Clinical trial data for all drugs in current useBMJ 2012; 345 doi: https://doi.org/10.1136/bmj.e7304 (Published 29 October 2012) Cite this as: BMJ 2012;345:e7304
All rapid responses
Dear Dr Godlee,
I understand that during your speech at the Faculty of Pharmaceutical Medicine event you commented that “the ABPI had ignored Ben Goldacre’s book – Bad Pharma – which is shameful”.
I was extremely surprised and disappointed to hear this. As you are aware, the ABPI – and indeed the broader pharmaceutical industry – has been proactively engaged in examining and addressing many of the issues raised by Ben in Bad Pharma in partnership with multiple stakeholders, long before it was published. This has included disclosure of payments, industry funded education and the provision of promotional aids as well as clinical trial data transparency. We share many of Ben’s objectives and indeed I will be debating with him, you, and, among others, Sir Iain Chalmers at the Pharma Times event on 26 February.
The highly successful joint BMJ and ABPI conference that we funded in May last year – where issues on clinical trial design, implementation and dissemination were discussed candidly – was indeed part of our on-going commitment to address industry issues collaboratively, and one we shared with you.
Subsequent to the event, in September 2012 the ABPI Code of Practice was amended to require companies to disclose details of the results of all clinical trials in accordance with the International Federation of Pharmaceutical Manufacturers and Association’s (IFPMA) Joint Position on Disclosure of Clinical Trial Information via Clinical Trial Registries and Databases. This includes that all clinical trial results must be published within one year of marketing authorisation. Companies must also publicly register clinical trials within 21 days of initiation of patient enrolment.
In addition, the ABPI has led a major initiative in partnership with the Ethical Standards in Health and Life Sciences Group (ESHLSG) to ensure all payments to physicians are fully transparent and disclosed. Following aggregate disclosure this year, the ABPI has publicly stated its commitment to introduce disclosure of payments on an individual basis by 2016. Last week a consultation – run by the ESHLSG – was announced to establish how this can be taken forward in partnership with healthcare professionals and other parts of the life science community such as medical devices and diagnostic companies.
I had hoped that last May’s successful joint conference would further our collaboration with you as a leading authority in the medical sector and, that we would continue our dialogue to address issues in partnership. However, despite this we have progressed with our activities, which have included engagement with Dr Goldacre as well as discussions with many other interested stakeholders. I also have a meeting scheduled with Sir Iain Chalmers, following our recent interview on clinical trial data disclosure on the Today Programme.
As we have said in our communication about clinical trial data transparency – which is all publicly available on our website – we support increased transparency for clinical trial results, balanced with the need for patient anonymity and commercial sensitivities, and we support the work of the EMA on how to achieve this.
We remain fully committed to working with all stakeholders to ensure that we continue down our established path. The medicines that we discover are pivotal to human health and it is vital that confidence is maintained with the healthcare professional and the patient.
I very much look forward to receiving your response.
ABPI Chief Executive
Competing interests: No competing interests
We would like to add to the debate on neuraminidase inhibitor (NI) data recently initiated by the BMJ, by highlighting what lower efficacy may mean from a population-wide perspective. If NI agents are less effective than those given in published data, future influenza pandemic responses which rely on such efficacy data may be significantly compromised. Even following the 2009 influenza pandemic, such NI-based interventions remain an untested strategy.
We have used the limited published data on NI efficacy in a series of modelling studies based on a population of ~30,000 in Western Australia (1). These studies aimed to determine the population-level effectiveness of a range of NI-based intervention strategies, on their own and coupled with various social distancing interventions (2).
Recently, we conducted sensitivity analyses on the NI efficacy parameters, both for treatment and prophylaxis. This has allowed us to examine how the potential mitigating effect of NI-based interventions (in terms of illness attack rate (AR) reduction) might be impacted if their efficacy was lower than that suggested in the literature (3, 4, 5). The results indicate that if NI efficacy is lower than claimed, interventions involving NI use show a significant variation in outcome, see Table.
