Three years on from Cambridge’s Epic big bang go-live
- 23 August 2017
Few healthcare IT projects have generated quite as many headlines as the initially-troubled deployment of Epic at Cambridge University Hospitals NHS Foundation Trust. Three years on, Laura Stevens visits the organisation and finds no desire to go back to the pre-digital days – along with some hard-won lessons on how best to implement an EPR.
It’s easy to get lost among the maze of buildings, roads and ambulances that make up Cambridge University Hospitals NHS Foundation Trust. There’s even an airport-style food court. The scale of this world-leading teaching hospital – it covers 74 acres, and has an annual income of £661 million – means it’s no surprise the organisation took an epic approach to its health IT.
Cambridge’s deployment of Epic’s electronic patient record (EPR) was the first implementation of the US supplier’s EPR this side of the pond. To date, it remains the only one. Part of the trust’s £200 million eHospital programme, Epic went live on 26 October 2014.
Difficulties quickly followed. A major incident was declared a week after go-live, due to instability with the EPR. In the longer term, the eHospital project contributed to major financial difficulties at the trust, which in turn led to the resignation of chief executive Keith McNeil, a highly critical CQC report, and the imposition of special measures.
Nearly three years on, and the trust’s digital fortunes have changed drastically. Now one of the 16 acute global digital exemplars, clinical staff seem to have no desire to return to the pre-Epic days.
“Absolutely no way I would go back”
When I sit down with Cambridge’s chief information officer, Zafar Chaudry, and clinical chief information officer, Afzal Chaudhry, the benefits of having all of a patient’s information in one system is emphasised repeatedly.
Epic spans all clinical areas (both inpatients and outpatients) including A&E, critical care and laboratories. Chaudhry, who is the transplant surgeon on call when I visit, says Epic has “every possible thing that I can need”, which means more patient-facing time.
“The administrative burden is eased by the system substantially, and the clinical decisions that are made by the team collectively are much better because all of that information is there.”
The trust’s CCIO is expected to be an advocate, but I also find a belief in Epic’s virtues on the wards. Stephen Wallis, a consultant in elderly medicine, tells me there’s “absolutely no way that I would go back to badly filed paper notes”.
Expected and unexpected patient benefits
Some EPR benefits can be predicted. It’s no surprise, for instance, that paper patient notes have reduced by 99% since the system was implemented.
Charlotte Foster, a senior sister at the trust, says not having to share one set of hardcopy notes has been a huge help. “Whereas before you were running around almost queuing up for set of notes”, she said, it’s “just so easy now, because we don’t have to be fighting over the notes”.
Predicted patient safety benefits have been realised too. Through Epic’s electronic prescribing, there has been a 100% reduction in sedation-related prescribing errors in paediatric intensive care.
However, some improvements come from unexpected areas. Staff in orthopaedics, for instance, dreamt up a multi-disciplinary virtual fracture clinic. This has freed up 4,500 clinic appointments, as trauma nurses and consultants can view the case notes and images simultaneously and reduce unnecessary hospital visits.
Winning hearts and minds
It does seem, however, that staff have gone on a journey to become Epic converts. Foster admits it was “daunting” prior to the go-live as “everyone was very anxious about how it would work”.
But she too wouldn’t go back. She tells me handovers are better, documents can more easily be kept track of, and contending with clinicians’ illegible handwriting is a thing of the past.
“[It’s] just made everything a bit tighter and a bit safer, and made us a bit more organised. I wouldn’t go back”.
Chaudry, who was not in post at the go-live, believes change management could have taken a higher priority during the deployment.
“Maybe the change piece is something we should have focused more extensively on, because how do you win the hearts and minds of everybody?”
He describes the initial problems as “teething pains”, as “this is a massive change programme put into an organisation that’s the size of a city” with 13,000 users adapting to a brand new system in a big-bang go-live.
Investing money in managing the change would be his advice to other trusts implementing such a system. “[As] a lot of organisations will tell you, that can be as much as your programme in terms of cost”.
More than just crossed fingers
Chaudhry points out there’s no current mechanism in place for trusts to pre-emptively support others during an EPR go-live.
“There’s no way that the system can at present can help a hospital to go-live, in as much as if a number of hospitals around us had taken say between 0.5% and 1% of our activity for six weeks, that would have helped us.”
His example is fellow GDE and neighbouring hospital, West Suffolk NHS Foundation Trust, when it went live with Cerner last spring. “When it came to the day of go-live, the best that we could do was say, we’ve all got our fingers crossed, off you go”.
If the NHS wants to achieve its paperless goals, Chaudhry believes this is something that needs to be addressed.
“It isn’t possible just to say to every hospital just do this along with your regular day job because you just can’t do so many things at once.”
Epic fast followers
Chaudhry speculates that the GDE and fast follower scheme could potentially be the framework by which trusts can help others when implementing IT systems.
Cambridge was chosen in the second wave as a GDE and, despite delays to funds being released, has now received its first tranche of funding. The main clinical aim is to be a HIMSS stage 7 hospital, and the key themes are about sharing data, high quality care, and ensure systems are safe, Chaudhry tells me.
Each GDE has to select a fast follower, and provide a blueprint for that trust to help them with their digital ambitions and IT procurements.
