National engagement on data: cohort 3 report

The data deliberation: what we set out to cover

Thinks Insight & Strategy is a research and strategy agency. Early in 2024, we were commissioned by NHS England (NHSE) and the Department of Health and Social Care (DHSC) to conduct a national public engagement programme, to build and maintain public trust in how data is used across health and social care.

Public engagement is crucial to the success of digital and data transformation, as policy and services must be informed by a detailed understanding of public attitudes and shaped by public recommendations. Meaningful engagement can help to build trust and confidence in the decisions taken by Government and the NHS. During phases one and two of this programme of engagement we explored data use and access (Cohort 1), then a single patient record and the use of general practitioner (GP) data beyond individual care (Cohort 2), each with its own report and public recommendations.

The third phase of the data deliberation, reported here, focused on the opt out landscape. The ways health and social care data can be used has changed and continues to evolve. It can be used to provide better care, to plan services effectively, and to support research into new treatments. But the current opt out landscape is complex and can be confusing. It is unclear if it works in the way that people expect or want.

So, this cohort of the data deliberation focused on one big question:

What we covered in Cohort 3

We used a range of different methods, as shown in Figure 2, to ensure that the data deliberation generated a robust understanding of public attitudes. Our findings from all three tiers have been consolidated into this report. We explained the current opt out landscape in terms of The National Data Opt-out (NDOO), type 1 opt out, programme opt outs, and study specific opt outs. We also explained that the NDOO doesn’t apply where there is a legal requirement, an overriding public interest or a court order to share data.

Tier 1 participants explored the opt out landscape and developed recommendations across 18 hours and five workshops:

• Workshop 1: Introduction to the project, familiarising participants with the format and broad content; introduction to NHS data uses; surfacing spontaneous attitudes to opt out and choice.
• Workshop 2: Work through initial assumptions and expectations about choice; deeper deliberation and engagement exploring how current opt out landscape compares to and affects expectations.
• Workshop 3: Developing participants’ understanding of what choices people want; developing outline principles for testing in Workshop 4.
• Workshop 4: Debating principles from Workshop 3 and develop initial practical recommendations.
• Workshop 5: Exploring participants’ views on recommendations; refining recommendations in response to participant and stakeholder feedback; highlighting areas of divergence between participants.

The structure of cohort 3

Across the 18 hours of deliberation, the Tier 1 participants explored the opt out landscape through the use of case studies, specialist presentations with Q&A and small table discussions. The specialist presentations were from medical professionals, researchers and privacy groups covering the benefits and risks of secondary uses of health data, and some of the processes used to keep data safe. The deliberations took place across three two-hour online sessions and two six-hour in-person sessions. The overall question about what choices people should have was broken down into four sub-questions:

1. Are there situations where opting out shouldn’t be allowed? (exploring what should be in or out of scope for opt out decisions)
2. Should people have a choice about whether their de-identified data is used?
3. Should people have a more detailed set of choices than just in or out? (exploring what choices people want over the different ways their data might be used)
4. Should people make a choice once or have multiple opportunities to opt out? (exploring different types of opt out methods)

Tier 2 participants were asked to deliver similar recommendations, abbreviated in line with the reduced time spent scrutinising the issues and trade-offs. Tier 3 respondents answered a set of questions testing some of the issues emerging from Tier 1 and Tier 2.

We summarise the key insights and recommendations emerging from this cohort below. We’ve pulled these together under four main themes, which correspond to the respective chapters of this report.

Theme 1: Overall insights of what the public want from the opt out system

Throughout the deliberative strand of this process, Tier 1 participants engaged in substantive discussion about the principle of choice. Via more qualitative approaches, Tier 2 gave specific audiences the space and time to engage with the topic, particularly those in vulnerable circumstances. Tier 3, without opportunity to deliberate on these complex issues, gave a strong indication of wider public opinion and as such helped contextualise potential policy changes and how they would be received.

The recommendations in this report were primarily driven by the views of the more deliberative Tiers 1 and 2, based upon the following key participant perspectives:

Many people, across all three tiers, believed that choice about how health and care data is used is a fundamental right, and not open to debate. The importance of choice for these participants held irrespective of the purpose, security measures in place, or their own preference for opting out or not. This was especially important for the majority of the marginalised groups who took part in the Tier 2 engagement.

