CLINICAL TRIALS

thebmj.com n 1 January 2015 a watershed moment came for ЖЖOpen data campaign: thebmj.com transparency when a long awaited ЖЖEditor’s choice: Open data: seize the moment policy change at the European Agency (BMJ 2012;345:e7332) came into force. From this date all successful appli- ЖЖEditorial: to industry’s clinically cations to the agency will lead to publication of the Orelevant trials’ clinical study reports. relevant data (BMJ 2004;329:64) The change of policy opens a new era of transparency that has been hard fought for, over many years, by many researchers and a few very committed policy makers.1 It will be followed, in May 2016, by the new European Union Clinical Trials Regulation, THE PIONEERS OF which says that the clinical information in clinical study reports should not be considered commercially confidential, thus remov- ing the legal argument that drug companies have used until now TRANSPARENCY to justify withholding large amounts of trial data. But many battles are still to be fought. For now, the European This year sees the first major step towards full Medicines Agency can still make restrictions and redactions if it considers data to be confidentially commercial (such as proprie- disclosure of clinical trial data in Europe. Here we hail the tary secrets about the drug molecule), although the regulator has determined campaigners who pushed for this for years emphasised that this is likely to happen only in a small number of cases.2 And the new legislation will not apply retrospectively and the powerful people who listened. By Ben Adams to data from drug applications made before 1 January 2015,

THE PEOPLE WHO PUSHED FOR CHANGE

PETER GØTZSCHE TOM JEFFERSON SILVIO GARATTINI AND ALESSANDRO LIBERATI Peter Gøtzsche, director of Since the 1980s the other Tom Jefferson and his Garattini and Liberati, both researchers at the Mario the Nordic Centre cofounder of the Cochrane fellow Cochrane reviewers Negri Institute in Milan, have helped question attitudes and cofounder in 1993 of the Collaboration, Iain Chalmers, worked for years to try to get among academics as well as the drug industry. Liberati Cochrane Collaboration, has and his fellow researchers have access to data for Roche’s was a clinical epidemiologist and researcher who been working tirelessly for years been drawing attention to the H1N1 influenza treatment helped establish the Italian Cochrane Centre. He had to get the European Medicines issue of under-reporting of oseltamivir (Tamiflu) and multiple myeloma and died on New Year’s Day in 2012, Agency to release more data.3‑5 research results, which he says GlaxoSmithKline’s flu at the age of 5715 In 2004 he wrote a personal view in Gøtzsche worked on clinical is “unscientific, unethical, and prevention drug zanamivir The BMJ talking of his disease and the uncertainties he trials and regulatory affairs uneconomic.”6 7 (Relenza). faced.16 He wrote, “Research results should be easily in the drug industry between At the beginning, he explains, Jefferson says, “In 2009 accessible to people who need to make decisions 1975 and 1983. Since 2007, “Our exhortations made virtually we’d never heard of clinical about their own health . . . Why was I forced to make my Gøtzsche says, he has come up no difference,” adding that no one study reports. The first decision knowing that information was somewhere but against much resistance and was interested, “least of all the glimpse of this hidden not available? Was the delay because the results were frustration, but all that changed researchers who were guilty of this universe came to us from less exciting than expected? in November 2012 when the misconduct.”8 This all changed an old Roche submission to “Unfortunately this is possible in a world where European Medicines Agency held with ’s book Bad the UK National Institute for has become dominated by commercial a workshop at its headquarters Pharma and the resultant AllTrials Health and Care Excellence interests. When you are a patient you wonder how (we) that made history. Its new campaign, he says. “These (NICE) that had been leaked researchers can keep forgetting the principle that the head, Guido Rasi, started by have raised the profile of this to us. priority should be collaboration for better hypotheses, announcing, “We are not here to scandal, and we have begun “It contained fragments not competition.” decide if we will publish clinical to see tangible developments of four trials. Now, thanks Garattini is a member of the European Clinical trial data, only how.” towards greater transparency. to the Tamiflu saga, even Research Infrastructure Network (www.ecrin.org), a not “The industry representatives This is encouraging, but the battle politicians know about CSRs for profit clinical research project that was started in were stunned,” Gøtzsche recalls. remains far from won. I think [clinical study reports], and 2004. It provides clinical trial services only to clinical “I have never before seen the winning will require a combination EU legislation specifically trials that commit to be registered, to publish results mighty drug industry lose a of legislation and demands from mandates their disclosure— irrespective of findings, and to make their raw data public battle so completely as patients and the public for greater and all of this in less than five available to other researchers after the completion of during [that] afternoon.” transparency.” years.”8‑12 the trial.

