What is it about?

Data unavailability impedes research transparency and is a major problem for individual participant data (IPD) meta-analyses as it reduces statistical power, increases risk of bias, and may even preclude completion. The primary objectives of this study were to determine IPD sharing plans reported in recently registered clinical trial registration records, how data sharing commitment relates to clinical trial characteristics, and principal investigators' attitudes, motivations and barriers to data sharing. The secondary objective was to derive recommendations to overcome identified barriers to data sharing. This was a retrospective cohort study of all interventional trials registered on the Australian New Zealand Clinical Trials Registry (ANZCTR) from 1 December 2018 to 30 November 2019, and an online cross-sectional survey of their principal investigators.

Featured Image

Why is it important?

In the cohort study of all clinical trials registered on the ANZCTR in the study period (n = 1517), commitment to share data was low (22%, 329/1517). In the cross-sectional survey (n = 281, 23% response rate), principal investigators showed strong support for the concept of data sharing (77%, 216/281) but a substantially lower intention to actually share data from their clinical trials (40%, 111/281). Major barriers to data sharing included lacking informed consent to share data, protecting participant confidentiality and preventing misinterpretation of data or misleading secondary analyses. There is a gap between high in-principle support for data sharing, and low in-practice intention from investigators to share data from their own clinical trials. Multiple pathways exist to bridge this gap by addressing the identified barriers to data sharing.

Perspectives

What is already known • Sharing of individual participant data generated by clinical trials is considered an ethical obligation by the International Committee of Medical Journal Editors because trial participants have put themselves at risk; yet this ethical obligation is often not met. • Data unavailability is a common issue which impedes research transparency and limits the completion of powerful and non-biased IPD meta-analyses. What is new • Commitment to share data, as reported in registration records, was low (22%, 329/1517). • Support for the concept of data sharing, as reported in the investigator survey, was strong (77%, 216/281), but intention to actually share data was substantially lower (40%, 111/281). • There was a gap between high in-principle support for data sharing, and low in-practice intention from investigators to share data from their own clinical trials. • Major barriers to data sharing included lacking informed consent to share data, protecting participant confidentiality, preventing misinterpretation of data or misleading secondary analyses, and misunderstanding the concept of sharing IPD and need to share IPD.

Dr Aidan Tan
University of Sydney

Read the Original

This page is a summary of: Data sharing—trialists' plans at registration, attitudes, barriers and facilitators: A cohort study and cross‐sectional survey, Research Synthesis Methods, June 2021, Wiley, DOI: 10.1002/jrsm.1500.
You can read the full text:

Read

Resources

Contributors

The following have contributed to this page