In our latest interview we spoke with Dr Tanya Pankhurst, Deputy Director of Digital Healthcare from University Hospitals Birmingham NHS Foundation Trust (UHB).
What systems are used at UHB and how advanced are you with technology?
Here at UHB we are quite unique, we have a team of programmers and build systems ourselves, we are like a medium sized software business. The systems include a hospital Electronic Health Record, PICS (Prescribing, Information and Communication System); a hospital portal, a patient portal, (MyHealth@UHB) and a external referral system NORSE (Network of On-call Referral SystEm).
We have our own electronic health record called PICS which we sold to The Royal Orthopaedic Hospital NHS Foundation Trust and Birmingham Women’s and Children’s NHS Foundation Trust.
We use our patient portal MyHealth@UHB for letters, results and contact with patients, and we have our own clinical portal which is used for image sharing and as a document store. I like to think we are quite advanced as a Trust.
What are the challenges you face over the next 12 months?
We have two big challenges for the next 12 months.
The first is the ongoing work following our merger with Heart of England NHS Foundation Trust. The challenge here is where there are different systems used across each site. Going from paper to digital is one thing, but changing from one electronic system to another will be a challenge. Our plans are to consolidate onto one PAS and then share all our clinical systems going forward.
Another key challenge for us over the next 12 months is to share data with GPs, councils, social care, tertiary services and beyond. The priority is to be able to share information across our region and this is being led by our STP.
It’s vital that all systems use open APIs and follow data standards, gone should be the days when suppliers try to charge you for your data.
Years ago nobody wanted to share information, GDPR didn’t exist and you couldn’t get the right people at the table. The time feels right now to really make a difference with technology and with the new focus from Government I think everyone believes digital should be high on the agenda. A challenge will be resources but that hopefully will be less of a challenge than previously; it doesn’t feel like pushing against the tide anymore.
What advice would you give to other organisations?
Our organisation is a GDE and we are duty bound to share information, our advice to any organisation would be to go visit a hospital and see the software in use. Ask frontline staff what they think and be alone with them! Spend time with Doctors, Nurses and Patients and ask them what they think and ask them if the technology works for them. We are big believers that technology should be designed by and with Doctors and Nurses.
What’s the future of health technology?
There’s a huge opportunity for artificial intelligence and machine learning to push boundaries. Lots of people who use the words artificial intelligence don’t mean what it truly is, we’ve been using data and creating rules for a long time. There is a massive opportunity to utilise the data but you need to know how you will use it, but the potential here is huge.
We are conducting extensive research which uses the large datasets within the hospitals:
A lot of information is ‘hidden’ in text in letters and laboratory reports. We are using natural language processing to extract information and triangulate this against coded data enabling us to present likely diagnoses to clinicians. An ability to use coded and un-coded data will greatly increase what we know about patients, allowing us to run complex clinical decision support rules which prevent mistakes and drive up quality of care.
Another example of innovation research is to study patterns in our data for patients who later become unstable and require intensive care admission. We want to detect these patterns early, before people become very ill in order to prevent deterioration.
A final example is that we are creating a research PACS where many anonymised images can be held. This allows for pattern recognition in patients and where this can be matched to phenotypic information enables machine learning in big datasets.
