Data sharing - The time has (not yet?) come

Clyde W. Yancy; Robert A. Harrington; Robert O. Bonow

doi:10.1001/jamacardio.2018.1939

Data sharing - The time has (not yet?) come

Clyde W. Yancy^*, Robert A. Harrington, Robert O. Bonow

^*Corresponding author for this work

Medicine, Cardiology Division

Research output: Contribution to journal › Editorial › peer-review

4 Scopus citations

Original language	English (US)
Pages (from-to)	797-798
Number of pages	2
Journal	JAMA cardiology
Volume	3
Issue number	9
DOIs	https://doi.org/10.1001/jamacardio.2018.1939
State	Published - Sep 2018

ASJC Scopus subject areas

Cardiology and Cardiovascular Medicine

Access to Document

10.1001/jamacardio.2018.1939

Cite this

@article{45da2b2e3bda46f38dedbdf083689217,

title = "Data sharing - The time has (not yet?) come",

author = "Yancy, {Clyde W.} and Harrington, {Robert A.} and Bonow, {Robert O.}",

note = "Funding Information: The CSDR recently announced its partnership with 4 major nonprofits: the Bill and Melinda Gates Foundation, Cancer Research UK, the Medical Research Council, and Wellcome Trust.17 The best-chronicled effort has been from the Sprint Challenge that was promoted by the New England Journal of Medicine. In this public solicitation of proposals to access a well-phenotyped and carefully configured database, 279 groups expressed interest, yielding 143 proposals and 3 winners. Specifically, approximately 3 of 279 (1%) of data requests were deemed appropriate when held to a high methodological bar for access, reanalysis, or new discovery.18 Recent experience with YODA has also identified issues related to investigator access and data use.19-21 Thus, despite the theoretical and ethical imperatives, the effect of actual use and output has been limited. This early experience is most consistent with a tentative “toe-in-the-water” approach. Funding Information: For several years there have been well-articulated discussions of the need to share data that come from randomized clinical trials.1-7 The voices have come from clinical research-ers, those who fund research, journal editors, legislators, and, most importantly, pa-tients and their advocates. We have heard these calls for shared data and applaud the intent of open data repositories and full transparency while sharing the concerns of unlimited access to patient-level data that are obtained in highly structured clinical trials. Many have opined on this question. The International Committee of Medical Journal Editors released a definitive position in 2016.8 It believes that there is an “ethical obligation” to responsibly share data. The editors further suggested that clinical trial registrations should include a transparent plan for data sharing. Bauchner et al4 have supported this idea by arguing that we have a responsibility to study participants to optimize the use of data from clinical trials, stating, “The social contract for taking these risks and experiencing these harms imposes an ethical obligation that the results lead to the greatest benefit to society.” Rosenbaum9 stated the questions well: data scientists support data sharing as a means to uncover missed opportunities and guarantee that results from publicly funded trials are available for the greater good.9 However, trialists support a proprietary interest in the ownership of the data that facilitates future trials and mitigates concerns over misinterpretation.6 Many have referred to the sentiments of Rory Collins: “What problem are we trying to solve?”9 In this issue of JAMA Cardiology, we have invited respected leaders as data scientists and clinical trialists to opine on the potential benefits and risks of a more open and transparent data repository that is intended to validate known discoveries, generate new hypotheses, and uncover unforeseen consequences (or benefits) of novel therapies.10,11 Data scientists feel that the universality of science argues for a fully open and interoperable research platform with a free flow of ideas.12 Nascent efforts in this space have included the National Heart, Lung and Blood Institute BioLINC repository,13 the Yale University Open Data Access (YODA) project (a repository of nearly 200 clinical trials that are supported by 1 major pharmaceutical and device company)2,14 and the ClinicalStudyDataRequest.com (CSDR) platform (a repository of more than 3000 clinical trials that is supported by 14 pharmaceutical companies).15,16 Accessing the CSDR platform has been of interest. Of the 393 research proposals that were made to the Wellcome Trust (the purveyors of the CSDR), most have been for new hypotheses rather than confirming original results; data access was provided to 186 (47.3%), and only 21 proposals (5.3%) have been published.16",

year = "2018",

month = sep,

doi = "10.1001/jamacardio.2018.1939",

language = "English (US)",

volume = "3",

pages = "797--798",

journal = "JAMA cardiology",

issn = "2380-6583",

publisher = "American Medical Association",

number = "9",

}

TY - JOUR

T1 - Data sharing - The time has (not yet?) come

AU - Yancy, Clyde W.

