This key post in the Politics of Care series stems from an invite from Norman Fenton to give a lecture on December 6 to a group interested in the evidence swirling around vaccines. It is accompanied by The Handmaid’s Vaccine on RxISK, which gives a video of the talk, whose text and slides are below. The sound effects in the video are slightly mixed at one or two point and you might need the text to clarify the points made.
This talk is for all who are interested in evidence and how we generate it as well as for a group of people who are pro-vaccine, to the point of being volunteers in clinical trials, but who have ended up being harmed by them. They are the ones doing the science and demonstrating what science means – as I’ll explain – but their work is written off as misinformation.
The company handling of SSRI harms shows we came to classify real Evidence as misinformation. Many of the company tricks involved in this effort to persuade us the world is flat will be known to you but the brazenness with which they were deployed and the failure of physicians to spot what was going on may be new.
I have a doctorate in Serotonin Reuptake and was keen to try the SSRIs early on. Two men I put on Prozac became suicidal. Their problems cleared after stopping and re-emerged on starting another serotonin reuptake inhibitor and stopped again after stopping. See Creaney et al.
This is as clear a causal connection as you can get anywhere in science. I sent the cases to Lilly for comment and presented them in forums. No-one offered alternate explanations. Others reported similar cases during the year to publication of this article.
This rash of cases forced FDA to require Lilly to defend their drug. As my article came out, Lilly published this Beasley et al article in the BMJ. It came out on the same day the company presented their case at an FDA hearing – September 20, 1991 – stating:
- The plural of anecdote is not data
- It’s the disease not the drug
- Are you going to believe the misinformation or the science?
- It’s all the fault of the Church of Scientology (1991’s Anti-vaxxers).
The BMJ article shows more suicidal events on Prozac but the paper said these were not statistically significant and so there was no problem. FDA talked about heart-breaking cases reported to them but concluded the science didn’t link Prozac to the problem.
Here is Tony Hill, who created RCTs, saying 20 years later RCTs can help assess one of the 100 things a drug does – something we might be able to use for treatment purposes. This, by definition, means RCTs are not a good way to evaluate a drug. See Clinical Trials are not Safe.
Saying RCTs are not the way the truth and the light, these days, is like saying the Bible, the Koran or the US Constitution aren’t reliable.
Slide 5 Watch this.
In a depression trial, investigators focus intensely on one thing – does Prozac have an effect on mood. Pretty well everything else is ignored. The statistics we use don’t work unless there is an intense effort to collect everything we can about this one outcome.
And so, depression trials miss something that happens to almost everyone who takes an SSRI within 30 minutes of the first pill – your genitals go numb. You can search the RCTs on these drugs and all you will find is that perhaps 5% of people have sexual issues on these drugs.
Emotional numbing is another extremely common effect almost completely missed. This is how these drugs help. This is how these drugs might help someone diagnosed as depressed but the key point is that it is much more common than depression recovery.
Similarly in the vaccine trials, the common thing is a multiplicity of Spike protein effects – doing this we hope might help but if we are hypnotized by what is hoped for we will miss and have missed what these Spike proteins are actually doing.
If we just depend on RCTs, we end up knowing almost nothing about a drug.
The idea that an RCT shows there is a favourable Risk-Benefit ratio for a drug or vaccine can only hold true if the thing we are looking at is the commonest thing this vaccine does – like a parachute for instance. The commonest thing is a life saved and the Risk Benefit is favourable but we don’t need an RCT for parachutes.
If what we are hoping for is pretty rare – as in vaccine or SSRI trials – and in particular if we don’t know what we are missing, such as an obliteration of our ability to make love, perhaps for all time, then claiming a favourable Risk Benefit ratio is psychotic.
The first rating scale for behaviour was the Hamilton Rating Scale for Depression. An aid to make sure physicians checked on most of the things that might be abnormal in depression while they were interviewing a patient. An aid to help a doctor do an interview that would help the patient to live the life they wanted to live.
If you cleave to the checklist you will do very standardized but possibly disastrous interviews. For instance, on the Hamilton Scale, there is an item on suicide which could stem from the illness or from the drug – it needs a judgement call as to which of these is responsible. Ditto for sex, for sleep. Just checking yes for suicidality risks going badly wrong.