Data in the Table indicate how reduction in NI efficacy from a baseline of 66% for treatment and 85% for prophylaxis alters the AR of an influenza pandemic having a basic reproduction number of 1.8 and an unmitigated AR of 32%. These were the most recent efficacy estimates which we could find, relying on an analysis (3) of two oseltamivir trial studies (4, 5). The left hand column in the Table indicates assumed NI efficacy, as % reduction in infectiousness when used for treatment, and % reduction in susceptibility when used for prophylaxis; the latter being bracketed. We have examined the five NI-based intervention strategies described in the Table under a range of efficacy estimates lower than the baseline. Two examples are given; one having 11% treatment efficacy (and 14% for prophylaxis) and the other with 33% treatment efficacy (and 42% prophylaxis efficacy).
Depending on the strategy applied, lessening NI efficacy from the baseline alters the population-wide effectiveness of interventions differently, depending on whether social distancing interventions are combined with the NI-based interventions or not.
If combined with rigorous and sustained social distancing (continuous school closure and reduced workplace and community contact; right hand column) lower NI efficacy does not alter the AR reduction achieved (from 32% to 6%); the significant reduction is due to the rigorous social distancing applied. However, weaker social distancing involving only 2 weeks of school closure when combined with NI interventions (3nd column from right) gives a strategy much more reliant on the mitigating effect of the NI antivirals, resulting in ARs of 20% and 23% with the two efficacy reductions, compared to the high efficacy baseline of a 15% AR. All other intervention strategies presented in the Table also become less effective due to reduced NI efficacy, as illustrated by proportionate increases in their AR.
Experience of NI-based interventions used during the 2009 pandemic give little guidance as to their effectiveness. If the NI efficacy is lower than claimed, detailed modelling studies such as (1, 2) may be used to give guidance on how this negatively impacts on pandemic preparedness plans which are partially reliant on NI-based strategies. If a future influenza pandemic has higher morbidity and mortality rates than occurred with the “mild” H1N1 2009 pandemic, investment in creating and maintaining NI stockpiles may be viewed as costly and ineffective.
1. Milne GJ, Kelso JK, Kelly HA, Huband ST, McVernon J (2008) A small community model for the transmission of infectious diseases: comparison of school closure as an intervention in the individual-based models of an influenza pandemic. PLoS ONE 3: e4005.
2. Kelso JK, Halder N, Milne GJ (2010) The impact of case diagnosis coverage and diagnosis delays on the effectiveness of antiviral strategies in mitigating pandemic influenza A/H1N1 2009 PLoS ONE 5: e13797.
3. Yang Y, Longini IM, Jr., Halloran ME. Design and evaluation of prophylactic interventions using infectious disease incidence data from close contact groups. Appl. Statist. 2006;55:317-30.
4. Hayden FG, Belshe R, Villanueva C, Lanno R, Hughes C, Small I, et al. Management of influenza in households: a prospective, randomized comparison of oseltamivir treatment with or without postexposure prophylaxis. JID 2004;189(3):440-49.
5. Welliver R, Monto AS, Carewicz O, Schatteman E, Hassman M, Hedrick J, et al. Effectiveness of oseltamivir in preventing influenza in household contacts. JAMA 2001;285(6):748-54.
Competing interests: No competing interests
We note the British Medical Journal’s pledge to only publish results of clinical trials where access to patient level data is made available on reasonable request from January 2013. We are keen to understand in more detail how decisions to publish clinical research data will be applied.
