When negative data fails patients by publication omission

Publication of negative trial results are as valuable as positive ones for healthcare professionals

Half of all clinical trials never see the light of day. The complicit combination of keen clinicians, medical journals and bottom-line driven pharmaceutical industry has not served patients well. All clinical trials must be registered and published in full if healthcare professionals are to make a fully informed decision of what care to provide patients. Yet it is not just stringent regulations—or indeed their enforcement that are needed—but rather handing ownership of clinical data back to patients themselves.

Millions of patients took part in such trials thinking that they were helping others, but they were misled. If a drug company does not like the results of a trial, it can hide them from doctors and patients, so that a drug’s true worth or side effects remain hidden. “This situation is a massive betrayal of patient trust,” says Sile Lane at UK science advocacy group Sense about Science, located in London.

“The idea that people are volunteering for research, putting themselves at risk, and then that is not being used by anyone or being made available to researchers is just disgusting,” says Lane. Her organisation is involved in the All Trials campaign – this is also led by the Cochrane Collaboration, a global independent network of researchers, professionals, patients, carers, and people interested in health, James Lind Initiative, a patients, carers and clinicians partnership and open access journal PLOS and in the USA by Dartmouth’s Geisel School of Medicine and the Dartmouth Institute for Health Policy and Clinical Practice. The AllTrials petition has been signed at the time of writing this article by 82,682 people and 540 organisations.

Meanwhile, the current situation leaves a doctor prescribing a drug on best available evidence, not knowing whether this drug is really the best option or if they have been deprived of half of the data. Studies show that trials with positive results are twice as likely to be published as negative results. “Comparisons of trial protocols with published papers have shown widespread selective reporting” says Peter Gøtzsche, director of the Nordic Cochrane Centre based in Copenhagen, Denmark. “The benefits of drugs have been much over-rated.”

Long term campaigning

The issue has been the object of continuous efforts by WHO to increase transparency on clinical trials by establishing the International Clinical Trials Registry Platform (ICTRP) in 2005. They also proposed that the findings of all clinical trials must be made publicly available, as far back as 2008. Implementation remains tricky as this platform remains voluntary.

Get high quality articles on Science policy, innovation and society delivered directly to your inbox.

We respect your privacy.

Medical journals have also favoured compulsory registration of all clinical trials. And since July 2005, journals that adhere to the guidelines of the International Committee of Medical Journal Editors (ICMJE) will consider reports of clinical trials only if the trials were registered before patient enrolment began.

In the wake of these international efforts, UK-based Sense about Science has helped lead the international All Trials campaign for the last 2 years; the organisation is part of a growing band of activists, doctors and scientists pushing for all clinical trials to be published in full. At the time of writing this article, about 540 organisations have signed up to the campaign and 80,000 people have signed the petition. The campaign is demanding that planned clinical trials should be registered, with a summary of the trial protocol, before the first participant is recruited; a summary of results should be publicly available where the trial was registered, within one year of completion of the trial; trial sponsors or others who produce a full report for marketing authorisation or any other purpose should make this publicly available.

“Science that happens and then no one knows about it or no one can repeat it is not science done properly,” says Lane. “Scientists and researchers must have access to scrutinise the data independently.”

Clinicians’ complicity

The trouble is that scientists, publications and industry are part of the problem. Blame falls upon trial investigators, who fail to publish negative results, journals that push for only positive trials, and the pharmaceutical industry, especially, according to Roberto Frontini, president of the Brussels-based European Association of Hospital Pharmacists (EAHP). “They have hidden data because it would compromise the sales of new drugs and that is really unethical,” he says. This is not an ethical nicety. It contravenes the Declaration of Helsinki, guidelines for the ethical behaviour of researchers.

Gøtzsche does not mince his words either. He compares big pharma to criminal enterprises and says the tobacco and drug industries have much in common. “The morally repugnant disregards for human lives is the norm,” is what he writes in his book, Deadly Medicines and Organised Crime: How big pharma has corrupted healthcare. “The business model of big pharma involves organised crime,” says Gøtzsche. “It also involves cheating with research. If we got access to the data, we would be able to show when the companies cheat and this would harm their sales while benefiting the patients and our national economies.”

To conceal unwanted findings, all sorts of statistical “tricks” have been used by the industry. These include ending a trial early, reporting on one outcome that was not the primary outcome you had set out to study and focus on subgroups. GSK notoriously failed to reveal trials showing the drug Paxil (paraoxetine) was ineffective for treating depression in children. And trials that revealed it increased risk of suicide. Another example is that of an unpublished study in the 1980s, which could have predicted that anti-arrhythmic drugs should not be given to patients who had a heart attack – perhaps 100,000 people died unnecessarily as a result. French pharmaceutical company Servier was also confounded by the French judicial system in 2010 for underplaying the potential heart-related adverse effects of Mediator (benfluorex), prescribed off-label for obesity. These are just a few examples, among many.

Campaigning successes

The All Trials campaign has notched up some wins. Lane explains that clinical trials regulation was being discussed in the European Parliament when the campaign began. But the focus was on decreasing red tape with nothing about data transparency. Thousands of supporters wrote to MEPs and health ministers during 2013. And a new law, due to come into force in 2016, will now include rules on clinical trial transparency. Specifically, all drug clinical trials in Europe will have to be registered before they begin. And a summary of the results will have to be published on the register within a year of the trial’s end. Meanwhile, information in clinical study reports will no longer be considered commercially confidential.