In contrast, it is notable that by the end of the deliberations, around a quarter of Tier 1 participants felt there was a strong argument not to allow opt out in any circumstances. For these participants, a system where the public access free at the point of care services involves a ‘social contract’. They feel that allowing data to be used – especially when it enables the service to run effectively – is a responsibility that users of the NHS should accept.

The security considerations already in place for many secondary uses of data – for example, de-identification and the ongoing implementation of the Secure Data Environment (SDE) model – were reassuring to most Tier 1 participants. These measures gave participants confidence that the security of their data is taken seriously. For those whose primary concerns were about data being misused, this meant they were more comfortable with its use for secondary purposes and, therefore, less likely to insist on there being an opt out.

But, for others, security alone was not enough to give them confidence in their data being used for secondary purposes. They feared that even de-identified data stored securely could be used to their disadvantage or for purposes that they do not approve of. Both of these issues – fear of misuse and the desire to express disapproval of certain uses – were particular concerns for many of the marginalised groups in Tier 2, particularly transgender audiences, sex workers and domestic abuse survivors. Therefore, for these participants (and echoing what we heard from Tier 2 audiences in the previous two cohorts), it was essential they be given an opt out choice, regardless of security measures in place to protect their data.

Taking all these factors into account, most Tier 1 participants concluded that an effective opt out system must balance the need for choice with the need for data to be used for the greater good. There was widespread agreement that an opt out system (rather than an opt in system) best balances these competing requirements. Tier 2 participants could also see the benefits of an opt out system, alongside education and transparency about this system to encourage others to stay opted in. Tier 2 participants felt that while an opt out system protects the principle of choice they find so important, this only works if there are clear, honest and timely communications with the public about the system to enable opt out for those who wish to do so. The choice between an opt out vs opt in system was not tested in Tier 3 given the clear recommendation reached in Tier 1, which was consistent with the strong preference of most Tier 2 participants.

Several other factors also influenced how participants thought about the scope and shape of an opt out system:

Theme 2: Recommendations for an opt out system and the choices available

Tier 1 and Tier 2 participants suggested that any redesigned approach to choice should remain an opt out (as opposed to an opt in), but also that it should be redesigned in a way that balances the greater good and the need for choice. They considered whether or not there should be choices over data being used for planning and improving services and for research, respectively. They also had clear views on how exemptions should work in a redesigned system.

Participants produced several recommendations for ways in which they wanted opt out to change, as well as aspects of the current system that they wanted to keep:

An opt out system should be maintained. An effective opt out system must balance the need for choice with the need for data to be used for the greater good. Tier 1 participants felt that an opt out approach to choice, as opposed to an opt in approach, provided the best way of balancing these needs. This aligned with Tier 2 perspectives of a desire for control over the use of their data through the opt out system.

Remove the ability to opt out of data being used for the purposes of planning and improving services. Most Tier 1 participants (77%) felt this was an acceptable, and often expected, use of data, consistent with their feeling that they have a ‘social contract’ with the NHS as free at the point of care. Participants felt the definition of planning and improving services need to be clear so that their data would be used for the greater good, with direct benefit for the NHS, and used by the NHS or with NHS oversight.

However, any changes to opt out must be accompanied by rigorous safeguarding of data from misuse, particularly for vulnerable people. Tier 2 participants started from a more sceptical position of maintaining opt out for their data for any purpose. Where planning uses were segmented from other secondary purposes, some Tier 2 audiences felt comfortable removing the opt out for planning, so long as safeguards are in place. This was driven by transparency about the specific purpose and meaning of ‘planning’, the users of the data (with participants more comfortable if this was the NHS) and the potential for this data to improve the health service. However other Tier 2 audiences felt that planning should remain as part of the opt out. This was driven by a desire to mitigate the potential for misuse of their data by restricting access, rather than a particular concern with use of data for planning purposes.

