Using AI to help identify long stay patients and improve outcomes

New tool uses artificial intelligence to assess hospital data and help predict patients at risk of becoming long stayers, with resulting negative implications

The Accelerated Capability Environment (ACE) supported the NHS AI Lab to develop a proof of concept (PoC) which could help identify patients at risk of unnecessary long hospital stays and the associated negative outcomes. This would allow clinicians to try and avoid this by adjusting treatment plans.

Gloucestershire Hospitals NHS Foundation Trust is the lead acute provider for a population of 660,000. Of all admissions to the hospitals within the trust, 4% go on to stay for 21 days or longer – defined as long stayers – which comprise 34% of all bed stays.

Extended stays lead to more negative outcomes, and long stayers face:

  • an 11% mortality rate (compared to 5% of all admissions)
  • A 23% chance of becoming unwell again after being deemed fit for discharge (compared to 1%)
  • In the over 80s, loss of 10 years’ muscle mass in just 10 days in hospital

Yet many long stays are avoidable, with no medical reason for bed rest, and interventions, such as an increase in walking or daily physiotherapy, are known to reduce overall stay.

Using patient data to highlight risk

The key questions for this commission were whether systems could be taught to identify people at risk of long staying, and whether patient data already collected had enough predictive power to highlight risk. The answer to both questions was yes.

ACE worked with Polygeist from its Vivace community to develop a long stay stratification tool, which could use an artificial intelligence (AI) model trained on 460,000 anonymised records to identify those at risk of becoming long stayers from initial patient data collection.

This produced an immediate long stayer risk score, which could be available to all reception and clinical staff to help avoid known risk factors. For instance, if a doctor suspects a patient to be at a higher risk of becoming a long-stayer, they may opt to avoid catheterisation, admit to a different ward, or immediately refer the patient to a geriatrician or physiotherapist to avoid decline.

Developing a predictive tool

ACE delivered at pace, producing a PoC in 12 weeks. The tool, which was very well received by the trust, as well as the wider NHS, detected 66% of long stayers within the highest risk categories. This has enormous economic, as well as health, benefits, as a single-day reduction in average stay yields £1.7m in savings for Gloucester Hospitals alone.

Following successful completion, the PoC was moved into a limited closed Alpha service phase, and integrated with the trust’s electronic health record system via application programming interfaces (APIs). This allowed further testing against Covid-era datasets, anonymously checking patients who had already been discharged to see if the tool could have helped, where it remained highly accurate.

Accelerated Capability Environment