ABPI member companies have supported clinical trial registries and results databanks, including the legislation that led to the development of databases such as ClinicalTrials.gov in the United States. The purpose of such transparency is to provide useful information about clinical trials for patients and healthcare professionals. ABPI member companies are also committed to disclosing clinical trial information at the time of patient enrolment, drug approval, and for medicines whose research programs have been discontinued. These and other efforts to enhance the transparency of clinical research are reflected in the Joint Position on the Disclosure of Clinical Trial Information via Clinical Trial Registries and Databases (2009) and the Joint Position on the Publication of Clinical Trial Results in the Scientific Literature (2010) issued by the IFPMA, the European Federation of Pharmaceutical Industries and Associations (EFPIA), the Japanese Pharmaceutical Manufacturers Association (JPMA) and the Pharmaceutical Research and Manufacturers of America (PhRMA) and also the 2009 revisions to PhRMA’s Principles on Conduct of Clinical Trials and Communication of Clinical Trial Results. Indeed, many companies are continuing at pace to further broaden the modes of access to clinical trial data.
While supporting enhanced transparency of clinical research and safety information, it must be balanced with the need to ensure that disclosure policies protect patients’ personal data, intellectual property rights and confidential commercial information associated with innovative research and development. The public health is not well served by dis-incentivising research based biopharmaceutical companies and commercial organisations from other life science sectors from making the substantial investments and shouldering the risks that are necessary to develop new innovative medicines.
While the publication criteria for clinical research papers is a matter for the BMJ, having partnered BMJ Group in the Innovation in research methods and dissemination - ensuring best practice conference (May 2012) we would have appreciated the opportunity to discuss BMJ’s new position in advance of publication. In particular to understand:
• If the same criteria will be applied to studies by commercial organisations, academia and charities across diagnostics, devices and biopharmaceuticals?
• Does the pledge apply to all studies, (including those in healthy subjects), irrespective of where they are conducted or only those studies with centres in the UK?
• Do the criteria apply prospectively only to studies conducted from January 2013 or retrospectively on all studies?
• How is reasonable request being defined by the BMJ?
• Who will assess that a request is reasonable? How will this be determined in advance of publication by BMJ?
• If BMJ supports the use of confidentiality agreements where access to patient level data is granted following a reasonable request in order to protect commercial in confidence information?
• If the criteria will be applied to meta-analysis?
• What steps does BMJ envisage will be required in respect of protected personal data of patients, employees and other parties involved in a clinical study?
From a wider perspective, clinical research is truly a global process in bio-pharmaceuticals and across the wider research community. Our view is that work to continue the journey to greater access to trial data needs to be approached at a global level and in partnership with all relevant stakeholders. It is an iterative process which has been advancing rapidly over recent years and continues to do so.
In that context, we are keen to understand if your signature on the letter to the Times advocating changes to UK legislation on the issue of disclosure indicated your personal perspective or if BMJ views unilateral action in the UK beyond globally agreed positions on disclosure as the right approach for clinical research in the UK? I am sure the ABPI and the BMJ are united in their desire to continue to see the UK as a leading centre for clinical research. Part of that equation is ensuring that we play an active role in the international journey towards greater clinical trial data transparency whilst ensuring our regulatory environment is aligned with European and Global standards.
I look forward to hearing your views and hope that our organisations can continue to keep an open dialogue on the issue.
 See PhRMA, Principles on Conduct of Clinical Trials and Communication of Clinical Trial Results (2011), available at http://www.phrma.org/sites/default/files/105/042009_clinical_trial_princ....
Competing interests: No competing interests
In an era of evidence based medicine clinical trials hold the key to validate or negate a medical intervention, whatsoever is its nature. However, clinical trials have been under increased scrutiny in the recent past due to concerns over violations of ethical concerns and research related injury. Thus it has become imperative on the part of those at the helm of affairs of clinical trials (drug makers as well as non-governmental organisations and governments which funds such studies) to make available the full and complete information of clinical trial available in public domain.
Making clinical trial data available right from the initial phase’s right till its end in the public domain not only enhances the transparency and validity of the trial it is also beneficial to those running the trial itself. It helps to improve the quality of clinical trials by making it possible to identify potential problems as well as allow health care practitioners to identify researchers related to their field.