Checklists like these however became viewed as scientific instruments. They look better to hospital managers than DH asking about your daughter or partner. Without judgements, in medicine we call these diagnoses, rating scales are meaningless – other than to help a doctor to help you to live the life Pfizer want you to live.
The latest twist on this story are the rating scales for adverse events companies now run electronically which let people rate up to 12 things that happen after the vaccine – such as sore arm, headache, nausea etc. This ensures certain events become statistically significant – and are put forward as a result as the only things we know for sure happen on the vaccine. See Johanna Ryan’s work on Virtual Trials.
There is no area for people to contribute anything else – so reports of other adverse events end up coming from outside the trial and are viewed anecdotes – misinformation. Companies like Pfizer tally anecdotes. What else would you do with misinformation?
Here is Fluvoxamine, an SSRI supposedly good for Covid. There have been lots of dropouts in the trials done on this, enough to invalidate the trial.
Side effects though could be endorsed on pre-populated lists that included cough, fever, nausea etc but not suicidality, homicidality, sexual dysfunction or the other things this drug causes that were likely responsible for the huge dropout rate.
Many look to drugs like this as an alternate to vaccines. Some doctors advocate them as our Hi Tech versus Albert Bourla’s Hi Tech. There are lots of low tech things that might be more helpful like getting you off some of the Tech you’re on rather than throwing more Hi Tech at you.
As we throw Hi Tech at you, we miss the fact that RCTs convert poisons, from whose use we hope to bring some good, into sacraments – something that can only do good. Most believers figure having as many sacraments as you can daily is a good thing where its seems equally obvious to most of us that taking more than one poison at the same time is unlikely to be all that safe.
One more quirk is companies always want their Ugly Ducklings to have an I’m a Swan moment – thalidomide was the fourth most profitable drug in the US last year but will be pushed down to fifth by Albertine this year. Thalidomide is a drug that causes suicidality, sexual dysfunction and birth defects – just like the SSRIs including fluvoxamine.
For drug companies, rating scales ensure you do an interview that produces figures which are the most seductive way to get the patient on their drug. The interview helps you to help them to live the life Pfizer want them to live.
This is not just true for rating scales, it is true for any measure – peak flow rates, bone densities, blood pressure or lipids, or sugar. It may be important to do something about some figures, but the goal is to help people to live the life they want to live – not the life Pfizer want them to live.
A stopwatch can be a wonderful motivator to achieve a dream – it provides data from one fraction of our lives. If we remain on top of that fine – but what about weighing scales? Just after they were introduced we got the first descriptions of anorexia nervosa. In the 1920s, they had norms for ideal weight attached to them and eating disorders mushroomed. They migrated into our homes in the 1960s and eating disorders became epidemic.
It’s very difficult to ignore figures for weight. Without data from every other aspect of our lives at the same time, we risk being trapped by this one data source. We become neurotic.
Can we let bone densities remain thin, or lipid levels remain high? Yes, we can. You think of post-mortems as something that reveal what we died from – they more often reveal what we can live with.
Figures create risks and pharma makes money from treating risks rather than diseases. We are seduced into taking drugs when we are healthy.
The Covid dashboards are a great mechanism to generate perceptions of risk and fear. The vaccines of course treat risks – not disease.
The Meatloaf title Paradise by the Dashboard Light is what Pfizer sees but its Hell by the Dashboard Light for us – this now extends to the evaluation of lectures and ensures we pander to people rather than challenge them.
It’s extraordinary how little we have put into treating SARS-Cov, the disease in this case and its associated pneumonia. Curing diseases is not a good business model.
Slide 12 As Peter Drucker, the Guru of Marketing Science put it 50 years ago:
The goal of marketing is to make selling superfluous. The aim is to know and understand the customer so well that the product or service fits and sells itself.
Imipramine was the first antidepressant. It beats the later SSRIs in RCTs. It treats melancholia – they can’t. They are useless for severe depression. Melancholia comes with a high risk of suicide.
Imipramine was launched in 1958. A year later at a meeting in England, Danish psychiatrists made it clear that while it was a wonderful treatment it made some people suicidal.
Let’s do a thought RCT of imipramine versus placebo in melancholia. Even though it can cause suicide, we would expect it to reduce the number of suicides because it treats the condition. This RCT would be great evidence antidepressants do not cause suicide.
Here is the data on the trials in mild depression that brought the SSRIs to market – a doubling of suicidal events compared to placebo.
