Reviewing the ODSL Goldacre Review Special

As part of our #OpenDataSavesLives event series, we featured Jessica Morley, from the University of Oxford who spoke about her work on openSAFELY - the new secure analytics platform for electronic health records. To continue this conversation we suggested we could support Jess's work and that of the whole team at the University of Oxford's DataLab, by running a special live un-conference to ask the #OpenDataSavesLives community to submit their response and comments to the Goldacre Review. Specifically, Jess was interested in understanding blockers and enablers to the better, broader and safe use of health data for research and analysis. They wanted to understand how we can facilitate the creation of an open collaborative ecosystem around health analytics, and our #OpenDataSavesLive Goldacre Review special is a wonderful example of exactly the response they were looking for.

What Is the Goldacre Review?

The Secretary of State for Health and Social Care has asked Dr. Ben Goldacre to lead a rapid review on the more efficient and safe use of health data for research and analysis for the benefit of patients and the healthcare sector.

To inform this review, input and feedback are sought from a broad range of people working with data across the healthcare system, in different roles, and at different levels of seniority, as well as patient and public representatives.

Department for Health and Social Care

How did ODILeeds help?

#OpenDataSavesLives was started in response to the Covid-19 pandemic and over the course of the last year we have hosted over 25 events, reached 1000+ attendees, and attracted over 75 speakers. The initiative now has 3 partners, ODILeeds, Beautiful Information and the Health Foundation and its own dedicated website. #OpenDataSavesLives is an example of new & open institutional thinking in health data. These events, the interest they generate and the outcomes show what can be done with a well-functioning network in very little time.

The Session

ODILeeds has a huge collaborative workspace in the centre of Leeds. We are used to convening events and live streaming, hybrid events and un-conferences were all part of our remit even before the pandemic. We now have Live@ODILeeds, a dedicated online events team who have researched software platforms and built our own virtual platform; this allows us to be able to continue to share our work and our sponsors and partners whilst physical events aren't possible. So we used this expertise to host a live and virtual un-conference session for this special meet. We shared a working google doc before and during so that participants could easily collaborate and respond directly to the questions posed by the review.

The Munro House is the home of ODILeeds and where we used to host in-person events

The Response

We had to quickly turn this around but managed to generate plenty of interest with over 50 sign-ups & remote attendees and 35 participants on the day. Many attendees were new to the #OpenDataSavesLives series and we saw a combination of health professionals, GP's & data scientists not just from the UK but all around the world. It was an exciting showcase of the power of open data and the reach of our network.

One table connected professionals from the UK, Germany and Canada

The Outcomes

These sessions are special as they offer a neutral, independent, non-partisan space that is free and open to all. Because of this we often find people are very interested in engaging with us and trying to make a change. As a result, we had some wonderful discussions about the opportunities and challenges that healthcare data provides.

A lot of contributors highlighted that this conversation needs to go beyond data accessibility. Currently, a lot of expertise and experience are siloed. So, similar institutions (like different GPs) should be encouraged and enabled to openly share their skills and code with each other to simplify, streamline and speed up analysis processes. This may be an alternative to attempting to overhaul and centralise the entire healthcare data infrastructure. Indeed, a contributor proposed that understanding the existing decentralized structures and building new systems around them would be more realistic and effective. To drive this kind of change, participants campaigned for more informatics experts in high-ranking and senior positions. A lack of data experts in leadership positions means analytics issues are often overlooked. So, assigning appropriate value to the development of this kind of infrastructure requires political will and board-level maturity. Probably the most recurring idea of our event was the importance of generating and maintaining trust. People need to feel like their data is used safely and ethically to actually make a difference in healthcare and not to earn somebody money. Many participants echoed this by underlining that being clear and transparent in all communications and providing solid evidence of the value that health data-based research provides is crucial.

A good example of how good, accessible healthcare data can affect policy and practice, as mentioned by one of the attendees, is the National Cancer Registration and Analysis Service (NCRAS). They explained how this - linked with other useful data like mortality and primary care data - allowed for insights supporting cancer awareness campaigns and a deeper understanding of symptoms. On another positive note, a commenter remarked that the UK is already in a comparatively good position when it comes to population health management. The efforts that are already being made in terms of segmentation and social marketing are hopefully a good starting point for further data infrastructure improvements and innovative use cases.

Did It Make Any Difference?

Yes - the google document, this blog and all communications have been forwarded to the team at the Goldacre Review. The value and impact of #OpenDataSavesLives is widely recognised. Besides, our impact went beyond just the big players: Following the session, we were approached by a GP who was interested to know what data is gathered in the daily uploads and how it could ultimately lead to improvements in tracking actual patient journeys from 111 and 999 calls. Another was interested in being able to potentially use GP datasets to study childhood adverse outcomes to match health data about families, to identify at-risk children. So yes small things can make a change.

What Next?

You can find out more about #OpenDataSavesLives here, the next meeting is on the 22nd of April and will focus on social care - tickets are available here. Please follow us on Twitter @livesnotdata or @odileeds to keep up to date with our work and we have a mailing list if you would like to sign up to keep in touch with the latest updates from us.

Thank you :)