Organisations that hold health and care data often make ‘public benefit’ assessments to help decide who should access that data and for what purposes. But there isn’t a shared understanding or consistent approach to evaluating public benefit, how to account for it in applications, and what (or who) we should draw on to define it. Understanding Patient Data has worked on two projects to address this gap - one in 2021 with the National Data Guardian, and one in 2018 with Carnegie UK Trust and Involve. Read more about what we mean by 'public benefit' below.

Our work with the National Data Guardian

We teamed up with the National Data Guardian (NDG) on a project to bring the public into the conversation about how to assess public benefit. The project had two parts: the first was to bring members of public together with experts to understand how best to define public benefit; and the second is to use the findings to inform the guidance the NDG will be developing this year.

What did we do?


We held a series of online public dialogues with participants representative of the population from around England. The dialogues ran over several weekends in late 2020 and were designed to give participants sufficient time to discuss pertinent issues in depth; something that is impossible to do in a standalone focus group. 

What did we find out?

We didn’t get exactly what we expected. Participants pushed us to think beyond creating a ‘check-list’ for public benefit. Instead, people argued for the need to embed on-going evaluation of public benefit throughout the data life cycle.  

1. There’s no public benefit without transparency 

Participants initially surprised us. They encouraged us to see transparency as the key ‘litmus test’ for public benefit, before impact or other considerations were discussed. The dialogues found that people believe an absence of transparency is in direct opposition to achieving public good, meaning it’s just too important to neglect. 

The reasoning for this became clear early on. Participants explained that public benefit could only occur if there was sufficient information about a project or process that enabled people to scrutinise, evaluate and ask questions about what is being done. In a sense, they wanted to democratise elements of evaluating public benefit to ensure accountability. This is hardly surprising given the importance of the use of health and care data in the response to Covid-19. 

2. Motivation is as important and what’s done with data 

Participants were concerned that misuse of data could do great harm. This was highlighted especially when the motivations of data users were discussed and seen as a potential barrier to achieving public benefit. Some public bodies, media organisations and commercial companies are perceived as most likely to ‘manipulate’ data to serve their interests. For example, participants were concerned about the potential for social care data   to be used in a way that could exacerbate inequity in the provision of quality care across the country. 

To address this, participants advocated for safeguards: 

  1. Publication of data users’ credentials and sources of funding 
  2. Re-evaluation of public benefit when there are significant changes in research or the way data is used 

3. Public benefit is defined more widely than tick-boxes  

Participants pushed back at suggestions that public benefit could be defined simply by the number or extent to which some people benefit from the use of data. In this sense, ‘benefit’ didn’t mean meeting a numeric threshold. However, participants advocated for a few key measures data holders should take as part of a ‘working’ definition of public benefit. 

  • ‘Benefit’ should be evaluated over time, and incorporate both short and long-term impact. Evaluations should include the likelihood of generating improvements in services and/or health outcomes as well as more indirect and long-term impacts, such as advancing human understanding. 
  • Public benefit must always outweigh profit. Benefit can include income generation for NHS organisations (for example), but this should be always secondary, and the main purpose must be to improve health and care. 
  • Including the public is part of public benefit. Some participants suggested regular mini-public dialogues to review data use applications which are ‘edge cases’ and a commitment to include public representatives on all data assessment panels, committees or boards. 

What's next?

Download the full UPD/NDG report here.

Following on from this work and other efforts, the National Data Guardian published guidance on public benefit in 2022. Read this guidance here.


Our work with the Carnegie Trust

In 2018, Understanding Patient Data worked with Involve and Carnegie UK Trust to explore the tensions between the benefits and risks of data sharing.

About the research

 It's important that the sharing of personal data between public service providers will benefit the public. However, the term ‘public benefit’ is rarely, if ever, clearly defined. This research published in April 2018, unpacks and explains the benefits and risks around data sharing. 

The report is based on the findings from a series of six workshops in different local authority areas across England. The workshops brought together over 120 professionals from the public and voluntary sectors (working in housing, criminal justice, health, social care and welfare) to explore how they understand, define and value the public benefits which could be derived from the use of personal data.

The findings

The findings are available in the report titled 'Data for Public Benefit: Balancing the risks and benefits of data sharing.'

Three tests emerged from our research for public service providers to gain the social licence to share and use data more widely. These are that data sharing should be:

  1. Purposeful
  2. Proportionate
  3. Responsible

The report presents a new framework for public service providers to assess whether these tests have been met. The framework aims to provoke discussions among public service providers and encourage them to engage the public in decisions about data sharing.