Imipramine looks the same in mild depressions. Now it too causes suicides. So RCTs tell us nothing about cause and effect – they can give us diametrically opposite answers. This is because these aren’t drug trials. They are Treatment Trials and in any clinical Trial, the condition confounds the effects of the drugs.
People evaluating drugs pre-RCTs knew this. When a patient becomes suicidal in a trial you have to use your judgement to work out what has happened but you are told not to.
This is true in every clinical situation where drugs and conditions cause superficially similar effects – diabetes and glitazones both cause heart failure, osteoporosis and bisphosphonates both cause fractures – and this makes it impossible for an algorithmic exercise as most RCTs are to establish what is happening.
Here is what a real drug trial looks like. Companies ran these studies in the 1980s and found that SSRIs make healthy volunteers suicidal, cause dependence and sexual dysfunction but we heard nothing about this when the drugs launched. These Trials enabled companies to game their Treatment Trials to hide these problems.
Vaccine trials are healthy volunteer trials.
This slide shows data straight from a 2006 GlaxoSmithKline paper. GSK’s SSRI paroxetine was in trouble – the RCTs data for Major Depressive Disorder seem to show paroxetine causes suicidal events. The real data are likely worse that GSK admit to here.
But never fear RCTs come to the rescue. GSK also did trials in people with Intermittent Brief Depressive Disorders – IBDD. These are borderline personality disorder to most people – patients who have suicidal events much more often than anyone else. But these patients can meet criteria for depression and could be entered into Depression RCTs.
Prozac in these patients didn’t work. Paroxetine didn’t either and had a 3-fold higher suicidal act rate than placebo. GSK then did another trial in a similar group of patients. Why?
The answer is here. Here are IBDD data from the two GSK trials. I have seen other data for these two trials which make paroxetine look worse but let’s stick with GSK’s story. We could even add 16 more events to the paroxetine arm and still get the same magical outcome
When you add the IBDD data to the MDD data – all of a sudden paroxetine doesn’t cause suicidal events it protects against them.
Something like this must happen in every treatment trial with heterogenous patients – back pain, breast cancer, diabetes, hypertension, osteoporosis, parkinson’s disease. We can use an effect a drug causes to hide an effect a drug causes.
RCTs are not a way to work out what is going on. Back pain trials will insist you use analgesics rather than antibiotics – which is wrong for the 10% of backpains caused by infections.
You’ve seen the son of this slide before. Here are the parents. In all AD RCTs there was a 2-week washout period during which patients were whipped off prior medicines. We now know this was a tricky thing to do – it gives lots of suicides – a bit like the two-week post vaccine period.
But companies argued as the patients were on nothing, all these events should be counted as placebo events – as the diagram here illustrates.
The Prozac 1991 paper had an increased number of suicidal events – but hey not statistically significant. Undo this maneuver –– and they are statistically significant.
Here are the paroxetine data presented to FDA. We’d prefer the figures for paroxetine to be better than placebo but what’s a fraction between friends.
Undo the washout maneuver and this is what the data looked like. FDA knew what was going on and that it breached regulations and did nothing. And these figures don’t look like a drug that should be approved without warnings.
When that was rumbled, companies changed the game. Patients terminated from their SSRI who went into withdrawal and became suicidal were viewed as placebo, while those who stopped placebo and were put on an SSRI and committed suicide were classified as a placebo suicide – on an intention to treat basis. Regulators didn’t ask questions.
Sylvia Plath committed suicide a week into an antidepressant – a common timeframe.
This advert is for the type of antidepressant she was given, an MAOI, featuring a space shuttle, aimed to giving doctors the impression this drug will get their patients into orbit faster.
Here is a space shuttle – the safest transport ever per million miles travelled – but not so safe if expressed in terms of exits from and entries back into earth’s atmosphere.
Rather than express suicidal events per patient exposed companies stuck to events per thousand patient years – having taken care to ensure some patients doing well remained in extended follow-up for months or years.
When the RCT data turned tricky and we got Black Box Warnings – companies turned to Real World Evidence – like national suicide rates. Here you see the claims for Norway which were typical of all Nordic countries – as SSRI use increased suicide rates fell – which is not compatible with the SSRI data.
If you look though suicide rates are going up with pre SSRI AD use until about 1988 – 3-4 years pre SSRI when they begin to fall.