16 3 January 2015 | CLINICAL TRIALS

Laparoscopic colorectal surgery The role of eating fish in stroke prevention PLUS Managing tennis elbow Maintaining weight loss in adults Why WHO’s future hangs in the balance Insoles for medial knee osteoarthritis In search of beneficial drug reactions Management of chronic rhinosinusitis Supplements to prevent pre-eclampsia Gambling addiction: a patient’s journey Diagnosing active and latent tuberculosis Easily missed: giant cell arteritis

344:1-52 No 7838 ISSN 1759-2151 345:1-52 No 7881 ISSN 1759-2151 342:1159-1216 No 7808 ISSN 1759-2151 7 January 2012 | bmj.com 3 November 2012 | bmj.com 28 May 2011 | bmj.com To mark the first major breakthrough in this long battle, in this article The BMJ would like to highlight MISSING TRIAL DATA Why we need the full picture and thank the leading campaigners and policy makers in the United Kingdom and across Europe who have taken risks and persevered to help shape a new era of transparency. Of course, many other organisations have The struggle for access to thrown their weight behind this campaign and helped unpublished trial data to increase the pressure: the French drugs journal Pre- scrire, Transparency International, the International But many battles meaning­ that the full trial data on the vast majority of the Society of Drug Bulletins and Berlin Declaration 2012 are still to be drugs we currently use will still not be available. (http://chn.ge/13HlGYl), to name but a few. And drug fought. For now, However, the fact is that within two years the public companies would not of course have felt the full pres- the European and researchers will be able to read, in full, clinical study sure of the need for transparency without the tireless Medicines Agency reports for all newly approved drug trials, whether con- efforts of tenacious lawyers who exposed some egre- can still make ducted by industry or academia. gious cases of malpractice through the courts. The BMJ “It’s easy to focus on the hurdles ahead, but let’s stop has also supported AllTrials (.net), the cam- restrictions and and appreciate just how far we’ve already come and how paign for full disclosure of trial data, and will continue redactions if it quickly,” said The BMJ’s deputy editor Trish Groves, who is to do so. considers data to responsible for all original research published in The BMJ. Ben Adams freelance journalist, London be confidentially “Suddenly, everyone who used to think it was fine to hide [email protected] commercial data from clinical trials has realised that it’s not OK.” Cite this as: BMJ 2015;350:g7717

