PMF Interviews Natalie Banner, Lead – Understanding Patient Data (UPD), A Wellcome initiative

In the lead up to our Patient Week, we have interviewed several of the speakers that will be joining us for the event.

Natalie Banner is joining us for our Patient Week event on March 9 to discuss Patient-Centric Data and Sharing.

Natalie has shared some of her fantastic insight into the work that is being done at Understanding Patient Data, at the Wellcome Initiative.

During many of our virtual events recently, we’ve heard of the importance of data collection and sharing within Healthcare. From your perspective, why is the availability of this data so important? And how does Understanding Patient Data (UPD) enable all parties (patients, clinicians, researchers, etc.) to connect and discuss the need for it?

The Covid pandemic has highlighted in such a clear way how and why the availability of data is so essential for our care, our health, and for research to improve these. Data has been critical to the public health response, from identifying case numbers to recognising the importance of loss of smell as a symptom, to highlighting shocking disparities in outcomes for people from different backgrounds. The RECOVERY trial, for example, is an astonishing clinical trial that rapidly identified what treatments worked for hospitalised Covid-19 patients. It was able to set up so efficiently because it could draw on routinely collected data. So there is enormous potential for data collected in the health system to be used to drive benefits back to patients and healthcare, through research, innovation and service planning. But there are also understandable risks and concerns about its misuse. After all, data about your health is some of the most sensitive information there is.

UPD aims to take the fear factor out of conversations about health data and demystify it, by providing accessible resources to answer the typical questions people have – and by people I mean health professionals just as much as patients or the public. We also seek out people’s views and concerns, and channel those into policy and practice. So where a new policy or initiative is being set up that will use patient data, we try to ensure these are trustworthy, and responsive to people’s views and values.

There are several exciting projects happening at Understanding Patient Data, in particular ‘Defining the use of data for public benefit’. If possible, could you provide us with more details about this project and what it means for the patients and public? Additionally, what impact do you feel this research will have on improving the public’s trust in data collection/sharing?

When researchers or analysts want to access health data, they usually go through an application process to assess if their proposed use of the data is legal, ethical and safe. Often this includes assessors determining if what they want to do with the data is in the public interest or for public benefit. But there’s currently no clear guidance on what this means, or how members of the public would judge for themselves what counts here.

At UPD, we think it’s important that people have a say in the rules for governance and decision-making about data use. So we have been collaborating with the National Data Guardian and Sciencewise, to fund some deliberative research exploring how people think that ‘public benefit’ should be defined and assessed. The findings are due to be published around Easter, and will inform statutory guidance from the National Data Guardian due in the summer.

Our aim isn’t to improve public trust – it’s to improve the trustworthiness of the systems in place for collecting, managing and using data about people. Efforts to improve trust place the emphasis on individuals to ‘have trust’ but without necessarily being equipped with the tools, knowledge or capacity to determine whether or not they should place their trust in institutions, people, processes and so on. Data and technology are changing society very quickly and people have legitimate reasons to ask questions and be sceptical. By emphasising trustworthiness instead of trust, we argue it’s incumbent upon the system to demonstrate why and how people should trust it – and showing how decisions are made about whether a data use is for public benefit is one aspect of this.

The trust of patients and the public in the safety of these systems that store our health data is imperative to the success of data being shared across the NHS.  Can you give us more information about the safety measures that are put in place to protect a patient’s identity when they are asked to share their health data?

The first thing to note here is that in many cases, people aren’t explicitly asked to share their data – it is often data collected as part of their routine care, that can be managed and used by the health system and partners for other purposes such as research and planning. It is not a legal requirement to ask for consent in every case.

There are three types of measures to protect people’s identities: technical, legal and social or ‘people’ measures. Technical measures include removing identifying information at source, or encrypting this information to make it harder to connect the data to a specific individual. NHS systems also need to comply with high standards of information security to guard against breaches or misuse. In many cases, it is also possible to audit the data uses, to ensure data is only being used in permitted ways. But, as with any digital system we use in our day-to-day lives, these methods are not completely foolproof. I liken them to putting locks and alarms on your property: you may not be able to 100% guarantee against a burglary, but you can make it incredibly difficult for someone to break in.

Legal safeguards include the Data Protection Act and the Common Law Duty of Confidentiality. Any use of patient’s data for purposes beyond their care needs to meet these standards and there are big penalties for the misuse of personal data. Data Protection law also sets a very high bar for data to count as ‘anonymous’, meaning that most uses of health data will be protected by this law.

Finally, ‘people’ measures: training and setting a culture of respect and protection for people’s privacy is vital. The risks of human error can be reduced with good training to ensure people who handle data are properly accredited.

How do we come to a better understanding of how health data is being shared and used – is it purely education and advocacy? Or is there more to it than that?

It often surprises and scares people to hear that data from their health records can be used for purposes beyond their care without their explicit consent. Literacy about health data and how it is used is certainly very important, but it can be pretty dry in the abstract. We’ve found it’s far easier to inform people about how health data is used from within the context of their own care and interactions with the health service. For example, if you get a new medication, the information you receive about it could include something about how research using data from other people has helped determine the right dosage. These ‘nuggets’ of information may give people an insight into how data underpins their care, but without overwhelming them with detail or technical jargon. So that’s an important starting point.

But it shouldn’t end there: people have a stake in data about them and how it is used, it’s not enough just to inform them. Our research indicates people expect citizen involvement in decision-making about how data held by the NHS is used by third parties, for example. And there’s lots of scope for governance and decision-making about data to draw on the views of the wider public, if only we put the time and resource into listening. Accountability to society is only going to become more important as data-driven technologies become more embedded in our lives and our care, so they should reflect societal views and values in how they’re developed and implemented.

For data to give maximum utility we need to ensure diversity and inclusion. It is widely recognised there are disparities in healthcare, how can we overcome these challenges?

There is no such thing as “the public” and everyone will have their own views about the use of data depending on their background, experiences in and beyond the health system, and own preferences and values. Most engagement and communication work done with patients or public representatives tends to be only with those who are already engaged and interested, or have the capacity to spend time and energy delving into these questions. There is huge value to these expert insights, but much less done with non-expert audiences whose views may differ from those of a more ‘professionalised’ patient advocacy group.

We also have very little understanding of whether there are different perspectives or sensitivities around data use depending on, say, ethnic background, or age, or socio-economic status – how are we ever going to use data to address health inequalities if we don’t seek out the views and perspectives of those whom inequality impacts the most? Diversity and inclusion are not added extras here, they are central to ensuring data use is equitable and mitigating the risks that it might make things worse for some people.

Thank you to Natalie Banner for taking the time to answer our questions.

To hear more from Natalie, please join us for our Patient Week event that is running from March 8 – 11. Find out here and register here.

Please feel free to share this...