I would like to draw attention to the fact that WHO's Handbook of Good Clinical Research  practice clearly states that "the report should be complete, timely, well-organized, free from ambiguity, and easy to review." However as pointed out by Dr Godlee in her editorial  misrepresentation of clinical trial data has been persistently and systematically been done over a decade. However the trend of partial presentation or “half truths” has been picking up more in the last decade. What is more appalling is the trend of regulating agencies responsible for approving products being lackadaisical in the face of new evidence being made available. The Cochrane Collaboration review on neuraminidase inhibitor’s (NI) in July 2006 mentions  that “because of their low effectiveness, NIs should not be used in routine seasonal influenza control. In a serious epidemic or pandemic, NIs should be used with other public health measures.” Yet more than three years after that in 2009 governments worldwide, on the face of a flu pandemic, following WHO guidelines stockpiled the drug. Why was the result of the 2006 review by Cochrane Collaboration not considered when such guidelines were issued by the WHO in 2006-when the world was in grip of a swine flu pandemic?
Why do drug regulating/guideline formulating agencies from across the world repeatedly fail to make note of conflicts of interest when preparing such guidelines which affect millions? Also since 2009 when the case of missing data came into light it is abundatly clear that it goes against a principle laid down by the WHO in its Handbook of Good Clinical Research?
Why is it that the BMJ or the Cochrane have to fight for such missing data when it is clearly the responsibility of the drug regulating agencies to access the complete set of data from all trials both before and after giving approvals?
With questions on effectiveness of interventions and their safety being raised every now and then partial presentation of clinical trial data is akin to going against the golden principle of medical ethics - “Do no Harm.”
1.Handbook for good clinical research practice (GCP) : guidance for implementation.World Health Organisation, 2005: 17 Available online at : http://whqlibdoc.who.int/publications/2005/924159392X_eng.pdf (Acessed 10th November 2012 )
2.Godlee F.Clinical trial data for all drugs in current use.BMJ2012;345:e7304
3.Payne D.Tamiflu: the battle for secret drug data.BMJ2012;345:e7303
Note : The views /opinion expressed by author are personal and not necessarily subscribed by the Your Health of the Indian Medical Association or the Indian Medical Association itself.
Competing interests: No competing interests
The BMJ’s initiative requiring sponsors to make relevant anonymised individual patient data (IPD) available as a precondition for publication is highly commendable, as is GlaxoSmithKline’s announcement to grant access to such data under certain conditions. However, in our view this would at best solve the issue of reporting bias for individual trials or individual drugs, but not for clinical research as a whole. Conclusions of systematic reviews regarding a drug’s benefits and harms, especially compared to treatment alternatives, could thus still be incorrect.
In our opinion, comprehensive access to trial data on drugs marketed in Europe (at least to data submitted to the European Medicines Agency, EMA) can only be ensured by the routine release of all clinical study reports (CSRs) by EMA once a drug has been approved. EMA previously denied access to clinical trial documents to systematic reviewers [1, 2]. But the agency has since changed its position: In 2010 it introduced a new policy, including the release of documents submitted as part of marketing authorization applications (e.g. CSRs) after procedures concerning a drug had been finalized , and has since published an update and accompanying guidance on the handling of commercially confidential information and personal data [4, 5]. However, the format of the data to be released is still under discussion (“conventional” CSRs following the International Conference on Harmonisation E3 guideline  and generally including anonymised IPD listings as PDF files or CSRs supplemented by anonymised IPD stored in electronic databases [2, 7]). In addition, EMA plans to make the data available only on request.
We suggest the following approach:
Firstly, CSRs of new drugs should be made routinely available on the EMA website after marketing authorization. This could be done by including a link to the CSRs of studies considered in the authorization process and listed in the European Public Assessment Reports (EPAR). Further CSRs submitted to EMA later in the drug’s life cycle could be added consecutively.
