Cool Things the NHS is Doing with Data

The UK’s National Health Service (NHS)—the fifth biggest organization in the world after the US Army, Chinese Army, Wal-Mart, and McDonalds—has more than 1.5 million patient interactions per day. It’s ripe for taking advantage of new big data technologies. Yesterday at the London O’Reilly Strata Conference: Making Data Work, I listened to two inspiring talks about what is being done with big NHS data and what we’ll soon be able to do—and by soon I mean next year.

One talk worth listening to was Francine Bennette’s “Data Nerding in Public Health” in which she describes some of the work the big data specialist Mastodon C as well as Ben Goldacre (@bengoldacre) are doing with publicly available data on prescriptions (400 million data points) and characteristics of GP surgeries such as location, size and so on (21 million data points). (I recommend watching the video of this talk up on YouTube.) Although this data can be illuminating, it’s important to note this data is what one might call “clinical” rather than “patient” data.

What I want to tell you about is what the NHS is doing with clinical and patient data. Tim Kelsey (@tkelsey1) is the NHS National Director of Patients and Information (previously Advisor to the Prime Minister on Transparency in Public Data). You can catch that talk here as well, but he said a few things I really want to shout about. He’s looking for new ways to unleash the power of patient data for the patient, giving people real control over their health and care. He wants to empower people to take control of their own health through their health data the way online banking did for financial data over a decade ago. Indeed, in 1998 the first online banking came to the UK and was met with skepticism; today more than half of the adult population only do online banking.

So what’s the NHS doing with its data?

  • NHS Hack Days. These hackathons bring together doctors, developers, nurses, and “all geeks that love the NHS” to help solve healthcare problems through collaborative disruptive innovation. The last one was in Cambridge, which Springer’s very own Matias Piipari from Papers attended. Matias is working in collaboration with a cardiologist and other developers to create an app that allows patients to visualise the effects of their actions on their heart health. Checkout other NHS hackathon apps on the NHS Hack Days wiki page. Most of the code is available on GitHub.
  • My data. At present, we have innovators like Matias donating their time to develop amazing apps; the problem is the data isn’t flowing. For example, at present, if you want access to your medical records, you must apply for access, and you may have to pay a fee. That’s about to change. From April 2014 every patient will have access to his or her medical records online. This will give every patient access to their own data and to be able to do things with it (mash it up with other data sources, for example).
  • Connecting hospital data with GP data. Currently we have data coming from our hospitals (called Hospital Episode Statistics Database) as well as data coming from our GP surgeries, but they’re not connected. By next June, data from these two databases will be linked and made public. This has never happened before on this scale anywhere. Yes, I’m talking about the NHS!
  • Friends and Family Test. In April 2013 the NHS launched the Friends and Family Test. It’s a simple question: would you recommend this service to your friends or family? Now, we have available literally millions of comments from patients on the quality of their local service (a “TripAdvisor for healthcare”, as Kelsey describes it). You can see the latest data from this survey here.
  • Care Connect. Inspired by the 311 hotline service in some US cities, Care Connect launched as a small pilot allowing people to tweet, ring, or email their thoughts on local services. Going national next year, it will allow people to share their experience, ask a question, or report a problem and receive a response within 24 hours. The data collected from this new service will be collected and analyzed over time.
  • This program will unleash anonymized patient data and link it with the clinical data mentioned previously. This initiative is particularly exciting because of its relevance to researchers. Take the data we have on chemotherapy delivered through the NHS, for instance. We spent £1 billion on chemotherapy last year within the NHS, but we don’t have data on how many people actually get this therapy. We have financial data, but no data on actually assessing the impact of treatment, the outcomes, etc. We need more real-world evidence—what actually happens to people on a particular cocktail of medications. This is what researchers like the genomics community are missing. Prime Minister David Cameron has just committed to sequencing 100,000 gene sequences in the next 3 to 5 years. This will be worthless without real-world evidence found in the data the NHS holds. Indeed, our NHS is not the first to recognize this. The Global Alliance for Genomic and Clinical Data Sharing has taken steps in this direction as well.

Kelsey closed his talk with a story illuminating the impact of opening up NHS data. Sitting next to a mother of a young child with a learning disability, he heard the story of her repeated weekly, sometimes daily, visits to her local hospital. Although well-known by the hospital staff, she spent much of her time during these visits completing the same forms. The doctors knew her; the institution didn’t. The hospital’s inability to “remember” her daughter and to link GP and hospital data meant waiting hours for what might be a comparatively small service. When he asked her what one thing he could do to make the NHS work for her and her daughter, her reply was: “Make the health service remember who I am.” Opening up and using the data behind the NHS is not just about costs savings or the improved health that can come with empowering people to use their own data to make better decisions, but about restoring dignity to healthcare.


View the latest posts on the Research in progress blog homepage