Data Sharing Statements for Clinical Trials a Requirement of the International Committee of Medical Journal Editors Darren B
Total Page:16
File Type:pdf, Size:1020Kb
Opinion EDITORIAL Data Sharing Statements for Clinical Trials A Requirement of the International Committee of Medical Journal Editors Darren B. Taichman, MD, PhD; Peush Sahni, MB, BS, MS, PhD; Anja Pinborg, MD; Larry Peiperl, MD; Christine Laine, MD, MPH; Astrid James, MB, BS; Sung-Tae Hong, MD, PhD; Abraham Haileamlak, MD; Laragh Gollogly, MD, MPH; Fiona Godlee, FRCP; Frank A. Frizelle, MB, ChB, FRACS; Fernando Florenzano, MD; Jeffrey M. Drazen, MD; Howard Bauchner, MD; Christopher Baethge, MD; Joyce Backus, MSLS The International Committee of Medical Journal Editors (ICMJE) nisms are not in place to mandate universal data sharing at this believes there is an ethical obligation to responsibly share data time. Although many issues must be addressed for data shar- generated by interventional clinical trials because trial partici- ing to become the norm, we remain committed to this goal. pants have put themselves at risk. In January 2016 we pub- Therefore, ICMJE will require the following as conditions lished a proposal aimed at helping to create an environment in of consideration for publication of a clinical trial report in our which the sharing of deidentified individual participant data member journals: becomes the norm. In response to our request for feedback we 1. As of July 1, 2018, manuscripts submitted to ICMJE jour- received many comments from individuals and groups.1 Some nals that report the results of clinical trials must contain a applauded the proposal while others expressed disappoint- data sharing statement as described below. ment it did not more quickly create a commitment to data shar- 2. Clinical trials that begin enrolling participants on or after Janu- ing. Many raised valid concerns regarding the feasibility of ary 1, 2019, must include a data sharing plan in the trial’s reg- the proposed requirements, the necessary resources, the real istration. The ICMJE’s policy regarding trial registration is ex- or perceived risks to trial participants, and the need to protect plained at http://www.icmje.org/recommendations/browse the interests of patients and researchers. /publishing-and-editorial-issues/clinical-trial-registration It is encouraging that data sharing is already occurring in .html. If the data sharing plan changes after registration this some settings. Over the past year, however, we have learned should be reflected in the statement submitted and published that the challenges are substantial and the requisite mecha- with the manuscript and updated in the registry record. Table. Examples of Data Sharing Statements That Fulfill These ICMJE Requirementsa Example 1 Example 2 Example 3 Example 4 Will individual participant Yes Yes Yes No data be available (including data dictionaries)? What data in particular All of the individual Individual participant data that underlie Individual participant data that underlie Not available will be shared? participant data the results reported in this article, the results reported in this article, collected during the trial, after deidentification (text, tables, after deidentification (text, tables, after deidentification figures, and appendices) figures and appendices) What other documents Study protocol, Study protocol, statistical analysis plan, Study protocol Not available will be available? statistical analysis plan, analytic code informed consent form, clinical study report, analytic code When will data be available Immediately following Beginning 3 months and ending 5 years Beginning 9 months Not applicable (start and end dates)? publication; no end date following article publication and ending 36 months following article publication With whom? Anyone who wishes Researchers who provide Investigators whose proposed use Not applicable to access the data a methodologically sound proposal of the data has been approved by an independent review committee (“learned intermediary”) identified for this purpose For what types of analyses? Any purpose To achieve aims in the approved proposal For individual participant data meta-analysis Not applicable By what mechanism Data are available Proposals should be directed to xxx@yyy. Proposals may be submitted up to 36 months Not applicable will data be made available? indefinitely at To gain access, data requestors will need following article publication. After 36 months (link to be included) to sign a data access agreement. the data will be available in our university’s Data are available for 5 years data warehouse but without investigator at a third-party website support other than deposited metadata. (link to be included). Information regarding submitting proposals and accessing data may