THE PEOPLE WHO PUSHED FOR CHANGE

BEN GOLDACRE PETER DOSHI DOUGLAS ALTMAN KAY DICKERSIN AND The psychiatrist, academic, and author of Peter Doshi, assistant professor at the The founder and director of the Centre TOM GREENE two bestselling books about the misuse Department of Pharmaceutical Health for Statistics in and Cancer Kay Dickersin, professor at of science to mislead the public, Bad Services Research at the University Research UK’s Medical Statistics Group, the Johns Hopkins Bloomberg Science (2009) and (2012), of Maryland’s School of Pharmacy in Douglas Altman, has worked to ensure that School of Public Health, Ben Goldacre has used his anger, humour, Baltimore and an associate editor at the methods and results of clinical and Baltimore, and director of scientific rigour, and ability to write to The BMJ, leads the Restoring Invisible epidemiological studies are of high quality the Center for Clinical Trials force the issue that has long festered and Abandoned Trials (RIAT) initiative. and, most importantly, are reported fully and the US attorney Tom within the medical community on to the In 2013 Doshi and his university and honestly.13 Greene started digging for public agenda. He argued, in Bad Pharma, colleagues called for trial sponsors and He is behind the development of many the missing data on Pfizer’s that “the whole edifice of medicine is investigators of abandoned studies to research reporting guidelines, such as pain drug gabapentin broken,” because the evidence on which publish (or republish) their data and the CONSORT (Consolidated Standards (Neurontin). Green and it was based has been systematically proposed a system for independent of Reporting Trials) statement, which Dickersin found that Pfizer distorted by the drug industry. In January publishing if sponsors failed to respond. aims to alleviate the problems arising and its development partner 2013 he cofounded, with the Cochrane He was impressed by the European from inadequate reporting of randomised Warner-Lambert had used Collaboration, The BMJ, and others, Medicines Agency’s inclusive process for controlled trials, and the EQUATOR selective outcome reporting AllTrials (alltrials.net), a group that developing its policy but disappointed (Enhancing the Quality and Transparency for trials of the off-label use continues to campaign for total disclosure that the policy first proposed was of Health Research) online network (www. of gabapentin.14 They say of trial data. He explains that the effort to watered down. equator-network.org), which was launched that this practice “threatens put the transparency issue on the public He says, “I still remain uncertain about in 2008 and describes itself as “the first the validity of evidence for agenda has been slow not just because of what has been achieved, because it coordinated attempt to tackle the problems the effectiveness of off-label resistance from the drug industry but also still remains promissory in . Until of inadequate reporting systematically and interventions.” because of a “force field of tediousness” other groups go out and try to access the on a global scale.” Pfizer and Warner-Lambert around the idea of data transparency that data—and let all of us know how easy or “A research report is the only tangible had to pay $325m (£210m; has long discouraged public scrutiny. hard it is—we will not know whether we evidence that the study was done; if €265m) in May this year Regulators and professional bodies also are where we need to be.” research is not published it might as well to settle allegations that failed to draw the attention of policy not have been done,” Altman said as they fraudulently marketed makers to the issue and in some cases EQUATOR launched a print version of its the drug for uses it was not actively denied that there was a problem. reporting guidelines in October 2014. licensed for.

the bmj | 3 January 2015 17 CLINICAL TRIALS

IMPORTANT PEOPLE WHO LISTENED

GUIDO RASI BEATE WIESELER PATRICK VALLANCE AND ANDREW WITTY The executive director of the European Beate Wieseler and her colleagues The president of pharmaceuticals research and development at Medicines Agency, Guido Rasi, from Germany’s health technology GlaxoSmithKline, Patrick Vallance, has been at the coalface of GSK’s drive set out his intentions to increase assessment agency, the Institute for toward transparency. An important step came in March 2013 when it was the transparency in 2012.17 Quality and Efficiency in Health Care first (and still, at the time of publication, the only) drug company to sign the The former director general of (IQWiG), published a paper in 2013 AllTrials register. the Italian Medicines Agency and showing that unpublished clinical It was Andrew Witty, chief executive of the London based firm, who took a teacher and researcher at the study reports were more than this step, but Vallance, a former academic, has been the main champion of University of California, Berkeley, told twice as likely as publicly available transparency on the ground, while his peers across the industry have not The BMJ that although the new rules sources to provide complete been so welcoming of opening up data to the public. coming into force in 2015 were only information on outcomes relevant to Vallance says, “I know concerns have been raised about greater one “step in a process,” they did “set patients.18 transparency introducing a competitive disadvantage, but I don’t see it. I see a new standard for transparency in This finding helped add a compelling societal need to do it. Historically, industry hasn’t done a good public health and pharmaceutical momentum to the transparency enough job to be transparent.” research and development.” debate, with IQWiG also taking the He believes, despite the delays, bold step of signing the AllTrials “Suddenly, everyone who used to think that the agency “can certainly be petition in the same year.19 proud to have delivered” on its Wieseler was also a coauthor of a it was fine to hide data from clinical promise and has contributed to 2010 study of the available data on trials has realised that it’s not OK” the “general drive towards more the antidepressant reboxetine.20 transparency in Europe.”

