Using health data to identify and approach people about health and care research: A Public Dialogue

About the study

The UK has one of the largest and most comprehensive collections of health data in the world. Heath and care researchers could use this data to identify participants for health and care research studies based on their health experiences. With far larger and better-targeted samples, these studies could drastically improve the nation’s health outcomes. However, the UK’s data protection regulations prevent researchers from accessing this data without prior consent, severely limiting and skewing the data available for this purpose – though not without good reason.

NatCen’s CfD explored whether a Welsh public would accept processes that identify and contact people about research studies by accessing their health data without explicit prior consent. It further sought to investigate the balance between participants’ preferences for individual privacy and confidentiality on the one hand, and the public good on the other. Across various workshops, participants learned about the topic, and considered a range of perspectives through extended discussions with each other and subject specialists. 

Findings

CfD analyses uncovered a range of public views on whether NHS Welsh should use patient data to identify people for research studies without their explicit consent. Some participants remained concerned about being contacted by researchers in this manner, especially if they had knowledge of health conditions perceived as sensitive. However, many expressed support for such processes, citing improvements to health research and the potential for further care through research participation. For many participants, their support rested on, or could be brought about by, a series of safeguarding principles and conditions, including:

• Research quality: accessing and using patient health data must have a material impact on health and care research – improving recruitment efficiency or facilitating more precise or better-quality research outcomes.
• Equity: participants held that using health data in processes that identify and approach participants should contribute to efforts to make health and care research more inclusive, diverse, and accessible.
• Public benefit: uses of health data for such processes should prioritise public needs and generate benefits for the entire Welsh population – public benefits must have primacy, particularly over commercial or profit-driven considerations.
• Data safeguarding: processes that rely on patient health data must be carefully regulated and subject to oversight procedures.
• Public involvement: the Welsh public should be kept up to date on future initiatives via public engagement campaigns; provided with relevant information and consulted about potential plans.
• Individual autonomy and privacy: must be respected. For a minority of participants, personal autonomy and privacy outweighed all other considerations – in some cases, ruling out the possibility of permissive data-driven processes. However, for others, a commitment to protecting individual privacy was important, though not more important than the potential benefits of developing more effective processes to identify and approach people about research studies.

Methodology

Members of the Welsh public were convened over four dialogue workshops, including three two-and-a-half hour online learning sessions and one all-day face-to-face session. The three learning sessions covered the benefits and challenges of using health data to identify research participants, existing models for doing so, and examples of data restriction exemptions (including during the COVID-19 pandemic). In the fourth and final workshop, participants discussed three scenarios which invited them to consider the question of data access through different hypothetical perspectives, as well as the broader public benefit associated with more permissive practices. The day culminated in a group decision-making exercise about whether, if at all, processes could be implemented in future.

The sample consisted of 35 participants, stratified to broadly reflect the Welsh population in key demographics while ensuring the inclusion of voices seldom-heard in research.