Objective The COVID-19 pandemic may have signaled a positive shift in attitudes toward sharing individual-level patient data (IPD).1 This project aimed to obtain IPD from trialists of randomized clinical trials (RCTs) of pharmacological treatments for COVID-19 for the purposes of conducting an IPD meta-analysis under the COVID-NMA initiative.2

Design This single cohort study evaluated the effectiveness of accessing IPD from trialists of COVID-19 RCTs through email requests and online data repositories. Participants were the corresponding authors of RCTs of pharmacological trials for the treatment of COVID-19 who published their findings in a preprint or peer-reviewed journal between March 2020 and May 2021. Corresponding authors were emailed once, with at least 2 reminders over 9 months (November 2020 to September 2021) regardless of their data sharing statement. Online data repositories (including Vivli, Yale University Open Data Access Project, and ClinicalStudyDataRequest. com) were also searched. The primary outcome was the proportion of studies for which IPD were accessed; IPD was defined as a data set obtained directly from trialists or those that were made accessible online. The project team collaborated with Vivli, a global data sharing platform.

Results Fifty-six of 229 COVID-19 RCT trialists (24%) shared their IPD by the end of December 2021. Of these, 18 (32%) declared their positive willingness to share in the registry and 42 (75%) in their preprint or publication. Of those trials that did not share their data, 53 (31%) declared their positive willingness to share in the registry and 100 (58%) in their preprint or publication. Stratified by funding, trials that shared vs did not share IPD received funding from sources that were public or nonprofit (shared: 13 23%; not shared: 79 46%), private (shared: 10 18%; not shared: 36 21%), mixed public/nonprofit and private (shared: 20 36%; not shared: 37 21%), no funding (shared: 11 20%; not shared: 10 6%), and not reported or unclear (shared: 2 4%; not shared: 11 6%). Trials that shared vs did not share IPD were based in high-income countries (shared: 19 34%; not shared: 45 26%), low- to middle-income countries (shared: 31 55%; not shared: 117 68%), and both high- income and low- to middle-income countries (shared: 6 11%; not shared: 11 6%).

Conclusions Despite the positive shift in attitudes toward sharing IPD, less than one-quarter of trialists shared their IPD. Furthermore, data sharing statements often did not line up with the ability to obtain the data. This study emphasizes the need to mandate and/or reward timely data sharing while addressing remaining administrative, resource infrastructure, and cultural obstacles.

References 1. Larson K, Sim I, von Isenburg M, et al. COVID-19 interventional trials: analysis of data sharing intentions during a time of pandemic. Contemp Clin Trials. 2022;115:106709. doi:10.1016/j.cct.2022.106709

2. Boutron I, Chaimani A, Meerpohl JJ, et al; COVID-NMA Consortium. The COVID-NMA project: building an evidence ecosystem for the COVID-19 pandemic. Ann Intern Med. 2022;173(12):1015-1017. doi:10.7326/M20-5261

Conflict of Interest Disclosures Isabelle Boutron is a member of the Peer Review Congress Advisory Board but was not involved in the review or decision for this abstract. No other disclosures were reported.

Funding/Support This study, as part of the COVID-NMA initiative, received funding from Université de Paris, Assistance Publique Hôpitaux de Paris, INSERM, Cochrane France (Ministry of Health), the French Ministry of Higher Education and Research, Agence Nationale de la Recherche, and the World Health Organization.

Additional Information Isabelle Boutron is a co–corresponding author.