AU - Harrington, Robert A.

AU - Bonow, Robert O.

N1 - Funding Information: The CSDR recently announced its partnership with 4 major nonprofits: the Bill and Melinda Gates Foundation, Cancer Research UK, the Medical Research Council, and Wellcome Trust.17 The best-chronicled effort has been from the Sprint Challenge that was promoted by the New England Journal of Medicine. In this public solicitation of proposals to access a well-phenotyped and carefully configured database, 279 groups expressed interest, yielding 143 proposals and 3 winners. Specifically, approximately 3 of 279 (1%) of data requests were deemed appropriate when held to a high methodological bar for access, reanalysis, or new discovery.18 Recent experience with YODA has also identified issues related to investigator access and data use.19-21 Thus, despite the theoretical and ethical imperatives, the effect of actual use and output has been limited. This early experience is most consistent with a tentative “toe-in-the-water” approach. Funding Information: For several years there have been well-articulated discussions of the need to share data that come from randomized clinical trials.1-7 The voices have come from clinical research-ers, those who fund research, journal editors, legislators, and, most importantly, pa-tients and their advocates. We have heard these calls for shared data and applaud the intent of open data repositories and full transparency while sharing the concerns of unlimited access to patient-level data that are obtained in highly structured clinical trials. Many have opined on this question. The International Committee of Medical Journal Editors released a definitive position in 2016.8 It believes that there is an “ethical obligation” to responsibly share data. The editors further suggested that clinical trial registrations should include a transparent plan for data sharing. Bauchner et al4 have supported this idea by arguing that we have a responsibility to study participants to optimize the use of data from clinical trials, stating, “The social contract for taking these risks and experiencing these harms imposes an ethical obligation that the results lead to the greatest benefit to society.” Rosenbaum9 stated the questions well: data scientists support data sharing as a means to uncover missed opportunities and guarantee that results from publicly funded trials are available for the greater good.9 However, trialists support a proprietary interest in the ownership of the data that facilitates future trials and mitigates concerns over misinterpretation.6 Many have referred to the sentiments of Rory Collins: “What problem are we trying to solve?”9 In this issue of JAMA Cardiology, we have invited respected leaders as data scientists and clinical trialists to opine on the potential benefits and risks of a more open and transparent data repository that is intended to validate known discoveries, generate new hypotheses, and uncover unforeseen consequences (or benefits) of novel therapies.10,11 Data scientists feel that the universality of science argues for a fully open and interoperable research platform with a free flow of ideas.12 Nascent efforts in this space have included the National Heart, Lung and Blood Institute BioLINC repository,13 the Yale University Open Data Access (YODA) project (a repository of nearly 200 clinical trials that are supported by 1 major pharmaceutical and device company)2,14 and the ClinicalStudyDataRequest.com (CSDR) platform (a repository of more than 3000 clinical trials that is supported by 14 pharmaceutical companies).15,16 Accessing the CSDR platform has been of interest. Of the 393 research proposals that were made to the Wellcome Trust (the purveyors of the CSDR), most have been for new hypotheses rather than confirming original results; data access was provided to 186 (47.3%), and only 21 proposals (5.3%) have been published.16

PY - 2018/9

Y1 - 2018/9

UR - http://www.scopus.com/inward/record.url?scp=85053878848&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85053878848&partnerID=8YFLogxK

U2 - 10.1001/jamacardio.2018.1939

DO - 10.1001/jamacardio.2018.1939

M3 - Editorial

C2 - 29971326

AN - SCOPUS:85053878848

SN - 2380-6583

VL - 3

SP - 797

EP - 798

JO - JAMA cardiology

JF - JAMA cardiology

IS - 9

ER -

Data sharing - The time has (not yet?) come

ASJC Scopus subject areas

Access to Document

Other files and links

Cite this