to keep this marginally useless thing the only thing because it's better than nothing so what more do those whiny patients want?

BPS- and EBM-dominated research has caused medicine to embrace the cheapest pseudoscience so fully that it can barely do the real thing anymore, because the "it's better than nothing if you squint really hard" is the only game in town, it appears to be cheap (as long as you don't count everything), it's repetitive, it's immune from criticism and ultimately only serves a supply-side model of medicine, where need is essentially irrelevant. It essentially works like politics, which explains a lot.

And then of course there's the fact that peer review has degraded so much that a paper can be published saying something like "X is, in fact, Y" even though it is purely a statement of opinion devoid of any supporting evidence that just happens to be something the author believes.

Medicine needs to grow the hell up, it's stuck in so many ruts that driving in circles is the only available path, there are so few ways out. The field needs to adopt a model like physics did decades ago, with large projects involving thousands of collaborators in a unified effort. This scattered approach of cheap zero-quality all-quantity copy-paste of random beliefs, devoid of serious peer review, of actual criticism or even substance, is the main factor in the near complete stagnation in improving patient outcomes over the last several decades. It's gotten so bad that fields that have existed for less than a decade display more maturity and ability to grow and learn than what is probably the oldest field of expertise. Time to get out of la-la-land, physicians heal your damn methods, they are terrible for patients. It may be super fun to do minimal work in circles without oversight but for patients this is very bad.

Seeing how the system works, it's easy to see why outcomes are so bad. The outcomes can't be better than the system that produces them. The whole thing needs a thorough top-to-bottom reform. I wish people listened more to Hughes. He gets it. Why do so few even get close to getting it?

 

[Kalliope]

Another article from Brian Hughes on the pandemic.
I changed the title for this thread, so if he writes more on the same theme, there's now a thread where we can have them collected.

You're so vain, you probably think this global pandemic is about you

Many of the history’s great global crises took place in much different times. If burgeoning narcissism really is baked into our modern world, then we might struggle to marshall the solidarity needed to flatten the COVID-19 curve.

 

[Kalliope]

Once it became clear that the coronavirus could spread even when carriers have no symptoms (and so might not realise that they are carriers), public health authorities have consistently advised people to wear masks to combat COVID.

So why do some people — around 20% of adults — decline to wear them?

COVIDxHalloween: The Psychology of Wearing Masks

 

[Kalliope]

The Science Bit: On COVID-19, the economy, mental health, and suicide

If you have ten minutes to spare, here is an audio recording from my remarks at last week’s annual conference of the Psychological Society of Ireland:
...
In this extract I speak about the impact of COVID-19 on the economy, and how the resulting recession can be expected to impact on the population’s mental health

 

[Kalliope]

The Science Bit: Off the PACE and not NICE

Quote:
Here is a video of a lecture I gave in Belfast just before COVID, entitled “Off the PACE and not NICE: Challenges with Evidence in ME/CFS.”
The lecture was part of a conference organised by Hope 4 ME & Fibro NI(kudos to Joan McParland and team). It hasn’t been posted online before.

ETA: I've seen it now. It was absolutely brilliant! I hope it reaches out far.

 

[Trish]

I agree, it's a really excellent talk. I hope it gets wide viewing beyond those of us who already know the problems with ME/CFS psychological research, and PACE in particular.

 

[Barry]

Yes, very good.

A bit unsure where @Brian Hughes criticises PACE for not employing blinding. This can be easily countered by PACE advocates that PACE trialled interventions that cannot be blinded, so to criticise them for not doing so is a flawed criticism. The point, surely, is not that the PACE investigators should have employed blinding (they simply could not have done so), but that they should have designed their study to account for the lack of blinding! i.e. Utilised some objective outcomes, which by their nature are more or less immune to the biases due to lack of blinding.

Pretty sure that @Jonathan Edwards has pointed out that there are all manner of interventions that cannot be fully blinded (open heart surgery for instance?), but that is not an issue so long as objective outcomes are used and not just subjective ones ("answer 'yes' if you feel unwell").