SARAH WOLLASTON ILARIA PASSARANI EMILY O’REILLY GLENIS WILLMOTT The Conservative member of the UK The head of Food and Health As the new European Union The UK Labour member of the European parliament for Totnes, Devon, and Department at the European ombudsman investigating poor Parliament and rapporteur for the now chair of the House of Commons Consumer Organisation (BEUC) and and failed administration in EU EU Clinical Trials Regulation fought Health Committee, Sarah Wollaston, a scientific researcher at Maastricht institutions, Emily O’Reilly got both hard to get the transparency agenda a former GP, has been a champion of University, Ilaria Passarani fought the European Medicines Agency into law, citing scandals such as the the transparency debate in the United hard to ensure that the European and the European Parliament to one that emerged in the 1980s over Kingdom. Medicines Agency kept its promise accept that clinical information the anti-arrhythmic agent lorcainide, Wollaston says that commitment on opening up data. in drug trials should not be in which over 100 000 people died to data publication was resisted As the signatory of a strongly considered commercially sensitive unnecessarily because data from an at all stages on the grounds of the worded letter to Rasi last May, when or confidential and also shouldn’t early trial were not published.23 perceived threat to the future of UK it looked as if the agency might water be redacted from the clinical study Willmott says that her fellow MEPs pharmaceutical companies. down its policy on transparency, she reports. did understand the issues but that She said, “I’m delighted that argued passionately that it should Like Passarani, she too was their positions started at complete by repeatedly reintroducing it at not renege on its deal.21 She said concerned at the agency’s delays opposite ends of the spectrum, some every opportunity the government that the agency was honour bound in mid-2014 and put Rasi and the wanting every single piece of raw data finally conceded and that need for to do this so as to “restore public agency under intense pressure to published, while others wanted even transparency is now recognised in our trust and confidence in the Agency go ahead with its plans on data less transparency than the European own legislation, as well as at EU level.” by implementing a meaningful disclosure. Commission had proposed. Wollaston also notes that the proactive publication of data and by In a letter to Rasi, O’Reilly, an She says, “One of my main lack of scientists and former health enabling independent re-analysis.” Irish national and former journalist, achievements was bringing the MEPs professionals in the House of Trust was a major motive for the said that if the agency did not allow from different political groups together Commons “can make it difficult to agency when it eventually pushed its a new transparency policy this and agreeing on a sensible way forward drum up enthusiasm for debates about plans through. would “undermine the fundamental that would guarantee meaningful the importance of evidence based right of public access to documents transparency for clinical trial results.” medicine.” established by EU law.”22

18 3 January 2015 | the bmj PRESCRIBINGANALYSISBRIEFING

Transparency in drug company payments to doctors Rebecca Coombes sums up the new UK system for public disclosure of payments from drug companies to doctors

ince the new year drug companies Once the data pass to the ABPI for publica- in the United Kingdom have begun tion, it will send individuals a statement of the recording any payments they make payments that will be published under their to doctors for certain services, such names. Clinicians then have four weeks to as chairing a meeting, in advance of query this record. They can also opt out at this Splans to disclose the data to the public. This stage, even if they originally signed a contract move echoes similar initiatives in the United with the drug company that paid them. The States and the Netherlands designed to bring company will be notified by the ABPI that the transparency to financial relationships between clinician has withdrawn their permission. doctors, teaching hospitals, and drug compa-

