​NHS IT: Using the open standards and the cloud to drive digital transformation

How one of the UK's largest NHS trusts is developing its own infrastructure to put patients at the centre of healthcare.
Written by Mark Samuels, Contributor

Leeds Teaching Hospitals NHS Trust is building out an electronic patient records system which will help to deliver integrated healthcare for 800,000 people across the region.

The Trust has chosen to avoid off-the-shelf systems and instead create its own electronic health record (EHR); Leeds had already built a successful Cancer Information System for the regional Cancer Network. This system is now being used as the basis for the home-grown EHR and an associated digital transformation programme.

"Having a system that could support the way in which healthcare is delivered locally is a huge and important goal," says Richard Corbridge, the trust's chief digital and information officer. "At the time of the decision, an evaluation of the market place showed that many of the off-the-shelf systems supported a very generic form of health and care delivery, often one often based on the US model of care delivery."

Building a platform for change

The EHR delivers what Corbridge refers to as a "digital fabric" for 800,000 people across the Leeds region. In the hospital, the system supports bar-code scanning of patients, beds, rooms, drugs and samples, removing potential errors in transcribing and aiding real-time flow analysis of patients.

Medications and observations are recorded digitally at the point of care, meaning the bulk of information about patients is collected at the bedside. "The EHR provides the opportunity for us to deliver integrated care across the hospital," says Corbridge. "The information we hold puts the patient at the centre of healthcare."

SEE ALSO: NHS IT: Can better use of tech give healthcare a shot in the arm?

Corbridge says the bespoke-developed EHR is also well-suited to the demands of the Trust's workforce. A recent survey of clinical users placed satisfaction with the current system as high as 70 percent, although the survey did indicate additional functionality is required. Corbridge says the EHR provides a boon here, too.

"By owning the ability to evolve the system in an agile manner, Leeds maintains the ability to react quickly to changes that are required and to ensure it can be a good test bed for information standards and changes to integration and interoperability across the entire healthcare system," he says.

Corbridge says the bespoke EHR at Leeds helps meet his desire for a more integrated approach to healthcare technology because it is built on open standards and can be integrated with more than 40 additional systems across the hospital.

Later this year a patient portal will go live, allowing patients across Leeds to add information to the EHR. Once again, the key is interoperability. Corbridge says there must be the capability to take bespoke developments in the NHS, like the strategy he is pursuing at Leeds, and to apply those benefits generally once success is proven.

"There are more than five organisations now with successfully designed, built and tested home-grown systems," says Corbridge, pointing to a growing trend. "Any NHS organisation heading down this route must consider open data and system standards, using expert resources, such as action group INTEROPen for technical assistance and not-for-profit Apperta Foundation for key guidelines."

Corbridge says that despite Leeds being one of the three largest trusts in the UK and one of the largest training organisations in the NHS, the organisation has one of the cheapest technology functions. The budget for the EHR is just £3m a year. Corbridge estimates supporting a comparative off-the-shelf solution would cost £35m over five years.

While the Leeds EHR is currently run on-premise, the system will migrate to cloud provision in the next three years. Crucially, the system will make use of both vendor-neutral cloud provision and archive solutions due to the open nature of its architecture and the standards adopted.

"The solution is already utilising vendor-neutral archive solutions with optical character recognition functionality built in for historic scanned paper notes," says Corbridge. "Cloud functionality is also being utilised as part of the solution for data analytics in particular in the field of clinical research and population health analysis."

Supporting further innovation

Corbridge is currently working with clinicians on future developments. The EHR is due to go mobile later this year, passing the power of data into the hands of healthcare professionals on the ward. Corbridge paints a picture of the kind of organisation he'd like to be part of in three years' time.

"Clinicians would come in and boot up the EHR on their mobile devices. AI would help us to look at the acuity of patients within the hospital and plan ward rounds automatically. As they get nearer to the patients, the records would be brought up automatically and the most recent information," he says.

"When they get to the patient, the proximity of the clinicians brings up more information and -- as they interact -- more information is added to the EHR to create a contemporaneous record. None of that is technology that's beyond our grasp -- all of it exists. Pulling that information together and putting it in one hospital is the vision for the next three years."

The Trust has started to open conversations with several organisations to help push developments in emerging technology, both smaller start-ups and more traditional vendors. Those conversations are focused on how Leeds can reduce the amount of screen time for clinicians who deliver healthcare.

SEE: Special report: How to automate the enterprise (free ebook)

Corbridge told ZDNet recently that the Trust's first foray into machine learning will be a robotic process-automation development which aims to make it easier to find the right medical records among hundreds of scanned documents, saving time for clinicians.

"We need to think about how we can use machine learning to hear, listen, learn and apply the knowledge it collects to inform what's displayed on the screen," says Corbridge. "And that's potentially exciting -- it's already being done in other industries. Our aim is to bring those advances to healthcare."

Leeds plans to use a chat bot to help clinicians search the EHR for patients notes by voice request. The bot handles all identity, search and optimisation requirements. It will discover and learn who the clinician is and what patients and conditions they are searching for. The system will then present patient records to the clinician's mobile device.

"Not being able to search scanned paper is one of the biggest blockers to clinical engagement," says Corbridge. "Imagine a person with a long-term condition -- they could have several hundred scanned PDF files that have to be scrolled through to reach the piece of information that the clinical team needs. Automation will help reduce the time clinicians need to spend in front of a screen."


NHS IT: Can better use of tech give healthcare a shot in the arm?

Is improving tech the key to to making healthcare more efficient?

How smart contact lenses will help keep an eye on your health

Health tech: Researchers are working on combining sensors with off-the-shelf contact lenses for a range of medical applications.

DeepMind's AI spots early signs of eye disease

Initial results of DeepMind's partnership with Moorfield's Eye Hospital in London suggest that a scanning method that uses artificial intelligence could provide quicker diagnoses and help prevent sight loss.

Healthcare security nightmare: UK's NHS lost nearly 10K patient records last year (TechRepublic)

Despite having electronic record systems in place, 94% of NHS Trusts still use handwritten notes for patient record keeping, according to a report from Parliament Street.

This swallowable chip uses glowing bacteria to spot hidden illnesses

MIT researchers have created swallowable chips that can help identify blood in the gastrointestinal tract.

Government websites hijacked by cryptocurrency-mining malware (CNET)

Over 4,000 websites worldwide were affected by the malware.

Editorial standards