Edit: See my correction in post 13 below.

 

[Trish]

A bit unsure where @Brian Hughes criticises PACE for not employing blinding. This can be easily countered by PACE advocates that PACE trialled interventions that cannot be blinded, so to criticise them for not doing so is a flawed criticism. The point, surely, is not that the PACE investigators should have employed blinding (they simply could not have done so), but that they should have designed their study to account for the lack of blinding! i.e. Utilised some objective outcomes, which by their nature are more or less immune to the biases due to lack of blinding.

I thought he did explain it very well that the lack of blinding was a problem both
- because the patients in the CBT GET groups were told they were in the active treatment group and that it was a proven effective treatment, ie they weren't given neutral messages about what to expect
- and because they used subjective outcome measures and ignored the objective ones in their reporting of efficacy.

 

[Barry]

I thought he did explain it very well that the lack of blinding was a problem both
- because the patients in the CBT GET groups were told they were in the active treatment group and that it was a proven effective treatment, ie they weren't given neutral messages about what to expect
- and because they used subjective outcome measures and ignored the objective ones in their reporting of efficacy.

I've listened to that bit again, and I think I agree with you. I was very focussed on what has been said here in S4ME before - that it is impossible to blind participants to their treatment. But what Brian points out is that even though participants might know what their treatment is - CBT or GET etc. - they could be blinded to those treatments' efficacy, and most certainly not have them actively promoted.

It is actually an important distinction @Brian Hughes makes, between blinding of a treatment versus blinding of a treatment's effectiveness; even if you cannot blind the former then you can make significant efforts to blind the latter.

ETA: Would not avoid the need for objective outcomes though, because such blinding could never be wholly effective. And double blinding would not be possible, because it would be very challenging to blind therapists to the presumed effectiveness behavioural treatment they were administering.

 

[Invisible Woman]

It is actually an important distinction @Brian Hughes makes, between blinding of a treatment versus blinding of a treatment's effectiveness; even if you cannot blind the former then you can make significant efforts to blind the latter.

You see, I may be wrong but I think in PACE that point is moot.

Presumably all trial participants, including the so called control arm, had access to the treatment promotional material published by the PACE authors telling them that is was safe, effective etc.

As far as I'm aware - & please jump in if I'm wrong - participants in the control arm knew that they could try the other treatment at the end of the trial if they wished. So not only did they have the incentive of knowing they were the control arm which in itself causes a negative bias, not only would they need ongoing medical support in the future and so want to keep the medical team onside they also had material available to them telling them of all the wonderful treatment they were missing out on.

This adds up to an incentive to report negatively in the hopes they'll give you the treatment that works at the end of the trial.

Given the trial compares outcomes of both arms, one way of making the treatment look better than it is is to make the control arm outcome look worse. I think they set the scene quite well for that.

 

[Barry]

You see, I may be wrong but I think in PACE that point is moot.

Presumably all trial participants, including the so called control arm, had access to the treatment promotional material published by the PACE authors telling them that is was safe, effective etc.

As far as I'm aware - & please jump in if I'm wrong - participants in the control arm knew that they could try the other treatment at the end of the trial if they wished. So not only did they have the incentive of knowing they were the control arm which in itself causes a negative bias, not only would they need ongoing medical support in the future and so want to keep the medical team onside they also had material available to them telling them of all the wonderful treatment they were missing out on.

This adds up to an incentive to report negatively in the hopes they'll give you the treatment that works at the end of the trial.

Given the trial compares outcomes of both arms, one way of making the treatment look better than it is is to make the control arm outcome look worse. I think they set the scene quite well for that.

I think the point @Brian Hughes was making is that in PACE the participants should have been blinded to the extent of not knowing which arms were receiving interventions being investigated. But the PACE investigators seemed to go out of their way to not hide anything.