Here is Norway again where you see suicide rates falling from 88 or so and what you see if them rising as autopsy rates rise and then falling in step – as ill-defined deaths fall and rise.
Slide 30 This is true for all the Nordic countries – See Reseland et al.
We routinely hear that SSRI use is escalating. It’s not – the same number of people go on them each year. The increase speaks to the growing numbers who are dependent on them.
This has implications for suicide rates – you are only likely to see an effect on an index like this during the first few years. In the case of vaccines, this years rate of myocarditis and thromboses will become the new normal – See Healy and Aldred 2005.
In 1999 I was asked to participate as a speaking at a company symposium in London – speakers would need to produce articles for a supplement. I said yes and soon after had an email with my article. It was a great Healy article saying the things Healy say in the way he says them with Healy references. No one who knew my stuff would pick it out as not mine.
I emailed back saying I figured on writing my own. There was surprise at the other end but they said okay. I sent it on to them and they said this is rather good but there are some important commercial messages in the other one – we’ll get Siegfried Kasper to put his name on it.
Here it is – only one word changed from the original paper – the name of author Kasper. Everybody in Vienna knows this but its done no harm to SK’s career. You can find materials saying you can trust doctors like Kasper because they have written a 1000 articles or more. Its still a great Healy article etc.
A year later I was in Pfizer’s archive where even the loo-paper was stamped confidential. I came across this working document – which was the articles on Pfizer’s SSRI Zoloft being managed by Current Medical Directions – a medical writing company.
Inside there are pages listing the articles published or in train on Zoloft for the anxious, the depressed, the young, the old etc – here you see the PTSD page.
You will see on the right – the articles were written for these essentially negative studies saying the drug worked wonderfully well. One would go to NEJM – the other to JAMA. And on the left – you see TBD – authors names are To Be Determined. Pfizer’s marketing department will work out who would be the best sales people for the drug.
This is not just a mental health issue. It holds for all treatments across medicine.
Here is the most famous RCT of all time. It has a distinguished authorship line and is in the journal with the highest impact factor in child psychiatry and says paroxetine works wonderfully well and is entirely safe for children who are depressed.
This internal GSK document from 1998 shows the company knew the trial had shown the drug didn’t work and proposed taking out the good bits of the data and publishing those which is the article you have seen. New York State took a fraud action against GSK on this basis who were also fined $3 billion which led to access to the trial data and what you are going to see next.
Slide 38 The full story is on Study329.org
Through this legal action we got access to company data no one ever sees. The efficacy data is pretty irrelevant, but it was still possible to show that no matter which way you cut the data paroxetine was not more efficacious than placebo.
The tricks used to hide the problems were the real interest in these data.
The original article had 10 pages. Regulators see an 800-page Clinical Study Report (CSR) plus nearly 5000 pages of appendices – these are notional they are there but no-one in MHRA or FDA will look at them. We saw these and a further 77,000 Clinical Record Form (CRF) pages.
Point 2 points to data that just didn’t get transcribed from the 77000 pages to the 5000 pages. Point 10 is patients on placebo got SSRIs – I can explain how. But I want to focus on the coding issues.
The psychiatric adverse events all got grouped in CNS or neurological events – into which the groupers also put headaches and dizziness. The dizziness was not neurological – it was cardiovascular because it this case the comparator drug lowers BP especially when used in double or triple the adult dose.
The effect of this was to drown out the signal from psychiatric adverse events. So there is an issue about grouping. We were sensitized to this by Elizabeth Loder, the BMJ editor handling out paper – which took over a year to publish with 7 review rounds and 7 reviewers – who objected to our every mention of headache. As it turned out was a headache-ologist, who was an opinion leader for GSK but above all was the wife of an attorney working in Ropes and Gray who had been the lawyers defending GSK against the $3 billion fine.
So here, leaving out headache and dizziness, in the lower bar you see the number of suicidal events in the Keller paper – once you decode them from emotional lability. In the middle bar, GSK revised this after being asked to do so by FDA when a fuss blew up. In the upper bar you see that we found more again – and there were more than we missed as I’m about to tell you.
How does this fit Co-Vaccines? Well, here you see Pfizer’s report of their adverse event data – a ton of them have disappeared into a higher order coding group called General Disorders. The crimson half of the bar shows you these are serious, potentially lethal. General Disorders is a meaningless group – it needs unpacking.