Ensure people can opt out of data being used for research. Participants across both Tier 1 and Tier 2 had most desire for choice when their data is used for research, because of concerns about who, how, and why their data is being used. These concerns, such as the potential for financial gain by private companies from data access, led to more participants feeling they would expect to have a choice about these uses – even if they may personally decide not to opt out. Tier 2 participants, in particular, felt that there were uses and users associated with research that they would want to be able to opt out of regardless of whether data was de-identified.

Yet all policy decisions about the scope of the opt out should be closely informed by all the factors that influenced participants’ views: NHS involvement, the involvement of external organisations, how a use impacts ‘the public good’, and the extent to which planning and research are interlinked.

In a redesigned opt out system, exemptions should be limited in scope, clearly and precisely defined, and well explained (through examples and with justifications). While participants understood the need for some exemptions, they were concerned about ‘definition creep’ – i.e. gradually expanding exemptions with less and less obvious justifications. The opt out system should be redesigned in a more effective way that limits the need for exemptions – participants stressed the importance of reducing the need for exemptions.

Theme 3: Recommendations for how to make an opt out choice

Participants had clear recommendations to improve the way they make their opt out choices. They felt the current system is over-complicated and not transparent. They also suggested that the choices they want to be able to make are not reflected in the way the system works now.

The following recommendations were some of the strongest and most widely supported, across and within Tiers, of all – participants saw a shift to a better understood opt out system as critical to its legitimacy. The recommendations were:

Create a simpler, unified way of opting out that makes it clear to the public what is in and out of scope, giving genuine choice. Participants emphasised the need for a single ‘touchpoint’ when choosing how your health and care data is used. They wanted to change from the current system of opt outs in multiple places and having to make multiple decisions.

The opt out system should be developed to allow people to tailor their opt out choice if they want to, or to make a single choice if they don’t. Across all three tiers, there was a clear message that opt out should be developed to allow people to tailor their opt out choice if they want to, alongside making a single choice if they don’t. Further work is required to test the design (and feasibility) of the ‘drop down’ element with the public.

People should also be reminded of their opt out status – regardless of whether they are currently opted in or opted out – from time to time. The design of any system of reminder contact should minimise the burden on the system and on the public. Further work is needed to test the detailed design of this with the public.

Theme 4: Recommendations for informing the public

Participants placed a great deal of emphasis on how little they knew about why the opt out exists or how it currently works before the sessions began. They also stressed that they knew little about how or why their health and care data is, or could be used. Learning about the use of data and the benefits it can bring to their own care and the way the NHS functions was eye opening for all participants. Greater awareness also tended to lead to increased comfort with secondary uses of data, and an understanding of the importance of planning and improving services and research.

For Tier 2 especially, participants strongly believed that it is the duty of the NHS to ensure the public are properly informed about what is happening with their health and care data, and their opt out choices. They felt that this duty is not being fulfilled at the moment, leading to an assumption that there is something being hidden from them.

As a result, participants across Tiers 1 and 2 strongly recommended that more be done to raise awareness among the wider public (this was not explored in Tier 3 as other topics were prioritised given their limited time with the information):

Increase awareness of both the possibility of opting out and the safeguards in place. Learning about the use of safeguards (including de-identification) built trust and confidence among participants. Opt out choices should be a more integrated part of how people interact with health and care services to enable people to make informed choices. Trust in, and the legitimacy of, the opt out is undermined by the current low levels of knowledge and awareness.

Increase awareness of why secondary data use is important and valuable. Learning the benefits of data use built trust and confidence among participants. Awareness of the usefulness and importance of data should be a more integrated part of how people interact with health and care services, and participants advocated for a proactive comms campaign to achieve this.

Download the full report

National engagement on data - Cohort 3 report, PDF (1.4MB)

Download the workshop discussion guides

Workshop 1 discussion guide, PDF (228KB)

Workshop 2 discussion guide, PDF (282KB)

Workshop 3 discussion guide, PDF (263KB)

Workshop 4 discussion guide, PDF (227KB)

Workshop 5 discussion guide, PDF (254KB)

Download the stimulus packs

Workshop 1 stimulus pack, PDF (1.5MB)

Workshop 2 stimulus pack, PDF (3.1MB)