Developing a strategy to improve data sharing in health research: A mixed-methods study to identify barriers and facilitators

Background: Data sharing presents new opportunities across the spectrum of research and is vital for science that is open, where data are easily discoverable, accessible, intelligible, reproducible, replicable and verifiable. Despite this, it is yet to become common practice. Global efforts to develop practical guidance for data sharing and open access initiatives are underway, however evidence-based studies to inform the development and implementation of effective strategies are lacking. Objective: This study sought to determine the barriers and facilitators to data sharing among health researchers and to identify the target behaviours for designing a behaviour change intervention strategy. Method: Data were drawn from a cross-sectional survey of data management practices among health researchers from one Australian research institute. Determinants of behaviour were theoretically derived using well-established behavioural models. Results: Data sharing practices have been described for 77 researchers, and 6 barriers and 4 facilitators identified. The primary barriers to data sharing included perceived negative consequences and lack of competency to share data. The primary facilitators to data sharing included trust in others using the data and social influence related to public benefit. Intervention functions likely to be most effective at changing target behaviours were also identified. Conclusion: Results of this study provide a theoretical and evidence-based process to understand the behavioural barriers and facilitators of data sharing among health researchers. Implications: Designing interventions that specifically address target behaviours to promote data sharing are important for open researcher practices.