Cambridge has not yet chosen its fast followers, but the trust has been in conversations with the three other English trusts that have selected Epic as a preferred supplier: University College London Hospitals NHS Foundation Trust, Great Ormond Street Hospital for Children NHS Foundation Trust and Royal Devon and Exeter NHS Foundation Trust.
While mainly happy with the fast follower initiative, Chaudhry says he has concerns with the notion of Cambridge “doing due diligence on another NHS institution”.
Potential to be “absolutely outstanding”
I end my trip with a visit to the noisy intensive care unit (ICU), where patients are hooked up to ventilators and an orchestra of beeps provides the working soundtrack for clinicians in green scrubs.
In the ICU an enormous amount of data is produced continuously and Andrew Johnston, consultant anaesthetist, tells me how the team has to assimilate the data to treat patients who are severely unwell.
Epic, with all the information in one place, is described as “completely invaluable”. By having the test results, charts, consultant notes, Johnston says the impact on patients has been huge.
“In terms of patient safety and the care we give the patients, it’s made an enormous difference, especially in critical care which is such a data-rich environment. It really has helped a lot.”
The Epic deployment, and staff’s relationship to it, has been on quite a rollercoaster since the go-live. But there now seems to be a confidence that the trajectory will only be upwards from here on out.
As Johnston says: “It was not necessarily perfect when it came in, but it’s a lot more perfect now and has the ability to be an absolutely outstanding system over the coming years.”
11 Comments
There was never any instability in the EPR. all the early difficulties were hardware related which was nothing to do with Epic.
The project (largest single IT deployment in the NHS) was delivered on budget and on time (well 24 hours late to be accurate). yes it added to the financial difficulties at the time but so would have a simple PAS replacement given the financial climate across the NHS which persists today.
At the time, CQC didn’t understand the IT nor what we were doing. CUH was already one of the safest hospitals in the country at the time, and now it’s even safer. this is a journey everyone will need to go on sooner or later, and the GDE program will help ease the way considerably by sharing all the learnings and helping to set realistic expectations of front line staff and regulators/inspectors alike.
This is where the Core and non Core aspects of a hospital come into picture, in my opinion, IT Organization for a hospital is a part of its non core services, this can probably be Outsourced to Managed service firms who can keeps the IT lights on for Health Systems. From a hospital standpoint they need to set up a Liason in the form of Medical Informaticists who can understand the needs of the hospital and get them translated into requirements and get it delivered from these managed services firms.
Outsourcing… has anyone *not* found that this overtly diminishes quality?
I’m not just talking about IT but it applies here too. I’ve seen it throughout the NHS. It is extremely difficult for one organisation to hold another accountable for something as subjective as the quality of their work and, without a vested interest, why would any third-party that purely provides a service, choose to devote their resources toward reproducing in-house quality rather than just ticking the functional delivery?
Epic is a complete suite of application which can cater to all sections of the hospital Scheduling, OR, Path, Obgyn and also customization is a best aspect of Epic. Once can get valuable output in terms of reports however you need a well trained IT workforce to run the show and thought leadership to lead the IT organization for the hospital.
Unlikely – Epic runs on an Intersystems Cache database. Not sure what Lorenzo uses but very much doubt it’s that.
They also have different aims – Epic is a complete suite including clinical applications (pharmacy, pathology etc) whereas Lorenzo is purely a PAS and EPR. Opinion from the US also seems to be that Epic works great but you need a small army of back office workers entering data and managing the system to keep it tidy.
“Opinion from the US also seems to be that Epic works great but you need a small army of back office workers entering data and managing the system to keep it tidy.”
Incorrect, the opinion in the US is that healthcare workers are burning out turning into data entry monkeys rather than caring for patients.
I don’t think it’s easy to compare Lorenzo to other more clinically focused products. Agree with Patricia’s comments and probably put trakcare at same level as Cerner (above Lorenzo).
Happy to hear EPIC is enabling connected care and rescuing apointnents.
The reason I ask is I am curious, does either give value for money? Estimated costs for Lorenzo was £9bn by 2013 but guess it would have reached £10bn by now, plus the ‘bribes’ hospitals are being given to take Lorenzo.
On a per head basis and assuming every person in the UK (65m) is on Lorenzo then the cost per person is £151- not including running costs/migration to system.
What would EPIC cost (or any other system with similar functionality) cost by comparison?
If Lorenzo and EPIC do similar things, do they talk to each other. If a patient is on Lorenzo and treated in a hospital that uses EPIC can the information be extracted?
Epic is better. Unless you got Lorenzo under the NHS Digital funding deal, you’re paying a lot for a rather shaky system but it is configured to work for the NHS, which Epic certainly wasn’t.
Epic offers slicker, more project team friendly, tools for data capture to make all that pointless CQUIN reporting much easier.
But the huge rub is that Epic is much more expensive. If you have deep pockets, you’d choose Epic. If you don’t, you’re probably better off with Cerner IMO. It’s come a long way in the last few years.
I have worked on projects for both so can make a fair comparison.
you’ll be wanting usability scores and a free and transparent market place next David. Clearly you are some sort of dangerous heretic ; )
Lorenzo vs. Epic
Do they do the same thing? Which one is the best value for money? Advantages of each other?
Comments are closed.