Editorial Note: This is a cross post from 1boringoldman’s Wisdom of the Dixie Chicks. It raises questions about a Lancet editorial jointly written by GSK and AllTrials it would seem. It calls on Iain and Ben to tell us what they are up to. So far there appears to be no clear response from AllTrials. There is an ambiguous message here.
This repost omits the image of a Trojan Horse and a wonderful lyric from the Dixie Chicks which may have distracted attention from the call for a clear response. This is a further call to Iain and Ben to tell us what’s going on.
Here is the GSK AllTrials editorial. This editorial seems to me to entirely miss the point behind consent forms – whose origin was outlined in When Does Yes Mean No. At EMA’s meeting on Access to Clinical Trial data in November 2012, I made the point that most people signing consent forms assumed that their data would be open to scrutiny by independent scientists. Very few of them would be likely to consent to having their data sequestered by GSK or other companies. Iain specifically congratulated me on making the point. This seems to point to a change of mind/heart on his part.
Many people have sensed there is some difference between AllTrials and RIAT. This editorial on the AllTrials side and a talk I was invited to give to the House of Commons All Party Group on Involuntary Tranquilizer Addiction, attached here, which brings out some of the RIAT concerns helps to point up the differences. These need resolution. The appearances of differences need to be debated in forums like Biojest. It would be great if any with links to Biojest, OpenSociety and other groups could take this post and help initiate a debate. The issue is RIAT have asked GSK for patient level data for Study 329 and are being refused it. We are being offered a GSK tabulation of the data. Whose side are Iain and Ben on?
The wisdom of the Dixie Chicks
Publishing the results of all clinical trials, whoever funds them, is required for ethical, scientific, economic, and societal reasons. Individuals who take part in trials need to be sure that data they contribute are used to further knowledge, prevent unnecessary duplication of research, and improve the prospects for patients.
Endorsement of these principles is clear in the support received for the UK-based charitable trust Sense about Science’s campaign demanding that all clinical trials should be registered and reported. However, although the campaign recognises the advantages of analyses based on individual participant data (IPD), it is not calling for open access to IPD. The campaign recognises that risks to personal privacy must be taken seriously. These risks are not just theoretical: a recent study was able to identify 50 individuals from public websites that contained genetic information. The research community must work with others to define what constitutes appropriate protection of identifiable information if it is to retain public trust in the use of IPD.
The commentary goes on to discuss various ways to anonymize the data all of which have the trialists in the driver’s seat. And then there’s this:
PV is a President at GlaxoSmithKline, holds stock or stock options in GlaxoSmithKline, and is a board member of A*Star Board Singapore and Genome Research Limited. IC declares that he has no conflicts of interest. The authors would like to thank Martin Bobrow, Mike Clarke, and Ben Goldacre for helpful comments and critical review of this Comment
So why would Iain Chalmers and Ben Goldacre put their names on this article that feels so much like a Trojan Horse? Having looked at GSK’s proposed process for access myself, it gives GSK the choice of releasing the data based on the credentials of the applicant and GSK’s opinion of the worthiness of their research proposal. I don’t want access to their IPDs to do further research. I want it to check and see if they’re cheating again like they’ve done countless times before. This article talks about the problem’s solution without mentioning what the problem is. It doesn’t mention the gross corruption in the publication of Clinical Trial results, and the epidemic of jury-rigged or absent trials that have turned our literature into litter-ature. It implies that we want data transparency to further scientific exploration. That’s a worthy goal, but off the point.
So why would Iain Chalmers and Ben Goldacre put their names on this article that feels so much like a Trojan Horse? I don’t know the answer to that. Perhaps they can tell us why this isn’t a sell-out?
As a Doctor of Psychology who has lived long enough to observe the fickleness of research and public gullibility (though these are not the guilty ones), I’m pretty sure the same thing has happened in psychology research, which according to the eternal swing of the pendulum, now points to Behavioural methods, to the detriment of all else. Once a path in research is started, unfortunately, bad conclusions only build on themselves. The most lucrative grants go to those who have started a thread and are already immersed in pursuing it, under the belief that these researched must continue to be supported lest their historical funding investment be lost. Indeed, what can granting committees say when faced with fabricated, or at least doctored, near-perfect research results?
Thank you for this post.
Another question is: why doesn’t Alltrials campaign ask for IPD?
Especially when EMA is proposing to discuss IPD public availability.
Of course, publication bias is obvious, outcome reporting bias too. Getting CSR will provide scientific community with important information. And I signed the Alltrials campaign.
But there are examples of incomplete or wrong analysis in CSR used for approvals. Information showing bias was only available in FDA reports and obtained from IPD analysis. (Examples here in French : http://www.etudes-et-biais.com/individual-patient-data-ipd-une-necessaire-mise-a-disposition-de-la-communite-scientifique/).
Should we leave FDA as the sole authority able or allowed to check if CSR are correctly or completely describing the results of trials? EMA is not performing IPD analysis.
Shouldn’t investigators have access to the IPD of their centre to check if data included in the database do correspond to their patient file and sign for it?
Without IPD, the only thing we can say about a doubtful trial is that there is a risk of bias but we cannot describe how the potential bias distorted the results and if it is a bias or a fraud.
Search for a “risk of bias” is the only research we can do. But a risk of bias doesn’t induce exclusion of the trial from Cochrane Meta-analysis which is at the top of EBM pyramid.
IPD availability could add a new top over the EBM pyramid. I hope it is not the reason why IPD availability is excluded from Alltrials campaign.