nies. The information gathered over the next GETTY IMAGES What happens when a doctor opts out? 12 months, and in subsequent years, will be signs are encouraging—one drug company has In an ABPI consultation with 1000 healthcare uploaded to a publicly searchable database due already asked 500 healthcare providers to sign professionals, 86% said they were in favour to launch in July 2016.1 new contracts and only two have refused. of public disclosure of payments. “Early indi- The stimulus for this new openness comes After the database launches in July 2016 any- cations are that the initial response from pro- from Europe and has been adopted in a new one will be able to search for payment data by fessionals has been positive. We want this to code by the Association of the British Phar- health professional name, healthcare organi- become the professionally-accepted norm,” maceutical Industry (ABPI), which covers 121 sation, place of work, city, or drug company an ABPI spokesperson told The BMJ. But he (98%) drug companies in the UK. In wider name, or they can download the entire dataset. acknowledged that a ’s decision to opt Europe 33 countries are covered by a new dis- The data are all recorded on a central platform out of disclosure may change a drug company’s closure code agreed by the European Federation on the ABPI website and will be available for at view of “who they work with over time.” of Pharmaceutical Industries and Associations.2 least two years. In law, consent to share the data must be Transparency “is no longer a ‘nice to have’: it “freely given,” so a drug company cannot use is a societal expectation,” said Andrew Powrie- What will be disclosed? threat of withdrawal of payment to induce clini- Smith, speaking as director of ABPI Scotland.3 Any transfer of value to a doctor or healthcare cians to sign. The ABPI will annually publish the This shift is significant because there is no professional either of a “direct” or “indirect” number of people who did not give consent to overarching system in the UK for regulating nature. There is no minimum value unlike, for share payment data and the aggregate amount. sponsorship, payments for expert advice, and example, in the Netherlands, where payments other benefits that doctors and other healthcare of more than €500 (£395; $615) are recorded. What are doctors saying? professionals receive from industry. Even the Indirect payments include donations to char- The president of the Royal College of Physi- new ABPI platform—novel as Initial signs are ity made on a doctor’s behalf cians, Jane Dacre, welcomes the move. “The it is—will only provide a par- and payments to third parties, ABPI database is timely and reflects the grow- tial disclosure of interactions encouraging—one such as drug companies paying ing need for to be open about their with industry. The associa- drug company has someone to organise an event relationships with pharmaceutical compa- tion doesn’t represent medical already asked 500 for a doctor or healthcare pro- nies. This openness is in the best interest of device companies, or private healthcare providers fessional. Two types of direct the patients we serve, and I hope that it will hospital chains, for example, to sign new contracts payment need to be disclosed— become a mandatory responsibility in the near so any payment from these and only two have “events,” which includes future,” she told The BMJ. industries to doctors will not refused registration fees, travel, and Separate moves are under way to compel be recorded. Nor does the ABPI accommodation, and “services doctors to make their own declarations on have the power to compel doctors to make a full and consulting,” which includes, for example, a database led by the profession. A working disclosure of their conflicts of interests, including fees for speaking at and chairing meetings, train- group, including representatives from the shares in commercial companies. The General ing, and advisory board meetings, and expenses. Royal College of Physicians and The BMJ, is set Medical Council does have guidance on conflicts to meet this month to draft a code of conduct of interest but it does not currently oblige doctors Can doctors opt out? governing interactions with industry. A final to declare these on the medical register. Yes, and this option could significantly weaken version will be published in the summer.4 the database’s “brand” as a transparency tool. Attempts have also been made to produce How does the new database work? European and UK data protection legislation is independent registers by which people can Drug companies have, from 1 January, changed a major barrier to full disclosure of payments declare any conflicts of interest (financial or the way they contract with doctors and other to doctors because drug companies must get otherwise). One such effort is Who Pays This healthcare professionals. If an individual an individual’s permission before publishing Doctor? (www.whopaysthisdoctor.org), which agrees to do paid work for a company, they payment data. This is not the case in the US or currently has fewer than 150 doctors registered. should sign a contract granting permission the Netherlands, where separate legislation Rebecca Coombes magazine editor, The BMJ for the payment data to be shared publicly over-rides data privacy in the case of their open [email protected] (although they can back out at any time). In­itial payment databases. Cite this as: BMJ 2015;350:g7748 the bmj | 3 January 2015 19