Enablers

Our research question was: “What are the barriers and enablers for patients and the public participating in sharing personal healthcare data?”

This section of our report looks at the enablers, which we have broken down to:

• Trust
• Public benefit
• Information
• Relationships

A key enabler for patient and public attitudes towards data sharing is the perceived trustworthiness of different organisations and professionals. Research has found that “The overwhelming factor in considering whether or not to share data… is whether or not people trust the organisation asking for it” ⁶⁴.

It appears that trust is not distributed equally. Ipsos Mori has found that “Public sector healthcare providers are most trusted to use personal information” in comparison to other organisations such as insurance companies and social media ⁶⁵. This ties in with other research, which found that “The NHS is highly trusted compared to other organisations. Generally, GPs and the NHS are trusted the most, followed by academic researchers” ⁶⁶.

Trust is also not given unconditionally. A study by Healthwatch Surrey found a trusting attitude towards data sharing – but with caveats: “If the parameters of medical and care record sharing are kept within the confines of the medical and care professions, and there are procedures in place to ensure security, confidentiality and accuracy, there is strong support from survey respondents for sharing all or part of their health and care record with all or some of the health and care professionals involved in their care”.

Good regulation can help to engender trust, but social and political context also matters. One study noted that “Public trust is essential for any technological system that is deployed widely and with significant impacts across society. For interventions like digital contact tracing or immunity certification… Transparency, accountability, independent oversight and appropriate data protection are key". However, it also made the point that “when discussing trust, many members of the public veered away from specific technologies and instead spoke about the social and political systems in which they are deployed” ⁶⁷.

One example might be that “For some participants who came to the workshop with pre-existing concerns about privatisation, all other concerns were looked at through this lens” ⁶⁸.

Evidence indicates that people are likely to support health data sharing if the reason for doing so offers clear public benefit: “The vast majority of respondents reported that if their health information could help others (e.g. via research) they would be happy to share it…A number of respondents see increasing data sharing as the only way for the NHS to achieve higher efficiency in the future (43%)” ⁶⁹. This is confirmed by another report, which states that “A commitment to public benefit is likely to increase support, but which also warns that this “is key to increasing support for the use of shared health information for research, but has been little considered in engagement in relation to system planning” ⁷⁰.

The public benefit may not have to be specific. People are willing to accept “potential benefits, and to respond to broader ideas such as “altruism”. One study found that “Assurances that research would – or at least have the potential to – bring about public benefits were fundamental for ensuring public support or acceptance” ⁷¹. Another states that there is “widespread willingness to share patient data for research for the 'common good'. Participants [were] willing to share electronic health records (EHRs) for secondary purposes when there was a 'common', 'greater' or 'public good', and when there was 'social responsibility', 'altruistic attitudes' and 'giving something back'" ⁷².

A public engagement exercise expands on this, again looking at potential, rather than proven benefits: “important reasons given for voting “yes” from each jury were that the planned use [could] lead to improved treatments, services, and care delivery and eventually to better health outcomes and more lives saved… Could strengthen research and help identify health trends… lead to better diagnoses of conditions, more effective treatments, and improved health outcomes for patients… allow NHS to more efficiently target the use of resources for particular conditions or communities” ⁷³.

However one study disagrees: “Being clear about specific secondary uses, and providing examples that are tangible and meaningful to the wider public, is likely to gain more support than presenting it in a more abstract or generic way” ⁷⁴.

It may, however, be important to emphasise that the ultimate public benefit is improved patient care. One study looked at “reasons for optimism about accessing NHS patients’ and NHS operational data”. It summarised these as “Improved patient outcomes, Sharing knowledge and expertise, Efficiency savings and increased NHS revenue, Enhanced reputations for patient data partners". It also made the point that “Improved patient care has to be the main purpose for entering into data access partnerships in the eyes of Jurors…That should be the endgame” ⁷⁵.

The other side of the “optimism” coin is the opportunity cost of not sharing healthcare data. In this respect, “Jurors reflected on the cost of not using the data for access partnerships. They thought about potential missed opportunities afforded by data access partnerships due to bureaucracy, inadequate or ineffective processes and complex organisational structures. They saw this, together with NHS organisations not being in tune with the fast-moving world of digital and technological innovation as serious challenges to success” ⁷⁶.

The “public benefit” case appears to hold good for commercial access to health data as well: “Participants wanted companies accessing health data to demonstrate that they are making a contribution to public value and a real, long term benefit to health in the UK. They look to government and regulators to enforce it”. This means, however, that “we need a shared understanding of what that value actually is, and what kinds of public goods we want data analysis to create” ⁷⁷.

Learning acquired from the Care.data scheme suggested that better quality information could have enabled public engagement and understanding. One suggestion was about methods of communication: “There does need to be more effort to inform people of the programme in addition to the letter to each patient of the GP surgery. These should include posters and leaflets at GP surgeries and other public spaces, public meetings and an advertised phone line for further information” ⁷⁸.

A further suggestion was about the content of communications: “A more informative public awareness campaign — which better covers areas such as the benefits of care.data, as well as the ethics, data security and safeguards — would help to build trust in the process and the wider healthcare system” ⁷⁹.

Information on data sharing should not be confined to the detail of specific government initiatives, but should be part of a more general process: “data-driven technologies should include patients and the public as active partners. This means that patients and the public [should be] active partners in agreeing priorities for, and determining the acceptability of, datadriven technologies as part of an ongoing process” ⁸⁰.

Information also needs to be clear about all sides of the data-sharing debate: “No use of technology is without risk, and honest discussions about the potential risks of the use of AI is necessary in order to ensure that progress on AI in healthcare is sustainable. Public trust cannot be engendered if only benefits are highlighted” ⁸¹.

Alongside transparency about risk, however, there is a need to reassure patients and public that risk is well managed: “Assurances of data security are important for public acceptability of research uses of data” ⁸². “It is important that the public are made aware of the safeguards in place to ensure the security of data” ⁸³. “Patients need to feel confident that their data is secure and in the hands of trusted actors” ⁸⁴.

Various studies suggest that the patient/professional relationship may be an important enabler for healthcare data sharing.

One possibility is that “the greater general level of trust in healthcare providers compared to other public bodies may be due to individuals’ existing relationships with primary healthcare providers, which are in some cases built up over many years, suggesting that having a familiar relationship with an individual is important for securing trust”. This study added that “trust in the NHS was expressed in terms of a perception that health professionals generally serve to help the public and are expected to abide by a moral code of conduct” ⁸⁵.

Another made the point that: “Older generations are more likely to use digital health technologies if recommended to do so by their healthcare professional. This indicates that healthcare professionals can play an important role in establishing data-centric habits among older healthcare system users” ⁸⁶.

This could also play into the shared decision-making agenda: “The majority of people (77%) felt that it should be a combination of patient and professional who decide how records are shared. (Patient and doctor together, 42%, or patient, doctor and social care practitioner, 35%)” ⁸⁷.

Relationships may also be important at the organisational, as well as the individual level: “While the NHS’ release of open health data will benefit both public and private-sector entities, those benefits can be enhanced if the beneficiary entities can help participate in shaping the government’s open data programme. The NHS should engage key stakeholders on a regular basis to determine which datasets have the highest priority for them; what new datasets should be released as open data; and which open data collections are particularly easy or difficult to use” ⁸⁸.