be found at (link to be provided). a These examples are meant to illustrate a range of, but not all, data sharing options. jama.com (Reprinted) JAMA June 27, 2017 Volume 317, Number 24 2491 © 2017 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/30/2021 Opinion Editorial Data sharing statements must indicate the following: must work toward fulfilling the other steps of best practice whether individual deidentified participant data (including as well—including data sharing. data dictionaries) will be shared; what data in particular will As we move forward into this new norm where data are be shared; whether additional, related documents will be avail- shared, greater understanding and collaboration among able (eg, study protocol, statistical analysis plan); when the data funders, ethics committees, journals, trialists, data analysts, will become available and for how long; and by what access participants, and others will be required. We are currently work- criteria data will be shared (including with whom, for what ing with members of the research community to facilitate prac- types of analyses, and by what mechanism). Illustrative ex- tical solutions to enable data sharing. The United States amples of data sharing statements that would meet these re- Office for Human Research Protections has indicated that pro- quirements are in the Table. vided the appropriate conditions are met by those receiving These initial requirements do not yet mandate data shar- them, the sharing of deidentified individual participant data ing, but investigators should be aware that editors may take into from clinical trials does not require separate consent from trial consideration data sharing statements when making editorial participants.2 Specific elements to enable data sharing state- decisions. These minimum requirements are intended to move ments that meet these requirements have been adopted at the research enterprise closer to fulfilling our ethical obliga- ClinicalTrials.gov (https://prsinfo.clinicaltrials.gov/definitions tion to participants. Some ICMJE member journals already main- .html#shareData). The WHO also supports the addition of such tain, or may choose to adopt, more stringent requirements for elements at the primary registries of the International Clinical data sharing. Trials Registry Platform. Unresolved issues remain, including Sharing clinical trial data is one step in the process articu- appropriate scholarly credit to those who share data, and the lated by the World Health Organization (WHO) and other pro- resources needed for data access, the transparent processing fessional organizations as best practice for clinical trials: uni- of data requests, and data archiving. We welcome creative versal prospective registration; public disclosure of results solutions to these problems at www.icmje.org. from all clinical trials (including through journal publication); We envision a global research community in which shar- and data sharing. Although universal compliance with the ing deidentified data becomes the norm. Working toward this requirement to prospectively register clinical trials has not vision will help maximize the knowledge gained from the ef- yet been achieved and requires continued emphasis, we forts and sacrifices of clinical trial participants. ARTICLE INFORMATION and Deutsches Ärzteblatt International (Baethge); Medicine, New Zealand Medical Journal, PLOS Author Affiliations: Secretary, ICMJE, and Representative and Associate Director for Library Medicine, The Lancet, Revista Médica de Chile Executive Deputy Editor, Annals of Internal Operations, National Library of Medicine (Backus). (Medical Journal of Chile), and Ugeskrift for Laeger Medicine (Taichman); Representative and Corresponding Author: Darren B. Taichman, MD, (Danish Medical Journal). Past President, World Association of Medical PhD, Annals of Internal Medicine (dtaichman Disclaimer: Dr Sahni’s affiliation as representative Editors (Sahni); Scientific Editor-in-Chief, @acponline.org). and past president of the World Association of Ugeskrift for Laeger (Danish Medical Journal) Published Online: June 5, 2017. Medical Editors (WAME) does not imply (Pinborg); Chief Editor, PLOS Medicine (Peiperl); doi:10.1001/jama.2017.6514 endorsement by WAME member journals that are Editor-in-Chief, Annals of Internal Medicine (Laine); not part of the ICMJE. Deputy Editor, The Lancet (James); Editor-in-Chief, Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Journal of Korean Medical Science (Hong); REFERENCES Editor-in-Chief, Ethiopian Journal of Health Sciences Disclosure of Potential Conflicts of Interest. Taichman DB, Backus J, Baethge C, et al. Sharing (Haileamlak); Editor, Bulletin of the World Health Dr Peiperl reported receiving travel reimbursement 1 clinical trial data: a proposal