• Trial reports significant reduction in emergency department attendances through AI-targeted health coaching 
  • CCG commissions new care model in response to findings

Commissioners in Yorkshire are to roll out a scheme which used artificial intelligence to help cut A&E attendances by a third in one patient group which received health coaching.  

Vale of York Clinical Commissioning Group green-lit the scheme following a randomised control trial featuring around 1,000 patients treated at York Teaching Hospitals Foundation Trust over the last four years.

The trial used AI to identify patients, often with long-term conditions, who were at risk of an unplanned hospital admission. These patients then received coaching from nurses for up to six months to help them take greater control of their health.

The trial reported a 30 per cent reduction in unplanned hospital admissions and 25 per cent decrease in planned admissions within the patient group which received the intervention, compared to the cohort which did not.

The CCG will now fund health coaching for up to 1,800 patients.

HSJ understands the initative is being closely watched by national leaders charged with tackling lengthening hospital waits.

AI finds at-risk patients

Health coaching has been used in some parts of the NHS for nearly two decades, but the use of AI to pinpoint which patients should be targeted is a newer phenomenon. The York trial is significant because it is one of the largest, and longest running, randomised control trials used for this purpose. 

Fiona Bell, the CCG’s lead officer for primary care, described the trial’s findings as “really impressive” and said the service had had a “huge impact on patients’ everyday lives”. 

She added: “High emergency admissions are a system problem, not just the responsibility of secondary care. We need to find a way of supporting patients to help them understand their conditions better, make life changes which will improve their health and know which services to access when they need support.”

The York project is part of a wider ongoing randomised control trial which is also being run at five other NHS trusts (Mid Essex Hospital Services, Royal Wolverhampton, University Hospitals of North Midlands, University Hospitals of Derby and Burton FT, and East Kent University Hospitals FT).

750 patients are part of this wider trial and its first findings are scheduled for publication in April next year.

The interim findings were presented by Health Navigator, the scheme’s technology provider, at the Health Excellence Through Technology conference earlier this month. Health Navigator chief executive Joachim Werr claimed that 1 per cent of the UK population accounts for 50 per cent of non-elective bed days, which means there is “huge potential for new models of care such as this” to reduce avoidable admissions.

Story updated at 4.17pm on 28 October to clarify that the 30 per cent reduction in emergency department attendances was within the patient group which received the intervention, compared to the patient group which did not.