 

[Sly Saint]

"
the behavioral intervention research in mecfs
has given us lots of examples of this
it's a very good teaching opportunity
it's a learning moment it's a
a chance for us to get it right.
but we have to engage in a sensible
discussion around it
as hard as that is given human nature
i am concerned that the nice processes
have good intentions
and have a positive sort of vision of
what humans are like.

and i would hope that the nice
guidelines will have enough
kind of insight to be skeptical towards
what people say about the quality of evidence
and that that needs to be looked at very
objectively and that we need to acknowledge that the
starting position for most people is to
be biased and if we accept that we can have better
conversations and better exercises
of uncertainty reduction
um the current review needs to take
great care in that regard
and to manage the expectations not just
of clients because the outcome might not
be the way patients want it to be
but also of clinicians the outcomes of
the nice review may not be to clinicians liking.

if nice review said that cbt was
probably not recommended that would have massive
implications for service providers
given the number of clinics providing
services currently but if that is the
outcome that is objectively most robust then that is the outcome we
should have.
so clinicians need to be prepared that
maybe it's not going to be the way it
was before because now we know more about how these
things go wrong.

you can tell a lot about a researcher by
the way that they respond
to criticism and i would say that the
pace researchers and me cfs researchers
have an opportunity to show us what
they're made of now"

https://www.youtube.com/watch?v=v-ZzfZgu8Y4

 

[alktipping]

is it only me that sees pace as outright fraud they altered it for their own financial benefit . will everyone involved have to be dead and buried before it is called out for the wilful and deliberate fraud that it is . brian come across as an apologist for understandable human failings if greed and outright sociopathy are just human failings we might as well have no law whatsoever .

 

[rvallee]

is it only me that sees pace as outright fraud they altered it for their own financial benefit . will everyone involved have to be dead and buried before it is called out for the wilful and deliberate fraud that it is . brian come across as an apologist for understandable human failings if greed and outright sociopathy are just human failings we might as well have no law whatsoever .

There is always the risk of professional retaliation. We know these people are petty and able to dispense punishment, they have done so in the past, it's not just speculation or "conspiracy". Even as far as Berkeley Tuller did carry some risk, it's far greater in closer proximity.

I have no qualms calling it overt fraud. They knew what they were doing, they knew why they were doing it, and it was entirely for their self-benefit in delivering false promises. But I carry no risk in saying the obvious. Unfortunately that's been the formula: they can bully and be disrespectful as much as they want while we have to remain nice or they will DARVO.

 

[Barry]

is it only me that sees pace as outright fraud they altered it for their own financial benefit . will everyone involved have to be dead and buried before it is called out for the wilful and deliberate fraud that it is . brian come across as an apologist for understandable human failings if greed and outright sociopathy are just human failings we might as well have no law whatsoever .

I do think there is a good deal of fraudulence that has been exposed with PACE, and I hope in due course it gets aired in a legal setting. Mainly because I don't think patients who suffered the consequences all these years will be properly acknowledged and recompensed until that happens.

I think (and this is purely guesswork of mine, nothing more) @Brian Hughes may be endeavouring to get the core messages of the PACE failings across to more people (maybe some at NICE for instance) using a softly-softly-catchee-monkey approach, rather than wading in guns blazing. If that works then I'm all for it; the more scientist stop being spellbound by the BPS BS the better, and the sooner the better. The BPS crew have been very adept at convincing many they are up against anti-science, anti-scientists activists. If a softly, softly approach gets the bewitched to read on, and to read the science bit, then that could be very good; get them to gain new insight. If it works then that's fine by me.

 

[Sean]

PACE may or may not have started out as fraud, but it sure as shit ended up as fraud.

As late as the release of the long-term follow-up results they could have saved face, and even been seen as honest scientists by accepting that there was no long term benefit, that their model had not delivered, and it was time to look elsewhere for answers.

But they chose not to, and instead kept doubling down ever more dishonestly and viciously on their failed claim.

They chose their fate.

So be it.