In a Pfizer trial, one man poured petrol over himself and set a match to it, intending to kill himself. He died 5 days later from his burns. His death was coded as burns. But the company had to write a Serious Adverse Events narrative and if you got to see that you could work out that he should have been coded as suicide.
Companies have found a way to get around this, as found out after Study 329 finished. Here is a young man on a street waving a gun. Its Kyle Rittenhouse. In Study 329 a boy of 15 was picked up out on the street waving a gun around and threatening to kill people. He was hospitalized and should therefore have had an SAE narrative but the company coded him as intercurrent illness.
Four children dropped out of Study 329 coded as intercurrent illness – all were taking paroxetine. Add them into the picture you have just seen and things look a lot worse.
What is intercurrent illness? This was almost certainly an adverse event on paroxetine but invoking an intercurrent illness that means you really should not have been entered into this study means there is no need to write a narrative. This loophole has been there for 25 years and FDA have not moved to close it.
We know Astra-Zeneca broke the blind and got rid of serious adverse events like the ones that happened to Bri Dressen – see New England J of Misinformation. Here you see intercurrent illnesses turning up in this same Astra-Zeneca trial.
Here you have Maddie de Garay who has been tube fed and needs a wheelchair since a few days after the second dose of Pfizer’s vaccine in their trial for 12-15 year olds. But the company says no serious vaccine-related adverse events happened in this trial. They claim she has hysteria and of course that antedates the trial and so the vaccine can’t have caused it.
Few people know that FDA approved paroxetine for children – here is part of FDA’s 2002 letter of approval to GSK. The key bit is typed up so you can read it. The date is important – the Keller paper was 2001.
You can see here GSK told FDA that Study 329 was negative and you see that FDA agreed to approve the drug on the back of three negative studies and also agreed that there was no need to mention this in the labelling. Why would FDA do this?
Here are the published results of adult trials of antidepressants nearly a decade earlier. The picture looks pretty good.
But as Erick Turner has shown, this is how these studies looked to FDA. A different picture. Companies don’t leave negative studies unpublished, they know FDA are happy to let them publish negative studies as positive.
Why? Did GSK tell FDA – if you tell the world Study 329 was negative, we might get sued for fraud – which they did and fined for $3 billion. FDA don’t feel inclined to blow this whistle and MHRA and EMA have even less incentive.
So here is Study 329 again. The author is not listed. In the case of trials done in children, pretty well the entire literature was company written. The mismatch between what articles claim and the data when we see is the greatest known divide in medicine but likely not atypical. Study 329 was a good and ethical trial compared to some of the current vaccine trials.
There are now 45 negative trials for antidepressants in minors – out of 45 trials done. Yet antidepressants appear now to be the second most commonly taken drugs by teenage girls.
In this New England J of Misinformation article, the first thing to note is the author is not listed here. Twenty of the 29 apparent authors are company people. Few are clinicians and none are likely to have seen anyone harmed.
Second the trial was run by I Con rather than Pfizer who subcontracted to Palladium Research, who subcontracted to Ventavia and we know Ventavia ran a shit-show.
Anyone with experience of company trials knows that it is worth looking at the centres involved because for instance in a trial of aripiprazole where there might be 33 centres with 30 producing results for the drug that would not get approval, but perhaps 3 in places FDA won’t visit that found every patient put on the drug did fabulously and every on placebo was seriously injured or died a horrible death and adding both together produces a result that can squeak by FDA.
There is scope to wonder if something similar happened in this trial.
What we do know is that more people died on vaccine than placebo and lots more people disappeared on vaccine and FDA’s current leadership for whatever reason would prefer to be dead before anyone gets access to the data and their correspondence with Pfizer.
Is there a House in the Doctor? The medical drama House was watched and loved by many. Dr House was good at solving puzzling clinical cases by pulling on the thread of some minor detail which led to the answer.
Doctors today have close to lost the ability to say an evident X causes an evident Y – largely down to the mantra that only RCT evidence tells us what a treatment does and we can’t believe the evident anymore. Wife shoots husband point blank in chest – did she kill him? Who knows. In perhaps 1 case in 100 he had a heart attack just beforehand – we need to pass the 100 cases on to those experts in CDC, FDA, EMA etc and let them work it out.
As a result docs report maybe 1 in 10 or 1 in 100 serious side effects to regulators who file these away and do nothing with them.
This is unlike airline pilots who also report near misses and refuse to fly if these reports are not taken into account – after all if the customer dies the pilot does too. This is not true of doctors.
