GSK’s ‘Transparent’ Deception

May 23, 2014

(DAVID HEALY)  –  A year and a half ago this blog ran a series of posts about access to clinical trial data – reporting on how industry were going to engineer the appearances of transparency.  See Won’t get Fooled Again, Access to Clinical Trial Data, and  The Data Access Wars.

Editor note: potentially offensive words have been [changed]


Several months later, soon after being fined $3 Billion, GSK trumpeted their endorsement of transparency by signing up to the AllTrials campaign and declaring their intention to put in place a method to allow researchers access to clinical trial data that would go beyond the wildest dreams of researchers.

See April Fool in Harlow, and GSK’s Journey. Its all too easy to imagine a marketing department figuring that academics don’t have very wild dreams. When GSK signed up to AllTrials Ben Goldacre rolled over and purred.

The BMJ featured Andrew Witty on their front cover as the candidate of hope.


In contrast, on this blog, 1boringoldman and on RxISK a small group have warned consistently that this was not good news.  That what would be put in place was a mechanism that gave the appearances of transparency but in fact would lock academics into agreeing with GSK and other companies as to what the outcomes of their trials have been. No one wanted to rain on the AllTrials parade – it never seems like a good idea to fracture a coalition. RxISK put the AllTrials logo on its front page.

Not content with a few academic ghost authors, GSK’s maneuver has put industry well on the way to making Academia a ghost, a glove puppet manipulated by company marketing departments. Meanwhile Iain Chalmers co-wrote an editorial with GSK endorsing the GSK approach (The Attitude of Chicks to Trojans and Horses) and the British Government produced a document on clinical trial data access that could have been written in GSK central.


The great hope for those dismayed at all this lay with EMA who following Peter Gotzsche’s initiative and a European Ombudsman’s ruling looked like a beacon of hope.  But this week EMA has come out and said it is going to put in place the GSK model of data access. Everyone is in a spin.  AllTrials are asking for more donations to continue their successful campaign.

As someone who has been working the GSK system, I can say with confidence that this is a disaster. The key thing that companies are trying to hide are the data on adverse events.  To get to grips with the adverse events in a clinical trial is a bit like playing the children’s game Memory – where you have a bunch of cards with faces turned face down and you get to pick up two and then have to remember where in the mixture those two were when you later turn up a possible match.


In the same way, picking up adverse events is about recognizing patterns – patterns of events, and patterns of deception. To do this you have to be able to spread maybe a hundred documents out over a big area and dip back into them if something in one document reminds you of something in another.  The new GSMA-ESK remote access system simply won’t allow this. Not only will it not allow this but it is about to make things far far worse than they are at present.

At the moment when it comes to studies like Study 329, GSK have been stuck by a Court order with putting the Company’s Study Reports up on the web where they can be downloaded and pored over – all 5,500 pages of them for Study 329.  They have refused to do the same for the 77,000 pages of raw data from Study 329, making it available to a small group of us through a remote desktop system.

For all other trials – future and past – investigators won’t even be able to get the Company Study Reports in usable form.  They too will only be accessed remotely. For anyone who wants to look at the efficacy of a drug this might just about work for outcomes that involve rating scale scores or lipid levels.   The efficacy of drugs is pretty well all that most Cochrane groups, Iain Chalmers and Ben Goldacre are interested in.  The Cochrane exceptions have been Tom Jefferson, Peter Doshi and the Tamiflu group. But this system is a bust when it comes to adverse events and it won’t work if the efficacy outcomes are in any way complex.


The first point to make is this.  Clinical trials are not all they are cracked up to be.  Even if well designed, not using surrogate outcomes, of sufficient duration, done on patients who actually exist, and not written up by ghostwriters, clinical trials systematically get the wrong answer, especially on adverse events.  Clinical trials are the gold standard way to hide adverse events.

One of the risks of the data access wars is that it will put an unwarranted premium on clinical trials and their data – and in this way play straight into pharma’s hands. This is what led “Crusoe” to warn Peter Gotzsche a year ago that his data access crusade might backfire – See Marilyn’s Curse.

Let’s make no mistake here – it’s morally indefensible that there is not full access to the data from scientific experiments.  In this sense Peter is right and his outrage is well-placed and close to magnificent. But, ceteribus paribus, it would be better for mankind if all clinical trial data were sunk to the bottom of the sea rather than being made visible to academics stuck in a submarine and only able to view things through a periscope, which is what the GSMA-ESK system offers. It might have been better if AbbVie had won their legal action.  Instead EMA’s accomodation with them has [screwed] us all.

Rape is a loaded word these days but in so far as what is happening is an abuse of consent and will primarily do harm to women and children it perhaps come close to being the best word.  Consent processes in clinical trials were about telling you you were on a new drug that might be dangerous or might be involved in a marketing trial.  Instead they have become a way for companies to justify hiding your data on the basis of a confidentiality clause they have slipped into the forms. See When Does Yes Mean No.   Iain Chalmers, Ben Goldacre and AllTrials appear to have signed up to this.

Whether raped or screwed, the Dan Markingson case is a stunning example of what has gone wrong – the Markingson petition is probably a much better petition to sign if you really want to save yourself and others you know and love than the AllTrials one.


Second the data companies are really hiding is their adverse event data.  There are other ways to collect adverse event data. We invited you 18 months ago to join an AbbVie – Let’s Do The AbbVie Again.  This Irish invention is the reverse of a boycott – another Irish invention. A boycott of a company’s drug would mean you don’t get the benefit.  If you decide to AbbVie a company’s drug, you – we – can all make these drugs better by reporting on the effects they have.

Company efforts are geared more than anything else to ensuring that doctors and patients don’t report adverse events or ensuring that these events don’t register.  If you really want to get up their nose, if you really want to send the marketing departments into a spin, if you really want a company CEO to blow a fuse, this is what you need to do.

Companies want to transform adverse events into non-information.  You can stop this happening. If an event happens to you on a drug, you are in possession of the missing information.  It’s our tolerance of the patients who have Disappeared in clinical trials that is killing medicine as we have known it.

AbbVie with us and then sit back and take pleasure in a marketer who says

Thank you for helping us make our drugs better – without you we couldn’t do it“.

Now there is of course a huge conflict of interest here. was set up precisely for this purpose – to register adverse events. But we will hand all events on to FDA or MHRA or whoever you want us to.  What we will also do though, and we invite any doctors or others out there with backbone to help us do it, is to decide when a drug is causing an event – this is something no regulator will ever do for you or for anyone.


The other option is a Boycott.  Doctors could refuse to prescribe drugs for which the information was not fully available.  The Panalba and Thalidomide cases have shown that this is the one thing industry is scared of – Report to the President.  If I made claims about a drug to my colleagues but refused to show them the data, they’d have no problem telling me to get lost. I’d be boycotted from here to kingdom come.  But when it comes to industry, 99% of doctors lack [courage].

Doctors have been given a license to degrade us by treating us like addicts – the origins of prescription-only status.  They have been given a license to print money – we can only get our drugs through them.  The very least they could do in return is show some backbone. But this is a decadent situation and decadence rarely breeds courage.


The boycott was likely invented by Irish women.  The Abbvie was too.  There is one bulwark still standing in the way of GSMA-ESK.  It’s the European Ombudsman, an Irish woman – Emily o’Reilly.