Sara Tannenbaum,1 Joseph S. Ross,2,3,4,5 Harlan M. Krumholz,2,4,5,6 Nihar R. Desai,2,6 Jessica D. Ritchie,2 Richard S. Lehman,2,7 Ginger M. Gamble,2 Jacqueline Bachand,8 Sara Schroter,9 Trish Groves,9 Cary P. Gross3,4,10
Although the International Committee of Medical Journal Editors (ICMJE) recommendations for trial data sharing have been controversial, little is known about the attitudes and experiences of authors published in journals with existing data sharing policies.
We conducted a self-administered online survey of the authors of clinical trials published January 1, 2012, through March 1, 2016, in 3 high-impact journals with policies either requiring all clinical trial authors to share data (PLOS Medicine) or publish a statement specifying whether they were willing to share data (The BMJ and Annals of Internal Medicine). For the latter 2 journals, we only contacted authors who specified that they were willing to share data. We contacted the corresponding author and then an additional author if no response was received. The survey addressed sharing plans, receipt of sharing requests, and effort required to respond to sharing requests. We also asked respondents about willingness to share data in 6 hypothetical scenarios. Each hypothetical request occurred 1 year after publication of the original study but varied by type of request. Survey results are for all respondents unless otherwise indicated.
Among the 154 trials for which we contacted authors, 90 responses (58.4%) were received. Respondents and nonrespondents did not significantly differ by journal, year published, region of the corresponding author, or funding source. Half of the respondents had a data sharing plan (n = 49 [54.4%]), and about one-third had received at least 1 sharing request (n = 31 of 89 [34.8%]). Out of the 68 data requests that were received in aggregate, only 4 (5.9%) were denied. Most respondents indicated that they would be willing to share data for a meta-analysis (n = 87 [96.7%]) or for replication of the primary study outcome (n = 66 [73.3%]) 1 year after publication. However, in response to scenarios indicating that data were requested for a secondary outcomes analysis or predictive modeling study, willingness to share was largely influenced by author intent to conduct similar analyses (Figure). For a secondary outcomes analysis, 70 authors (77.8%) responded that they would share if they had not planned a similar analysis, but 15 authors (16.7%) responded that they would share even if they had planned a similar analysis. Among authors who had granted at least 1 request (n = 25), a median (range) of 18 (3-125) person-hours were spent to prepare data for sharing.
Figure. Willingness to Share by Request Type for a Trial Published 12 Months Ago
Among respondents to a survey of clinical trial authors, we found that data sharing is taking place under journal data sharing requirements but that willingness to share data depends on the type of request and intent to publish similar analyses.
1Yale University School of Medicine, New Haven, CT, USA; 2Center for Outcomes Research and Evaluation, Yale-New Haven Hospital, New Haven, CT, USA; 3Section of General Medicine, Department of Internal Medicine, Yale University School of Medicine, New Haven, CT, USA, email@example.com; 4Robert Wood Johnson Foundation Clinical Scholars Program, Department of Internal Medicine, Yale University School of Medicine, New Haven, CT, USA; 5Department of Health Policy and Management, Yale University School of Public Health, New Haven, CT, USA; 6Section of Cardiovascular Medicine, Yale University School of Medicine, New Haven, CT, USA; 7UK Cochrane Center, Oxford, UK; 8Yale School of Public Health, New Haven, CT, USA; 9The BMJ, London, UK; 10Cancer Outcomes, Public Policy, and Effectiveness Research (COPPER) Center, Yale Cancer Center and Yale University School of Medicine, New Haven, CT, USA
Conflict of Interest Disclosures:
Dr Ross, Dr Krumholz, Dr Desai, Ms Ritchie, Dr Lehman, Ms Gamble, and Dr Gross receive research support from Janssen and the Pharmaceutical Companies of Johnson & Johnson, to develop methods of clinical trial data sharing. Dr Ross, Ms Ritchie, and Ms Gamble receive research support from the Blue Cross Blue Shield Association to better understand medical technology evidence generation. Drs Ross, Krumholz, and Desai receive research support from the Centers for Medicare and Medicaid Services to develop and maintain hospital performance measures that are used for public reporting. Dr Ross, Dr Krumholz, and Ms Gamble receive research support from the US Food and Drug Administration to develop methods for postmarket surveillance of medical devices. Dr Ross receives research support from the US Food and Drug Administration to establish the Yale-Mayo Center for Excellence in Regulatory Science and Innovation. Dr Gross receives research funding from 21st Century Oncology and the National Comprehensive Cancer Network-Pfizer. Dr Krumholz chairs a scientific advisory board for United Healthcare. Drs Schroter and Groves are full-time employees at The BMJ, and Dr Groves is a member of the Peer Review Congress Advisory Board but was not involved in the review or decision for this abstract. No other conflicts were reported.
Research reported in this publication was supported by National Institute of Diabetes and Digestive and Kidney Diseases of the National Institutes of Health (award No. 5T35DK104689-02).
Role of the Funder/Sponsor:
The funder had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the abstract.
The content is the sole responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.