Darren B. Taichman, MD, PhD; Joyce Backus, MSLS; Christopher Baethge, MD; Howard Bauchner, MD; Peter W. de Leeuw, MD; Jeffrey M. Drazen, MD; John Fletcher, MB, BChir, MPH; Frank A. Frizelle, MBChB, FRACS; Trish Groves, MBBS, MRCPsych; Abraham Haileamlak, MD; Astrid James, MBBS; Christine Laine, MD, MPH; Larry Peiperl, MD; Anja Pinborg, MD; Peush Sahni, MBBS, MS, PhD; Sinan Wu, MD
Feedback may be posted at www.icmje.org by 18 April 2016.
Note: This editorial is being published simultaneously in Annals of Internal Medicine, British Medical Journal, Canadian Medical Association Journal, Chinese Medical Journal, Deutsches Ärzteblatt (German Medical Journal), Ethiopian Journal of Health Sciences, JAMA (Journal of the American Medical Association), Nederlands Tijdschrift voor Geneeskunde (The Dutch Medical Journal), New England Journal of Medicine, New Zealand Medical Journal, PLOS Medicine, Revista Médica de Chile, The Lancet, and Ugeskrift for Laeger (Danish Medical Journal).
Disclaimer: Dr. Sahni's affiliation as representative and past president of the World Association of Medical Editors (WAME) does not imply endorsement by WAME member journals that are not part of the ICMJE.
Disclosures: Disclosures can be viewed at www.acponline.org/authors/icmje/ConflictOfInterestForms.do?msNum=M15-2928.
Corresponding Author: Darren B. Taichman, MD, PhD, American College of Physicians, 190 N. Independence Mall West, Philadelphia, PA 19106; e-mail, email@example.com.
Paul Dendale (MD;, PhD), Ines Frederix (MD), Dominique Hansen (PhD, FESC)
REVAL – Rehabilitation Research Centre, BIOMED- Biomedical Research Centre, Faculty of Medicine and Life Sciences, Hasselt University, Diepenbeek, and Jessa Hospital, Heart Centre Hasselt, Hasselt, Be
February 2, 2016
Reply to Editorial ‘Sharing Clinical Trial Data: A Proposal From the International Committee of Medical Journal Editors’ (Ann Intern Med. Published online 20 January 2016 doi:10.7326/M15-2928)
In this Letter to the Editor, we would like to reply to the recently published proposal of the International Committee of Medical Journal Editors to share clinical trial data.Even though we applaud the Committee of Medical Journal Editors in its endeavor to minimize the risk for publishing biased data and scientific misconduct, some legal, ethical and practical concerns are raised when clinical trial data sharing becomes mandatory to be able to publish articles in highly ranked medical journals.For example, according to European legislation1 it is stated that (European) Member States shall provide that personal data must be collected for specified, explicit and legitimate purposes, and not further processed in a way incompatible with those purposes, except when each member state puts sufficient safeguard in place. This law implies, at least in Europe, that the sharing of clinical trial data is illegal when it is without a priori verification and stipulation where the data will be stored, with whom and when they will be shared and for what legitimate purposes, and in explicit agreement with the data generator. Obviously, anonymizing data is imperative but several studies have shown that, whenever you have access to sufficient databases, any database containing sufficient data can be de-anonymized easily.Also, some ethical issues remain to be resolved. It is proposed to authors to share their data, no later than 6 months after publication, to others. What is the definition of ‘others’? Many clinical studies are financially supported by governments, patient groups, medical societies, legates,…all financial resources coming from the community, with lots of effort of researchers to obtain these grants and after long hard work in laboratories/hospitals. Is it then ethically justified that corporations can use those data to optimize their devices/products, or develop new devices/products, and profit from it, without investments or effort? In line with this reasoning, how can piracy of data be prevented? In this respect, the public sharing of research data may stimulate dishonest competition. It is also important to define what credits should be given to investigators collecting and sharing data. One credit may be to become co-author on any publication that results from such data sharing and after written approval of the investigator. It is clear that data sharing is an endeavor that will affect the fabric of trials and how data are used…References1. Directive 95/46/EC of European Parliament and of Council of 24 October 1995, on the protection of individuals with regard to the processing of personal data and on free movement of such data.
Sunder M. Lal, M.D.
