EMIS Group Uses AWS For Highly Secure, Scalable Analytics

UK-based EMIS Group is at the forefront of healthcare software and its products are used by over half of UK general practitioners (GPs). It provides a range of tools to help doctors, hospitals, pharmacists, and other care providers deliver more integrated services through better access to patient data using EMIS-X Analytics, a suite of cloud-based analytics products. The first application to be released as part of the suite is named Explorer.

Explorer is an analytics program that provides doctors, researchers, and life sciences companies with a secure way to allow the healthcare records of 4,000 GP practices to be searched as one. This includes a dataset of 35 million patient records and 1.3 trillion clinical events and interactions. Researchers access the data under the Control Of Patient Information (COPI) emergency data access legislation. During the pandemic, Explorer was used to combine data from 4,000 GP offices and hospitals with data that patients entered directly into EMIS’s patient-facing website—Patient.info—to provide researchers from the University of Oxford with the data they needed to make important insights into COVID-19. 

Running such a system on premises would have meant weeks spent uploading, copying, and analyzing data. But by building on Amazon Web Services (AWS), EMIS was able to get its system to ingest data daily from various sources and store it automatically in a data lake. 

In addition, one EMIS customer used Explorer to create a near-real-time visualization showing the most common symptoms to help doctors diagnose the disease. 

In both instances, the ability to analyze data in real time helped doctors improve their understanding of COVID-19 symptoms and helped to speed up diagnoses in the early days of the pandemic. It also helped to improve operational decisions by using localized infection rates to predict likely future spread of the disease and outbreak hot spots.

During the pandemic, doctors also used Explorer to create simple dashboards to determine which patients were missing treatment appointments for other illnesses because of worries about COVID-19 and needed specific support to access services. In addition, researchers are using the system to identify which groups of people are not getting vaccinated to help tailor messaging to improve vaccination rates. 

Richard Jarvis, chief technology officer at EMIS, predicts that Explorer will impact patient experience from the operational intelligence it can offer. For instance, relatively simple automation allows the system to intelligently tailor the length of GP appointments. So a routine prescription renewal appointment would be shorter than a follow-up meeting for a patient with a complex medical history. 

Given the tens of thousands of GP appointments every day, even relatively small improvements in efficiency, or reduction in overruns, could have profound benefits for the patient experience and the health service overall. 

“These kinds of operational improvements might not be the most exciting, and they’re not hugely complex, but they are really useful ways to improve patient care,” Jarvis says. “The quickest wins are often quite simple analytics that can help people. A small improvement in GP appointments, when multiplied across thousands of appointments every day, can have a huge impact.” 

Explorer can also provide diagnostic help—it can identify patients who more likely to suffer from atrial fibrillation, a common condition that, left untreated, can lead to a heart attack. And it can provide tailored side-effect information when prescribing for patients, making it possible to personalize information by age, gender, allergies, and general health profile. 

Security lies at the heart of the system. Users have different levels of access to the data depending on their role and organization. All data is encrypted, both at rest and in transit. Jarvis says that a key reason for choosing AWS was its ability to secure the core infrastructure, encrypt all the data at all stages of the journey, and provide fast and scalable access to data when required. 

Jarvis also credits AWS—which follows a shared responsibility model with its customers—with helping to guide the company through the process of getting security right at every stage. “We received helpful advice from AWS. We shared the mission we’re trying to achieve, and they gave us tips to get there,” says Jarvis.

Explorer presents two technology challenges. One involves running a massive and constantly growing data lake—1.3 trillion entries and growing. The second comes from providing analysis for growing numbers of users, all of whom are making a variety of inquiries at different scales. 

These users might include doctors looking for interactions between two uncommon drugs to make sure there are no dangerous side effects for patients. A researcher’s request, by contrast, might require querying millions or even billions of rows of data and creating a complex visual output. EMIS was able to easily navigate these challenges while also ensuring the highest levels of data security and privacy compliance. 

EMIS runs all of this storage and analytics infrastructure on AWS. When things do go wrong, Jarvis says, auto failover and auto correction means problems are usually fixed before staff need to log in and investigate.

As its analytical capabilities continue to grow, EMIS-X Analytics will be able to augment existing expertise and provide insights to power personalized healthcare advice—both proactive and predictive—and treatment. 

Thanks to AWS, the EMIS platform is highly scalable, allowing companies and researchers to access huge amounts of data when they want to import information into their own in-house analytics systems. 

EMIS also provides a bring-your-own-artificial intelligence (AI) service model. For instance, a provider that has created a model to predict diabetes can submit it to EMIS, which can then run that model using the provider’s data and can even inform GPs of results—all while communicating safely without any data security concerns. 

Such advances will change how healthcare is delivered. For the NHS, this could bring a profound shift in how data is used. With faster access to data, information could be used to predict conditions for individual patients weeks or months in advance. Ensuring this is done with the highest levels of security and informed consent will help reassure patients that their data is safe. 

The insights the system provides are already fueling operational improvements and efficiencies across a variety of healthcare settings, through capabilities like predictions for better staffing level decisions or intelligent scheduling to minimize late or missed appointments. “What excites the people who work at EMIS is helping patients have better lives—that’s why they get out of bed in the morning, not to use the best technology, the most AI, or the biggest dataset,” says Jarvis.