Lane says some companies have responded better than others. For example, GSK and Johnson & Johnson (J&J) handed over clinical trial data on pharmaceutical research to a group at Yale University, in Connecticut, USA, called The Yale University Open Data Access (YODA), which is running research projects on those and giving access to independent researchers. Meanwhile, GSK has created its own website to share data from its own trials and that of other sponsors’. GSK, J&J and Bristol Myer Squibb have been leading the way,” says Lane. “At the other end are companies like AbbVie and InterMune, which both sued the European Medicines Agency a few years ago to keep trial data hidden.”

However, those campaigning for greater transparency and accountability, like the AllTrials campaign, are not just about getting trials registered and all results published from here on in. Even if this were achieved, trials and data for the vast majority of drugs we use today could still be kept in the dark. “Some of the retrospective data is going to be difficult to access, and there can be legitimate reasons why some [historical] data is never going to be made available,” says Lane, “but this old information is still really important.”

And it is not just industry that needs to put pedal to the metal. “The big health funders across Europe have skeletons in their cupboards,” she adds “so taking steps to go back through projects they funded and see what resources are needed to make the information available.” In this respect, researchers though have a valuable contribution to make too. A study published in the New England Journal of Medicine found that 62% of administrators at all 122 accredited medical schools in the USA said it was okay for a clinical trial agreement between academics and industry sponsor to be confidential.

Although there are now enough regulations in place in Europe and in the USA, according to Frontini, what will be needed is implementation. An independent US audit published in the British Medical Journal in 2012 showed that only one in five trials had met the compulsory reporting requirement to be posted within a year of completion on clinicaltrials.gov a measure in place since 2007. Despite this fact, according to Sense about Science, no fine has ever been levied against any company or researcher for failing to post results.

Misconduct by omission

In Europe, some believe that one solution is for all university contracts should follow the same format and forbid gagging clauses. This is the point made by British author Ben Goldacre in his book Bad Pharma, who is also involved in AllTrials. Also, he thinks that all professional bodies should deem the failure to publish trial data as research misconduct. And he believes that all research staff involved in any trial on humans should be jointly and severely liable for ensuring that it is published in full within one year of completion.

He goes through examples of how industry favours researchers, who know how to play the game. And the game is marketing not science. There are indications that companies tend to spend more on drugs that are not so effective. As he writes: “Marketing, therefore, one might argue, exists for no reason other than to pervert evidence-based decision-making in medicine.” He quotes some estimates that the pharmaceutical industry spends twice as much on marketing as research and development, with about €50 ($60) billion in the USA alone.

With free access to all data on a treatment, many eyes can perceive potential problems with statistics or flaws in methods. And there are already examples of this situation, where an independent researcher uncovered a link between a drug and serious side effects. This is important because doctors rely on meta-analyses and industry literature to help them make the best decisions. “Doctors and pharmacists don’t have enough time,” says Frontini, “so we rely on reviews. If the negative data is not published, then these reviews show only the positives. And all the reps will go to doctors with the published data and say this is a fantastic drug. They don’t lie; they show you the data, but there are lots of examples where after many years the evidence turned out to be the opposite [to what was published.]”

Given ownership of the data back to patients

Gøtzsche does not believe we are on the way to fixing the problem of industry hiding and distorting drug trial data. “No company is genuinely interested in sharing data, “ he says, “They only do it, hesitantly, because we have forced them to do it, and they set up all sorts of obstacles for getting access, which is not access in my opinion.”

Industry should not own the trial data. “Patients run an unknown risk by participating in trials and it is therefore obvious that companies cannot own trial data. They belong to all of us, which is also the view of the former European ombudsman and other lawyers I have spoken to. Anything short of this is unacceptable exploitation of patients and society,” he says.

He argues that the whole set-up is topsy-turvy. And industry needs to be shoved back from trials. “We need a major culture change where we see clinical trials as a public enterprise done for the public good and performed by independent academic institutions,” he explains, “The industry could still pay for trials but should have nothing to do with them.”

“This could break the vicious circle where drug companies choose investigators that have long-standing relations with the drug industry and don’t ask uncomfortable questions.” He cites the European Society of Cardiology’s estimate that university centres could perform drug trials for about one-tenth the cost of industry trials where there are numerous for-profit middlemen who tack on hefty surcharges.

The transparency struggle continues

Those involved in the All Trials campaign say patients should be clear about the trial that they are participating in and should interrogate any consent form to make sure that their participation will make a difference.

Gøtzsche notes the fact that advertising and PR firms are now running clinical trials in Europe and the USA is a clear sign that companies do not see marketing and research as separate silos. In his book, he writes that a patient consent should read as: “I agree to participate in this trial, which I understand has no scientific value but will be helpful for the company in marketing their drug. I also understand that if the results do not please the company, they may be manipulated and distorted until they do, and that if this also fails, the results may be buried for no one to see outside the company.“

“Finally I understand and accept that should there be too many serious harms of the drug, these will either not be published, or they will be called something else in order not to raise concerns in patients or lower the sales of the company’s drugs.” This might seem an exaggeration, but the pharmaceutical industry has some of the heftiest criminal fines ever paid out by commercial companies on its ledger. Whether this situation can indeed be reformed, remains to be seen.

Do you agree that all clinical data should be published?

Your thoughts and opinions are valuable, feel free to use our simple comment section below.

Featured image credit: CC BY-ND 2.0 by wr52351

EuroScientist is looking for contributors!

If you would like to write guest posts in EuroScientist magazine, send us your suggestions of articles at office@euroscientist.com.

Go back to the Special Issue: Trust

Anthony King

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

3 thoughts on “When negative data fails patients by publication omission”

  1. Of course negative clinical trial results should be published.  Indeed all negative results in the wider domains of science on general should also be published.

  2. It is quite obvious that all data from trials should be published so health care providers, government and regulatory bodies can make the most informed decisions possible. It is a matter of simple logic and ethics.