Secondly, as biased evidence is not only a problem of new drugs, but also of older ones still widely used in clinical practice, CSRs of older drugs submitted during previous approval processes should also be made available in a central repository at EMA to complete the evidence base. Pharmaceutical companies could also release these documents, thus underlining their commitment to transparency.
Thirdly, anonymised IPD stored in electronic databases could be provided for the above CSRs. As implementation of this step could be time consuming, this should not be merged with the issue of releasing CSRs, but should instead be discussed in parallel .
The first two steps could be taken quickly, providing systematic reviewers with invaluable information: CSRs represent the most comprehensive source for the evaluation of a clinical trial , and the addition of data from CSRs potentially changes the conclusions of systematic reviews based solely on journal publications and trial registry entries [9, 10]. Speedy access to CSRs is therefore required and would be a milestone for transparency in clinical research.
Of course the above comments do not only apply to Europe, which has been lagging behind the United States in terms of data transparency. The US Food and Drug Administration (FDA) Amendments Act of 2007 , which among other things, prescribes the online posting of methods and results of trials of drugs subject to FDA regulation, is currently the most comprehensive mandatory provision on access to clinical trials. The FDA itself has also taken substantial steps to increase data transparency (e.g.). Several researchers have used clinical trial documents available on the FDA website or unpublished data provided by the FDA to assess the effects of reporting bias [13-17]. EMA and the FDA already cooperate on a regular basis; a joint initiative on the issue of access to CSRs, possibly also involving other regulatory agencies, suggests itself and would ensure access to a substantial amount of the worldwide evidence on clinical trials.
1. Gotzsche PC, Jorgensen AW: Opening up data at the European Medicines Agency. Bmj 2011, 342:d2686.
2. Wieseler B, McGauran N, Kerekes MF, Kaiser T: Access to regulatory data from the European Medicines Agency: the times they are a-changing. Systematic reviews 2012, 1(1):50.
3. European Medicines Agency widens public access to documents. Press release, 30 Nov 2010 [www.ema.europa.eu/ema/index.jsp?curl=pages/news_and_events/news/2010/11/...
4. Principles to be applied for the implementation of the HMA/EMA Guidance on the identification of CCI and PPD in MA Applications [http://www.ema.europa.eu/docs/en_GB/document_library/Other/2012/03/WC500...
5. HMA/EMA Guidance document on the identification of commercially confidential information and personal data within the structure of the Marketing Authorisation (MA) Application - Release of information after the granting of a Marketing Authorisation [http://www.hma.eu/fileadmin/dateien/HMA_joint/02-_HMA_Topics/07-Transpar...
6. Guideline for Industry - Structure and Content of Clinical Study Reports [http://www.fda.gov/downloads/regulatoryinformation/guidances/ucm129456.pdf]
7. Eichler HG., Abadie E, Breckenridge A, Leufkens H, Rasi H: Open Clinical Trial Data for All? A View from Regulators. PloS Medicine 2012, 9 (4) e1001202(4 e1001202).
8. Wieseler B, Kerekes MF, Vervoelgyi V, McGauran N, Kaiser T: Impact of document type on reporting quality of clinical drug trials: a comparison of registry reports, clinical study reports, and journal publications. Bmj 2012, 344:d8141.
9. Eyding D, Lelgemann M, Grouven U, Harter M, Kromp M, Kaiser T, Kerekes MF, Gerken M, Wieseler B: Reboxetine for acute treatment of major depression: systematic review and meta-analysis of published and unpublished placebo and selective serotonin reuptake inhibitor controlled trials. Bmj 2010, 341:c4737.
10. Doshi P, Jefferson T, Del Mar C: The Imperative to Share Clinical Study Reports: Recommendations from the Tamiflu Experience PloS Medicine 2012, 9 (4) e1001201(4 e1001201).
11. FDA Amendments Act (FDAAA) of 2007, public law no. 110-85 § 801 [http://frwebgate.access.gpo.gov/cgi-bin/getdoc.cgi?dbname=110_cong_publi...
