Summary
The NHS has a significant waiting list of patients requiring elective surgeries. It is a priority to shorten the lists, but help is needed to ensure that patients most at risk of harm or mortality are at the right place on the list. This is where Artificial intelligence (AI) can be used with significant benefit.
"
I, like every surgeon, get a report each month detailing the complications we had and what we would have expected, based on the patient’s background. I saw waiting lists balloon through the pandemic, so went back to colleagues at C2-Ai, the technology firm we have worked with for many years now, and asked if we could get a forward-looking report predicting patient need, rather than looking at retrospective outcomes. And they could.
"
Professor Rowan Pritchard-Jones, Executive Medical Director at Cheshire and Merseyside Integrated Care Board
What the project involved
To reduce the elective backlog in Cheshire and Merseyside, the Innovation Agency, the AHSN for the North West Coast, identified three trusts that would benefit from the use of an artificial intelligence system.
While there is no shortage of data available to clinical teams to help them prioritise patient care and reduce waiting lists, it is often difficult to optimise the use of this information.
In particular, current methods of prioritising patients waiting for elective surgery do not identify individuals who deteriorate more quickly because of their pattern of morbidities, so a more accurate system for assessing such patients needs to be a priority.
This is where Cambridge-based C2-Ai’s Patient Tracking List (PTL) triage system can support hospitals.
It takes many factors into account, including social determinants of health. Its algorithm uses data from more than 200 million records from 46 countries from the last 30 years to review individual patient health profiles, as well as the type of surgery they face, to estimate the risk of complications during or after surgery.
This enables the surgical team to take steps to reduce these risks from happening or to prepare expert treatment if it does.
Outcomes
A report* studying the impact of PTL on 11,837 patients listed for elective surgery in Cheshire and Merseyside found the tool accurately predicted the risk of mortality and complication.
Qualitative feedback
- Users valued the tool’s accuracy and the time saved by replacing an onerous manual process.
- Waiting list managers said 98 per cent of surgeons agreed with the prioritisation metrics.
- There was anecdotal evidence of a 15-minute saving of surgeon time per patient whenever the waiting list was re-prioritised.
Elsewhere, the system has been successfully deployed in many NHS trusts. NHS England evaluation shows:
- 125 bed-days freed up per 1,000 patients on PTL
- an eight per cent reduction in emergency admissions
- a 100 per cent reduction in the avoidable cancellation rate
- a 27 per cent reduction in long-waiters and the highest urgency
Provisional findings from other sources show a three-day reduction in the average length of stay, post-operative avoidable harm events/ICU admissions minimised, and the conversion of many inpatient stays to day cases.
*A tool to prioritise patients waiting for elective surgery: an implementation report, corresponding author Mr Videha Sharma of the Department of Renal and Pancreatic Transplantation at Manchester University NHS Foundation Trust.
Next steps
The Innovation Agency is working with C2-Ai to scale up to meet the UK’s national health and care priorities.
The company is extending its work into Manchester, Yorkshire, the Midlands, and London.
