Empowering clinicians and researchers with unprecedented analytics

About

EMIS is one of the largest suppliers of electronic health records software in the UK. The company was founded more than 30 years ago, but its mission has remained the same – to develop the best possible technology to help healthcare professionals and ultimately, to improve patient health.

Challenge

Today, EMIS serves 10,000 organizations across every major healthcare setting and securely stores millions of National Health Service (NHS) records. EMIS supplies clinical software systems used on a daily basis by tens of thousands of doctors and nurses – whether for standard patient appointments or in life-threatening emergencies. Healthcare providers, academics, research groups, and commissioners look for ways to analyze the data generated during routine healthcare to uncover insights and trends. 

The challenge for EMIS – especially during the COVID-19 pandemic – had been finding the best way to make healthcare data and analytics available to providers and research organizations, and to do so in a way that’s secure, cost-effective, high-performance, and scalable. 

“We recognized that we can deliver greater value to our customers by enabling them with the latest technology to carry out analytics at scale,” explains Chief Analytics Officer, Richard Jarvis of EMIS. Clinical data is continuously streamed into a cloud data lake, ensuring that it includes the very latest information, including patient medications, allergies, symptoms and more. “Patients change over time, so we need to regularly update the data we hold about individuals and medical practices, and we need to do that efficiently, in a way that keeps costs down both for us and for our customers,” Jarvis says. 

Solution

To move away from a siloed, hosted environment to a wide-ranging cloud environment that could bring data from multiple sources and underpin its transformative analytics suite, EMIS-X Analytics, EMIS chose Starburst Enterprise to access its Delta Lake. 

EMIS needed a secure, high-performance query engine to query the available data in Delta Lake. Starburst’s Apache Ranger and native Delta Lake reader provide a secure, enables the EMIS-X Analytics platform to analyze data at scale, rather than just store it. The reliability of Starburst Enterprise also allows EMIS to scale compute usage to meet the needs of its customers.

Analytics with EMIS-X 

The EMIS-X Analytics suite includes a range of data exploration and visualization tools. As an example, EMIS was able to produce an online visualization delineating the most common reported symptoms of COVID-19 that updated daily, as new self-reported patient data streamed into its cloud-based data lake from a patient-facing application. “Researchers could look and see at any given point in time what the most common symptoms were,” Jarvis explains. “To get this kind of information in the past would have involved several weeks of copying and analyzing data. With EMIS-X Analytics, powered by Starburst Enterprise, researchers get the information in almost real-time.”

Results

Cost savings

EMIS wants to deliver the best service possible to its customers, and part of that means keeping its costs down. “What Starburst and Delta Lake allow us to do, along with the work we’ve done ourselves, is keep our costs down without compromising performance,” Jarvis explains. “This is important because we want to minimize the costs of this infrastructure for our healthcare customers.”

Joining data

Aided by Starburst technology, EMIS-X Analytics makes it possible to analyze both primary and secondary care data simultaneously. “The correlation between the data from GPs and the data from hospitals could be used to see trends and outcomes of the pandemic,” Jarvis says. “By bringing that data together for our customers, often for the first time, we’re helping them discover new ways to improve patient outcomes.”

Faster time to insight

“Being able to pull that data into our current platform, with the appropriate information governance and data permissions, we can give our customers a view across 100 or even all 4,000 EMIS GP practices,” Jarvis notes.

Support & optimizations

Some EMIS customers run very large, complex queries involving a lot of data. For example, looking at all patients to find symptoms that indicate COVID risk might involve querying hundreds of billions of rows of data. EMIS receives valuable guidance from the experts at Starburst. “We get great support from the Starburst team on questions like how to make a query run quicker,” Jarvis notes. “That’s important for us and our customers, because the quicker we can run our queries, the quicker a clinician can get to a new insight, and the quicker a patient can receive better care.”

Better outcomes, better quality of life 

EMIS envisions its customers using the platform to generate both clinical and operational insights. Different subject matter experts could generate algorithms which search for patients with specific risk factors for certain health conditions and alert their physicians. Practices could run analytics that optimize scheduling in order to minimize missed appointments — the sort of analytics that simply weren’t available to physicians in the past. Yet EMIS is also looking forward to seeing how their customers leverage this unprecedented access in unexpected ways. “The next bit is the interesting part,” Jarvis says, “where instead of sending packets of data out to researchers for them to analyze, we can bring their models to work inside our environment and provide insights directly back to the people who need them. If you can detect something early and treat it, then you don’t have lots of repeat appointments or the need to resort to expensive drugs. You can give people a better quality of life and better outcomes, while driving system efficiency and helping clinicians. So all of that centers back to the core vision of this program, and our reason for partnering with Starburst — to help our customers get more out of their data.”