12. Asamoah AK, Sharfstein JM: Transparency at the Food and Drug Administration. N Engl J Med 2010, 362(25):2341-2343.
13. Turner EH, Knoepflmacher D, Shapley L: Publication bias in antipsychotic trials: an analysis of efficacy comparing the published literature to the US Food and Drug Administration database. PLoS Med 2012, 9(3):e1001189.
14. Turner EH, Matthews AM, Linardatos E, Tell RA, Rosenthal R: Selective publication of antidepressant trials and its influence on apparent efficacy. N Engl J Med 2008, 358(3):252-260.
15. Rising K, Bacchetti P, Bero L: Reporting bias in drug trials submitted to the Food and Drug Administration: review of publication and presentation. PLoS Med 2008, 5(11):e217.
16. Lee K, Bacchetti P, Sim I: Publication of clinical trials supporting successful new drug applications: a literature analysis. PLoS Med 2008, 5(9):e191.
17. Hart B, Lundh A, Bero L: Effect of reporting bias on meta-analyses of drug trials: reanalysis of meta-analyses. Bmj 2012, 344:d7202.
Competing interests: Non-financial competing interests: All authors are employees of the Institute for Quality and Efficiency in Health Care, Cologne, Germany. In order to produce unbiased health technology assessment reports, the Institute depends on access to all of the relevant data on the topic under investigation. The authors therefore support public access to clinical study reports.
Dear Dr Godlee
I read your editorial with great interest.
The Health Research Authority is a Special Health Authority established on 1 December 2011. We promote and protect the interests of patients and the public in order to support both their confidence and participation in health research, and improvements in the nation’s health. We do this by ensuring that research involving members of the public is ethically reviewed and approved, that they are provided with the information they need to help them decide whether they wish to take part, and that their opportunity to do so is maximised by simplifying the processes by which high quality research is assessed.
We recognise that failure to publish the results of research not only presents a significant risk to patient safety, but also constitutes research misconduct.
The National Research Ethics Service is part of the HRA, and research ethics committee review already includes consideration of applicants’ plans to register their research (on clinicaltrials.gov for example) and publish research. The Integrated Research Application System (IRAS) – the single online system for applying for permissions and approvals for health research, including clinical trials – asks whether the research will be registered on a public database, and how researchers intend to report and disseminate the results.
The final report on the research should be submitted to the main REC within one year of its conclusion. This applies to all research, including Clinical Trials of Investigational Medicinal Products.
NRES conducted work in 2008 showing that many studies submitted to RECs were not, at the time, registered. It took the initiative to publish summaries of ethically approved studies submitted by researchers online. We recognise that the quality of these summaries is currently variable, but believe that NRES-approved research must be published online to promote transparency, encourage registration and support its publication. We believe that publication will also have the beneficial effect of encouraging participation in health research.
The decision by an ethics committee to give a favourable opinion will include consideration of plans for registration, publication and further use of data and tissue. Historically NRES has not been able to follow up on these intentions. The HRA now intends to monitor compliance, and identify researchers, funders and institutions who are not registering or publishing ethically approved research.
We will do this by routinely reviewing the final report to identify any failure to comply with researchers' declared intentions to publish, register or make study information or tissue available one year after the final report. We are currently exploring how best to implement these improvements and safeguards, and expect to establish a new system early in 2013.
Whilst publication can be assessed administratively, any consideration of failure to publish will need review by officers of ethics committees, who will determine whether the reasons not to publish are acceptable, or if not, whether the issue needs referring to the HRA Board, potentially with advice from the HRA National Research Ethics Advisors’ Panel. The HRA does not have powers to investigate research misconduct but it can request that the Sponsors of research investigate concerns identified by HRA or brought to its attention. We will at the very least make ethics committees aware when an applicant has not fulfilled the promise on previous applications, and it will be for the ethics committees to consider how this may influence their opinion when considering new applications from the same sponsor.