There is a profound misapprehension of the role of a regulator. They are not part of the health apparatus. Their job , perhaps easier to see in the case of food, is to decide if this yellow stuff is butter or lard colored to look like butter. If butter – it is not their job to decide if this is good butter or not or if butter is good for us or not. Ditto with drugs – the role is just to tick a box if certain criteria are met.
They have no abilities to or training in establishing if a drug or vaccine causes a problem.
Here is Walter Raleigh getting his head chopped off. After the fact legal systems recognized the injustice of convicting someone based on hearsay and said cases could only be decided on the basis of evidence in the room that can examined and cross-examined.
The first thing MHRA do with any reports is to remove the names of doctors and patients. This converts them into hearsay, anedcotes, misinformation. It means no-one can decide if there is a link.
MHRA will say till the crack of doom they are looking to find causal needles in the haystack of reports but faced with a needle-stack they can’t seem to spot a needle.
The key to determining cause and effect is an encounter between a doctor and patient. All the data is there. After a first run through there is a chance to follow up when oddities about the data come to mind. Remove the possibility for an ongoing two-way encounter like this and you remove the ability to establish cause and effect.
The one tool regulators have with anonymized drug reports is proportional reporting rates but as Matthew Crawford pointed out you can’t even use this for VAERS because you need lots of drugs in the mix for this to work properly. Besides proportional reporting rates are a cop-out. They might look more scientific than interviewing someone but they aren’t.
If someone commits suicide on an SSRI, their doctor will be advised by their insurer not to say the drug caused it or say anything that might lead to a further legal case. Insurance is supposed to be a business that supports us to take risks but is not doing this here.
If the doctor breaks ranks and blames the drug, a coroner, who can say a street drug caused a death, has no box to tick to implicate a prescription drug in a death.
Media guidance equally ensures journalists cannot say the obvious – the drug caused the suicide if the coroner hasn’t done so – and all this will apply to vaccines also.
If the coroner goes rogue and writes to the regulator and intimates that the evident cause of death was the drug or vaccine, the regulator will check on what the doctor has said and if the doctor didn’t finger the drug or vaccine – they won’t.
If the case is so Evident that both doctor and coroner go rogue, as in the Alana Cutland case, the regulator will respond, as MHRA did, that we only have a handful of reports like this – not enough to let us work out what might have gone on.
In the TV series House, the hero pays heed to tiny things that don’t fit the pattern and after twists and turns finds how it all hangs together.
When a wife shoots her husband in front of Dr House these days he seems unable to work out what’s gone on. Ok the guy may have had a heart attack at the same time and she shot him afterwards out of spite at being cheated out of a pleasant moment but 99 times out a 100 it’s pretty simple, she killed him. House though has lost this plot.
Some great doctors encourage their colleagues to report adverse events to regulators – which makes the problem worse. Regulators will file these reports away until the crack of doom. Unless doctors have the courage to say – look I know what I have seen and the vaccine or the drug killed my patient, they make things worse.
Dr House can’t get his head around the fact that with drugs and vaccines we hope to bring good out of the use of a poison, and sometimes people get poisoned. We and he prefer the idea he is giving sacraments – things that can only do good.
F Scott Fitzgerald once said that a sophisticated mind can hold two contradictory things in mind at the same time and still function – doctors could do this once but can’t now.
Science challenges Muslims and Jews, Xtians and Atheists to leave their biases at the door and come to a consensus about the data. But as in a jury trial, sticking with the data we still have to come to a verdict. A judgement. A diagnosis. It is not the role of a regulator to make diagnoses or deliver verdicts.
A verdict has effects in the real world just as much as shooting someone. There are evident effects from shooting on a husband who dies, and equally evident effects on the wife being tried.
This is important but more important for all of us just now are the effects on the doctor or failing to make any verdict, and just as much failing to make the diagnosis they know or suspect is the right one.
This failure transforms them into Model Doctors – a shrunken replica of the real thing.
When treating a patient, following the evidence can’t mean doing what ghostwritten fraudulent articles say. It has to mean following the person in front of me and coming to a consensus view just as a jury would.
If there is a mismatch between that and the so-called evidence – well all the books say that’s what moves science forward.
See The Handmaid’s Vaccine on RxISK – and its message about Albert and Ursula, the happy couple you see here.