Article Title

Stakeholders’ Views on Data Sharing and Multisite Research

Publication Date



human subjects research, confidentiality/HIPAA, observational studies, engagement of stakeholders, qualitative research


Background: Data sharing is a fundamental step in multisite studies, enabling research on rare outcomes, treatment heterogeneity and greater generalizability. However, data sharing entails costs and risks. Newly developed privacy-preserving analytic and data-sharing methods offer an approach to sharing data and conducting multisite research that eliminates the need to share identifiable patient-level information. Identifying stakeholders’ perceptions and synthesizing overarching themes about willingness to share data are important first steps toward increasing the acceptability and use of these new tools. We therefore sought to understand stakeholders’ views on data sharing generally and privacy-preserving methods in particular.

Methods: We conducted semi-structured group and individual interviews with a purposive sample of stakeholders to gather a variety of perspectives on data sharing in general and the use of privacy-preserving methods (PPM) specifically. Interviews were audio-recorded and professionally transcribed. Using content coding followed by thematic coding, we sought to identify factors affecting stakeholders’ willingness to share data, with particular attention to the potential impact of privacy-preserving methods.

Results: A total of 11 stakeholder interviews were completed, involving patients (n = 15), researchers (n = 10), institutional review board and regulatory staff (n = 3), multicenter research governance (n = 2) and health care system leaders (n = 4). Stakeholders’ perceptions of the benefit and value of the research was the strongest influence toward data sharing; perceived value was related to the relevance of the scientific question and the methodologic rigor. Influences against data sharing were primarily cost and data-security risks; the latter was mitigated by various safeguards (encryption, data use agreements, oversight), successful data sharing experience, established relationships and trust. The risk reduction obtained by sharing aggregate data rather than individual-level data was acknowledged as being potentially more acceptable to some stakeholders, but interviewees also expressed concerns about the increased cost and questioned whether aggregating data resulted in a loss of information that would in turn lessen the value or validity of the research.

Conclusion: The gains in privacy protection associated with the use of privacy-preserving methods in multisite studies involving data sharing were attractive to some stakeholders, but factors such as the value and generalizability of the research appear potentially more influential.




June 30th, 2017


August 10th, 2017