The HRA is committed to making it easier to do good quality ethical research in the UK, and sees great potential in gathering evidence of good conduct – including publication – that would enable a lighter touch when considering the suitability of researchers and sponsors who can show they have fully delivered on previously declared intentions.
Competing interests: No competing interests
Dear Dr Godlee
Congratulations on your editorial, which is good news for patient health.
Writing about GSK’s announcement, you say it will be important to know how many requests will be turned down and for what reasons.
If a pharmaceutical firm accepts that it should be making its raw data available to the scientific community, I think that it should also be the duty of our health agencies.
I have asked the EMA in the past for such data but did not get it from them (yet). Here is the story.
At the end of November 2010, after the publication by the EMA of the document “Output of the European Medicines Agency policy on access to documents related to medicinal products for human and veterinary use”, I decided to ask the EMA for some documents (CTD, raw data) about a product approved at the end of the last century. I received some pages without raw data. I eventually lodged a complaint with the European Ombudsman in April 2011. Unfortunately, the documents I requested “could not be retrieved from the archives”.
Since then, every 6 months, I receive a letter from the European Ombudsman telling me that they are doing their best, and I believe them.
Nonetheless, somewhat upset by the lack of transparency, I decided to publish my doubts on the product which had been approved on the basis of one phase III trial. The publication can be seen here: http://www.la-press.com/redirect_file.php?fileId=4504&filename=Approved-...
As you might guess, I believe that this trial might suffer from a bias.
The EMA and other health agencies, in order to avoid publication bias, should collate information concerning trials performed on drugs, make available that information, and also make publicly available all the data that helps them take any decision.
If the public and the experts have the same information, perhaps it will also minimize the risk of conflicts of interest.
Competing interests: cited in my paper
There is no doubt that medical research is crucial for improving health care; doing so, it contributes to save lives and to improve people’s quality of life. However, promoting integrity in the broad domain of medical research is not simple. Besides the issue of misreported data involving drug industries1,2, other problems are frequent in the research field: fraud and misconduct are not rare3. In this context, the recent scandal involving the Japanese researcher Hisashi Moriguchi4,5 is an example of fraud in medical research. Whilst the increased vigilance of editors must be highlighted, this impacts “palpable” frauds, such as plagiarism and duplicate publication. Yet, when more silent ethical issues are involved in the reporting of medical research, no vigilance system is currently effective.
This is the case for authorship criteria, for example. The veracity of the authorship disclosure is impossible to be verified. The use of the three authorship criteria of the International Committee of Medical Journal Editors (ICMJE)6, as well as the identification of the researcher “who is responsible for the integrity of the work as a whole”6, are a beginning for standardising contributorship, but they are certainly not infallible. It cannot be stressed too much that substantial contribution for each single article is a mandatory criterion for authorship.
The responsibility of “deciding” who meets authorship criteria must be shared among the corresponding author, the other co-authors and journals’ editors (even though the ICMJE indicates that “it is not the role of editors to make authorship/contributorship decisions or to arbitrate conflicts related to authorship”6). In this context, it would be useful, for example, that a broad discussion among editors and researchers help the ICMJE to re-evaluate its uniform requirements. As part of articles’ submission, Journals should systematically contact all co-authors to inform them about the ICMJE authorship criteria and give them the link of the ICMJE webpage for further details6. Moreover, Journals should ask all co-authors to supply detailed information on their real contributions to the article being submitted and how these contributions were taken into account (or not) in the submitted version: e.g., “after reading the paper drafted by the author AA, I, co-author BB, suggested the incorporation of the variable VI in the analysis. All the other authors agreed with this change, and the analysis was modified accordingly. I also suggested discussing variable VI in the way we have done in the paper. Finally, I suggested adding a paragraph discussing the lack of significant associations between variable VV and the outcome, but on the sake of word limitations this was not integrated in the submitted version”). Such a procedure, whilst more burdensome for co-authors (and for editors), would enable them to describe and voice their applied competency.