Boone Hospital Center, Columbia, MO
Response to the article: Sharing Clinical Trial Data
The International Committee of Medical Journal Editors (ICMJE) considers sharing research data generated from clinical trials to be an ethical obligation, and this has been highlighted in this editorial (1-3). Perhaps, the issue of data sharing wouldn’t have become relevant, if the clinical researchers had pursued long term follow up studies to compare the safety and efficacy of active intervention. Major clinical research centers, and the pharmaceutical industries have the financial resources to conduct such studies. Therefore, the research community could consider including this a requirement for registration with Clinical Trials.gov, and for publication in scientific journals. In order to avoid mushrooming of a new industry of secondary investigator(s) some of them may be more interested in seeking access to the individual patient data to further their academic credentials, the journals could accept more letters to the editors for publication. If space in the printed journal is a problem then the letters could be published online. Otherwise, the indiscriminate access to this research data repository may result in churning out plethora of articles on lines of meta-analysis. In the recent past, several articles have highlighted the pros and cons of reforms in clinical research regulation and sharing of clinical trial data (1-5). However, those articles don’t comment on the ownership of the research data. Are data the sole property of a pharmaceutical company and/or the academic institution where the clinical research was conducted by an individual investigator? Do solo investigators have any right to access, and publish the research data from the Institutional Review Board (IRB) approved projects after moving to another institution? Specifically, the IRB'S at a teaching hospital and a private dialysis unit have denied/ignored my repeated requests for access to long term (10-20Yr) follow up data for publication. Furthermore, they have ignored the submitted authorization letters from the survivors of deceased Medicare recipients. Can the institution(s) override a patient and/or their surviving family member’s authorization to release the relevant information? In order to access research data, the academic hierarchy at the institution recommended inclusion of a current faculty member even though he/she had no role in the conduct of that research. Is this not an act of academic dishonesty and violation of the authorship requirement policy of Journals (6)? Currently, neither the scientific journals nor the Office of Human Research Protections has a position on this subject. However, the latter agency suggested hiring a private attorney.References:(1) Taichman BD, Backus J, Baethge C, Bauchner H, de Leeuw PW, Drazen JM et al: Sharing Clinical Trial Data: A proposal from the International Committee of Medical Journal Editors. Ann Intern Med. Published online 20January,2016.doi:10.7326/M15-2928.(2) Institute of Medicine. Discussion framework for clinical trial data sharing: guiding principles, elements, and activities. January 22, 2014 (http://www.iom.edu/Reports/2014/Discussion-Framework-for-Clinical-Trial-Data-Sharing.aspx).(3) Eichler HG, Pétavy F, Pignatti F, Rasi G. Access to Patient-Level Trial Data — A Boon to Drug Developers. N Engl J Med 2013; 369:1577-1579.(4) Emanuel EJ. Reform of clinical research regulations, Finally. N Engl J Med 2015; 373:2296-2299.(5) Hudson KL, Collins FC. Bringing the Common Rule into the 21st Century. N Engl J Med 2015; 373:2293-2296(6) International Committee of Medical Journal Editors. Uniform requirements for manuscripts submitted to biomedical journals: writing and editing for biomedical publication (http://www.icmje.org/urm_main.html
Richard J Hayes, on behalf of colleagues
London School of Hygiene & Tropical Medicine
April 18, 2016
Response to Editorial “Sharing Clinical Trial Data – a Proposal from the International Committee of Medical Journal Editors”
We welcome the move towards greater sharing of data from trials but have concerns about unintended consequences of the proposal, particularly in community-based trials in low and middle income countries (LMIC). Our main concern is the short 6-month period proposed before data should be shared. Our studies often take many years of intensive fieldwork before the primary results are available. Published output during the trial is often limited to maintain blinding, and the careers of the researchers are critically dependent on post-trial outputs. The team focuses its effort on analysing and presenting the primary findings at the earliest possible date, especially when they are of major policy relevance. Following this, there will usually be a series of pre-planned further analyses. These studies – which often involve tens of thousands of research participants – will generally be conducted by teams which may include scientists from institutions where research capacity is being built. The latter will often take responsibility for preparing the secondary trial outputs, and this is important for their career development. If data are shared openly within 6 months of the primary publication, other groups (e.g. from well-resourced HIC centres) may easily outpace the work of these scientists - further exacerbating the asymmetry of opportunity between HIC and LMIC researchers.These concerns might be minimised if a 12 month period of exclusive use were adopted. Particularly for large studies, this may be a proportionate arrangement that protects the legitimate interests of researchers at the same time as ensuring