RFID tracking at Royal Glamorgan Hospital

Project aims

To launch the Digital Health Ecosystem Wales (DHEW), the team consulted Health Boards to identify their key challenges. Asset tracking – knowing where equipment has been left in a hospital – came up as a priority. In March 2018 DHEW held an event bringing together brought together NHS and industry to get under the skin of the issue and identify possible solutions. It was clear there was a range of technologies and options available to NHS organisations.

Following the event, the Welsh Government funded a pilot project to examine the impact of asset tracking technology on patients, staff and equipment The aim of the project was to identify where and what savings in time and resource could be made if staff could track assets more effectively.

Helen Northmore, Head of Innovation Adoption (Digital & AI) at Life Sciences Hub Wales, commented;

“There was a feeling that nurses and porters spend a lot of time in hospitals looking for equipment, which is impacting patient discharge and patient care.

“We wanted to understand the benefits of using technology to track equipment, so alongside securing a technology partner, we also commissioned an external evaluation to determine the benefits of adoptiing this technology.”

Getting the project off the ground

The Royal Glamorgan Hospital was chosen as the pilot hospital, and DHEW worked with the RGH team to appoint a technology partner and external consultant to carry out the evaluation.

As DHEW managed the procurement, the team decided to trial an outcomes based procurement process rather than a pre-determined technology specification.

Rob Salter, RFID project manager at Cwm Taf Morgannwg Health Board, said:

“Tracking software is a well-established technology, but we didn’t want to replicate what could be bought off the shelf, as they were all silo based. With DHEW taking the lead on procurement, we were able to go into procurement with a blue sky specification, which focussed on what was possible, rather than focusing on our problems which would drive us down a silo solution route. The specification was loose, which would never have been allowed in NHS procurement, and allowed suppliers to pitch their best solutions.”

Kinsetsu, a Belfast based software developer, was appointed to design and implement a passive RFID solution (KTRACK).

Helen commented:

“Kinsetsu were appointed because they really understood what the project was looking to achieve in terms of trialling the technology and assessing its impact. In addition, they proposed a different financial model, basing the costings around the number of zones in the hospital, rather than the number of tags. Taking this approach to the financial model, meant the project could be scalable if the hospital subsequently wanted to take on the technology after the pilot and also allowed us to understand exactly what we could test during the pilot.”

The pilot

Initially the plan saw passive tags placed on bladder scanners, wheelchairs and rental beds, along with software to provide staff with real time information on where the equipment was.

Helen said:

“Wheelchairs were a key piece of equipment as they are frequently needed and can take time to find. Bladder scans were chosen because they can impact discharge if a patient is ready to go home. Rental beds were also chosen as the hospital is charged per day on these beds - if the patient has been discharged you don’t know where the bed is, you’re still paying for that bed. There was a feeling this could provide cost savings.”

The introduction of zoning throughout the hospital meant that more items of equipment could be tagged. Zones were developed in the Emergency department and X-ray, as well as certain wards, where multiple readers were placed at exit and entry points and corridors to assess the direction of travel.

As Kinsetsu were keen to test the technology for themselves, the pilot evolved to include patient tracking, where 1500 patients were tracked through the A&E department.

On this additional element, Rob said:

“We’ve got a model of data which is exceptionally granular based on the movements of 1500 patients, from the moment they come into A&E, flow through the department and the exact times they spent in certain areas. The data model will be very important for us going forward. We can test systems with this real live data to see if the software holds up to having the variants that occur in healthcare.”

Project challenges

Although the pilot was initially planned to be delivered in 6 months, expected and unexpected issues arose including set up, communication and project management. Kinsetsu worked with the Board as challenges arose and extended support of the pilot to ensure the project had every possibility of success.

Helen said:

“The infrastructure for this type of project took time to set up. Tagging all the equipment, merging the technology with the current hospital systems, and implementing GS1 identifiers across all rooms in the hospital meant that the set up took longer than expected.