These words aim at stimulating the broad debate on promoting integrity in medical research. As for the drug industry issue1,2, we need to seize this moment to change people’s culture and practice about authorship.
1. Godlee F. Clinical trial data for all drugs in current use. BMJ 2012;345:e7304
2. Jackson T. Open data: seize the moment. BMJ 2012;345:e7332
3. Fang FC, Steen RG, Casadevall A. Misconduct accounts for the majority of retracted scientific publications. Proc Natl Acad Sci U S A. 2012 Oct 16;109(42):17028-33.
4. Promoting research integrity: a new global effort. Lancet 2012;380(9852):1445. [Editorial]
5. Cyranoski D. Stem-cell fraud hits febrile field. Nature 2012 ; 490(7420):321.
Competing interests: No competing interests
This is a topic that requires a revolution in practices. Currently when a manufacturer submits data to the regulator, they are required to provide all completed trials. The regulator may subsequently ask for further data, but once satisfied gives approval or denial for marketing. The data then provided to the public and medical profession is a subset of the actual data.
My suggestion is that firstly once approval is given it should be conditional on the provided data being freely accessible to the community for analysis and debate. This would enable other interested parties to test the manufacturer's conjectures of efficacy and safety. This would have the benefit of the data being vigorously perused by academics, competing manufacturers and other interested parties. Should their have been any faults in the manufacturer's and regulator's analyses, they would potentially be found sooner with consequent community benefits.
The second area that requires urgent attention is the nature of regulatory approvals. The emphasis is on safety and efficacy usually requiring at least two pivotal trials, often placebo controlled and not against competitive agents. For example in the field of osteoporosis these have all been placebo controlled and we thus do not have any data on comparative efficacy between treatments often of differing mechanisms of action. The solution, I believe, is for the major regulators to provide conditional approval for new medications, requiring the company to sponsor trial(s) to answer these and other important scientific questions. The manufacturer thus has the ability to market their drug and start to recoup their investment, whilst the public and medical community obtains further data as to the role and safety of the new medication.
Competing interests: No competing interests
Similar to drug companies, developers of complex interventions are also liable to conflicts of interest that may bias results. We thus support the BMJ’s decision to “publish only where there is a commitment to make the relevant anonymised patient level data available on reasonable request”, and we make suggestions that apply across all intervention fields.
Following other disciplines, for example economics [see 1], we hope that the BMJ considers extending such requirements to include statistical syntax. Choices in analysis methods can potentially influence findings in various ways, particularly for epidemiological studies and intention-to-treat analyses that employ advanced imputation methods.
Secondly, complex interventions  have particular reporting needs for both the intervention and control conditions in randomised trials, which are essential for replication . In addition to intervention design, it is also valuable to report how it was actually delivered, how participants complied, and the context of implementation.
As standard journal article word limits currently hinder such reporting, the BMJ could help address this issue by e.g. requiring an online supplement to every trial publication that fully describes the intervention(s). The health community could also benefit from an intervention registry, similar in format to clinicaltrials.gov, requiring authors to deposit replicable descriptions of interventions and report the id number in the publication.
In our opinion, additional requirements such as these would improve both transparency and replication.
TFS, SG, PM
 The American Economic Review: Data Availability Policy: http://www.aeaweb.org/aer/data.php, accessed 1st November 2012.
 Craig, P., et al. Developing and evaluating complex interventions: The new Medical Research Council guidance. BMJ, 2008. 337: p. 979-983.
 Glasziou, P., et al. Taking healthcare interventions from trial to practice. BMJ, 2010. 341: p. 384-387.
Competing interests: The authors work in the development of standards for reporting complex interventions, including work on implementation fidelity. Specifically SG & PM are working on a CONSORT extension for complex social interventions.