This is a murky story that like Study 329, to which it links, doesn’t leave anyone looking good.
First the journal – Frontiers in Psychiatry – which is one of the Frontiers group of journals. As outlined in Neo-Culturalism – a must read to get what is going on here and being done to all of us – there is no-one at home in Frontiers. There is a vacuum where once the soul (or at least the rear-end) of the medical establishment could be found. This is where the Surveillance Society meets academia.
Frontier operations recruit people for the most part probably flattered by the idea of being an editor for a day, perhaps occasionally knowing what they are getting into but with some vision of doing good.
In this case, Michael Hengartner, who being Swiss might have found it difficult to resist the overtures of a Swiss business, agreed to get involved in putting together an issue on antidepressants and children. Michael would generally be regarded as an all-round good guy.
Michael recruited Irving Kirsch, also regarded as one of the good guys, to help out, and then reached out to me. Irving dropped out when he realised that despite being an editor it would still cost him a lot to get his article published in the issue he was helping to organize – see Neo-Culturalism.
I initially recruited more contributors than Michael and Irving combined but then began to run into difficulties when Frontiers effectively started to insist it owned me – I had to keep to its deadlines rather than it fit around mine – see Neo-Culturalism.
As part of my recruitment efforts, I figured there were too many good guys being asked to write articles, all saying antidepressants were not a great idea for minors. Fine if you were going south of the border to sort out some pesky Mexican bandits but this was more complex. It would be helpful to get input from some of the guys pushing antidepressants for minors like Elias Eriksson, an aggressive pusher, and Andrea Cipriani. Eriksson refused.
Andrea, who is very charming and difficult to dislike, agreed to get involved. He has pioneered a technique he calls network meta-analysis. He also runs a Precision Psychiatry centre in Oxford University, linked to Evidence Based Medicine – the rough idea is that practice should be based close to exclusively on clinical trials. Andrea and his team would put together a network meta-analysis on trials of antidepressants done in children.
By this point, however, my difficulties with Frontiers were coming to a head and I withdrew from the process. I mentioned modern slavery to them – see Neo-Culturalism – but there was complete incomprehension on their side. There was no great ill-will – for me it was an insight on the way academia and publishing is going and emphasized the need for more co-operative, Samizdat-like publishing.
So, when I was asked would I be prepared to act as a reviewer for an article in the issue, I said yes. The Cipriani team’s article was sent to me. I was one of 4 reviewers – one dropped out early on.
The Cipriani article has just been published. Like Andrea’s widely cited 2018 article on antidepressants, the message in this article is that antidepressants work for children. If you read the fine print there are nuances about some not working that well for some conditions with one or two drugs thrown in that don’t really work at all and one or two conditions for which none work – but the bottom line is they work.
This was a review challenge for someone who has published articles in English and Spanish saying that 30 of the 30 trials done on these drugs in minors are negative. In the case of clinical trials in depression you can look at each of the 30 or so trials and if each are negative and none positive you can say this is the greatest concentration of negative trials for any condition in any age group ever in human history.
Or you can add up the ‘data’ from 30 negative trials and get a marginally positive result which is what Andrea does. This seems crackers to me, and goes completely against the rational of RCTs, but many people including the editors of the Lancet, BMJ, NEJM and JAMA don’t bat an eye at this.
But this wasn’t the core problem. The core problem is that, as in his adult work, Andrea and his team have been working from ghostwritten publications with no access to the raw data. In the case of the GSK paroxetine trials, including Study 329, and the Forest citalopram trials, these publications have led to fraud charges and Department of Justice fines ranging from hundreds of millions to billions of dollars.
It’s not the case that GSK and Forest were just the bad guys who were caught. GSK were pretty ethical, as companies go, when running Study 329. What they did was standard company modus operandi – back in the 1990s. Things are a lot worse now, and it’s almost for certain some of the trials the Cipriani team depend on for their ‘positive’ result are a lot shonkier than Study 329 was.
The Prozac trials are a case in point. They were all negative on their primary outcome – despite published claims that Prozac worked. Regulators in fact approved Prozac on the back of two negative trials, and there are more negative Prozac trials than there are for any other antidepressant given to children.
Then there is the supposedly independent Prozac trial run by the NIMH – the TADS trial, which may have essentially been run by Lilly. This returned something like 34 suicidal acts on Prozac versus 3 on placebo but these all go missing from the 7 publications of this trial. When Jon Jureidini and colleagues tried to get hold of the data in order to look at some of these more closely they were told it had been destroyed.
Pointing things like this out to the Cipriani team and asking is it really possible to do what you are claiming to have done and for it to be meaningful – drew a response from them like the quote Tom Lehrer, the satirical songwriter, put in the mouth of the Nazi and later American rocket maker Werner von Braun: ‘once the missiles are up who cares where they come down, that’s not my department’ says Werner von Braun.
It’s a technical operation, like constructing train carriages for transporting animals and ensuring there are systems to allow urine and effluent outflow – who cares if the animals are cattle or humans.
It’s now the job of academics to take the published literature as it stands, not asking if its fake or honest, and network meta-analyse it and see what comes out. This is a technical exercise not a moral one. Whether we put garbage or genuine material in, the technique guarantees publishable results.
I recommended rejection. The other two reviewers recommended acceptance – primarily on the basis that there were no technical faults. The question then became which way the editor and the journal – if there was anyone at home – would jump. The article was accepted.
In this way, decent academics, nice, even good guys, who are sticking to their job description, and being ethical by doing so, become, to borrow from the Covid lexicon, infection super-spreaders.
They take a ghostwritten, in some cases fraudulent literature, which breaches the basic norm of science which requires claims to be backed with data that can be scrutinized, and they whitewash it.
The image below is from one of the Cipriani network jobs – it even looks like the images for Covid super-spreaders.
When published, these papers get presented at primetime in media outlets with a tagline something like – Biggest study ever proves Antidepressants Work. Everyone it seems can compartmentalise the fact that with a big enough study Snake Oil can be shown to work.
This is not just Andrea spreading problems. Most of the Cochrane Collaboration are complicit in something similar. Around 1995 when they began Systematically Reviewing Drug Trials Cochrane could have insisted that they would only include those trials where the raw data was present – but they didn’t. Around 2005 Iain Chalmers and Richard Horton of the Lancet told the world that ghostwriting and lack of access to trial data were not a problem. Around 2015 the Collaboration were well on their way to being a publishing business – not unlike Frontiers.
Antidepressants appear now to be the second most commonly taken drugs by teenage girls. This weekend, the BBC website contains an article on a doubling of suicides among young women, which speculated on the reasons for this without mentioning antidepressants as a cause – perhaps because Andrea never has said anything other than these drugs work. It says to steer clear of drugs and alcohol which can disinhibit you without mentioning prescription drugs like SSRIs disinhibit too – perhaps because nothing Andrea has ever written mentions this possibility. The case featured in the article, ‘Lauren’, describes effects very like the emotional numbing found on SSRI antidepressants..
This is a generation of women who may never get off psychotropic drugs, who will have miscarriages caused by these drugs, pregnancies on the drugs with babies who have gross birth defects or subtle behavioural changes – autistic spectrum disorder, who may become alcoholic, end up with enduring sexual dysfunction and kill themselves. That’s not my Department says Werner von Braun.