“Stakeholder communication and engagement was a challenge that was expected, but what the team hadn’t foreseen was that the timescales for the project saw implementation land in August, which is a big change over time for doctors who move onto different rotations. It meant that those that had been consulted with and engaged at the start of the project moved on, and the new people coming into post had no prior knowledge. This meant more time had to be spent getting buy-in from new team members.

“The capacity to project manage this pilot also caused a delay, as Rob, the project lead, also had his day job to do alongside this pilot. What we learned was that a dedicated project manager for these kind of pilot projects is crucial.”

Project outcomes

The pilot uncovered some interesting but unexpected outcomes for the hospital. It identified that while being able to track levels of equipment was useful, it was actually the planning capabilities that it enabled which brought far more value.

Rob commented:

“The data showed us that the busiest time for the movement of equipment was in the middle of the night. We’re now able to model the equipment on this going forward. The ability to plan your levels of equipment and reduce capacity in some areas, brings far more value than the tracking itself.”

The pilot also showed that being able to claim that tracking items results in time savings for staff is not as clear cut as expected.

Helen commented:

“What we were hoping for was that there would be a really strong financial argument as to why the technology is worthwhile from a time saving perspective, but actually it’s not as clear cut as that. There are some strong financial benefits around managing stocks of equipment, but the time saving is much harder to quantify.”

Rob added:

“One of the benchmarks for the off the shelf software was ‘our system will save you 30 minutes a day per nurse’. We went into the pilot with that as a baseline KPI for understanding how tracking could benefit us. When we looked at our granular data, we uncovered that nurses are doing a dozen things at a time, so you can’t say they’re wasting 30 minutes looking for a piece of equipment, because they aren’t, they are multitasking and doing other things. We proved that nursing time is a highly multitasked time so difficult to quantify, and that modelling can’t base a task being from A to B. It’s A to Z with 24 tasks in between.”

Significantly, what started off as a project about asset tracking, evolved into a project around how to design services.

Helen commented:

“If you’re tracking patients and know that x ray is becoming a choke point, then you can redesign how you’re doing things. The ability to redesign services based on the data was something we hadn’t appreciated at the start of the project but it’s something that has really benefited the Health Board.”

What’s happened since the pilot

At the end of the pilot, the hospital were really pleased with the technology and took it on permanently, expanding it across other sites within the health board and thinking about ways it could be used to track other items, such as oxygen cylinders.

The API developed for this pilot is now being adopted by other health boards for a range of projects.

Since the onset of the COVID19 pandemic, the technology has proved to be invaluable due to the ability to track patients and equipment.

Rob said:

“With COVID19, we needed to identify if equipment had been in a COVID area and we also wanted to supply the infection control team with data to show if equipment was moving across areas. When you’re trying to track an outbreak, you need every tool in the box to understand what the link between two wards is in tracking outbreaks. We made some simple modifications to the tracking software so that we can change the designation of an area from COVID to non-COVID and suspected COVID. Each time a piece of equipment moves in or out of one of those areas, our asset management systems update to keep a history of when it went near COVID patients, so we’ve got that infection control tracking. It was something that we never anticipated but something that has been invaluable.”

What’s next

The rollout across Cwm Taf Morgannwg Health Board continues and other Health Boards across Wales are looking at how it can be implemented.

Rob said:

“The main progression continues to be bringing on board new stakeholders. They come to me saying ‘I really want to know where this piece of equipment is’, but it’s more than just a question of where it is. I’m encouraging them to think about ‘why they need to know where it is’ and ‘what other bigger problems can we fix by using that information’. Really, we shouldn’t be calling it RFID tracking as it does a lot more than just track. The audit function that it gives you is huge and if you use the audit function properly it surpasses the tracking functionality.”

On the future, Rob says:

“My long-term goal is that RFID technology is acknowledged as a key element in artificial intelligence for hospitals. So, if do we go down the route of having PAS systems based on AI, when making those difficult decisions on how patients should flow through a hospital, I’d like to see RFID feeding data into the system. It will mean that AI can see down to the millisecond what the movement of things looks like in the hospital and that can be factored into management strategies.”

Check out the full project timeline