that data can be shared without undue delay. If the main concern of the ICMJE editors is over the transparency of findings – and the ability to carry out replication analyses to check that data have been correctly analysed and reported - we feel that this could be ensured in other ways. For example, there might be a requirement for investigators to share the data for replication on a confidential basis at an earlier stage.Those advocating for greater data sharing correctly point out that we have a duty to the study participants to maximise the value of the data that they have voluntarily provided. But we must make sure that the proposed requirements do not make it so unattractive and unrewarding for researchers to carry out challenging long-term primary research that such studies are not carried out. This would be a grave disservice to the public.A full copy of our letter and signatories is on http://teg.lshtm.ac.uk/files/2014/01/Response-to-ICMJE.pdf
Alain Braillon, MD
University Hospital, Amiens France
April 22, 2016
Sharing data from clinical trials … whether published or not
The Journal must be commended for supporting the International Committee of Medical Journal Editors (ICMJE) proposal: “authors of manuscripts considered for publication in ICMJE members’ journals must agree to share de-identified individual patient data no later than six months after publication”.(1)Nevertheless, the problem of unpublished trials cannot be ignored as highlighted by a French experience. Initial results from controlled randomized trials suggesting a benefit of baclofen in alcohol use disorders (AUD) only included a small number of patients and were not confirmed in further trials. This, as the lack of pre-clinical pre-requisite did not preclude a skyrocking off label use. Presently in France, more than 100,000 patients with AUD are prescribed baclofen off label. Worst, only 6% are registered in the “Temporary Recommendations for Use” (RTUs) specifically enforced by the French Medicine Agency which requires data collection for patient monitoring.(2) Indeed, off-label drug use is usually not supported by sound scientific evidence, hinders the development of evidence-based medicine and therapeutic innovation, is costly to healthcare system, and exposes patients to unnecessary harm, including mortality, for uncertain benefit.(3-5) The French off label use of baclofen is even more alarming: extremely high doses (145 ± 75 mg/day, two or three fold higher than in other countries) while the marketing authorization for spasticity forbids the use of more than 40 mg/day and warns against associated alcohol use.(6) However, French prescribers bypass the specific framework aimed to protect the patients. A well-designed French trial with more than 300 patients (NCT01738282) was finalized in December 2014 (final data collection date for primary outcome measure). The sponsor, Ethypharm, the company marketing baclofen, has not released results yet. In June 2015 Ethypharm claimed it had bought the results of a trial funded by the Paris Public Hospitals Authority (NCT01604330) with 323 patients (Final data collection date for primary outcome: September 2014. Study completion date:October 2015).(http://www.ethypharm.fr/wp-content/uploads/2015/07/Communiqu%C3%A9_accord_APHP-Ethypharm_20150611.pdf) Again no results released yet, even in French meetings. Only claims stating the results will be presented in September 2016 during a meeting in Berlin with a marketing authorization planned for 2017.(http://www.addictaide.fr/actualite/congres-isbra-esbra-berlin-2-5-septembre-2016/) Meanwhile, in October 2015 the French private-equity firm Astorg Partners explored the sale of its company Ethypharm.( http://www.bloomberg.com/news/articles/2015-10-07/france-s-astorg-said-to-weigh-sale-of-pharma-business-ethypharm). Weird isn’t it?Competing interests: I was a late co-investigator for NCT01738282 and did not include patients1 Taichman DB, Backus J, Baethge C et al. Sharing clinical trial data: A proposal from the International Committee of Medical Journal Editors. Ann Intern Med 2016;164:505-6.2 Degrassat-Théas A, Bocquet F, Sinègre M, Peigné J, Paubel P. The "Temporary Recommendations for Use": A dual-purpose regulatory framework for off-label drug use in France. Health Policy. 2015;119:1399-405.3 Chiu Y, Bero L, Hessol NA, Lexchin J, Harrington C. A literature review of clinical outcomes associated with antipsychotic medication use in North American nursing home residents. Health Policy. 2015;119:802-134 Good CB, Gellad WF. Off-Label drug use and adverse drug events: Turning up the heat on off-label prescribing. JAMA Intern Med. 2015. Online Nov 2. doi: 10.1001/jamainternmed.2015.6068. 5 Eguale T, Buckeridge DL, Verma A, Winslade NE, Benedetti A, Hanley JA, Tamblyn R. Association of off-label drug use and adverse drug events in an adult population. JAMA Intern Med. 2015. Online Nov 2 doi:10.1001/jamainternmed.2015.60586 Braillon A. Baclofen and protecting the patients during research. Alcohol Alcohol 2012;47:631.
Taichman DB, Backus J, Baethge C, et al. Sharing Clinical Trial Data: A Proposal From the International Committee of Medical Journal Editors. Ann Intern Med. 2016;164:505–506. [Epub ahead of print 26 January 2016]. doi: https://doi.org/10.7326/M15-2928
Download citation file:
Published: Ann Intern Med. 2016;164(7):505-506.
Published at www.annals.org on 26 January 2016
Ethics, Infectious Disease, Nephrology, Neurology, Renal Replacement Therapy.
Results provided by:
Copyright © 2020 American College of Physicians. All Rights Reserved.
Print ISSN: 0003-4819 | Online ISSN: 1539-3704
Conditions of Use