AI-driven predictive analytics can help the heaviest users of health services to self-care and save the NHS from disaster, says HN’s Joachim Werr
A new set of data published by NHS England this month shows a staggering number of patients waiting 12 hours from their time of arrival at A&E.
Of the 1.2 million people arriving at A&E in February, 125,505 patients (10% of all attendees) waited at least 12 hours in A&E waiting rooms to be admitted, sent for treatment or discharged home. Until the publication of this new data by NHS England, existing recording methods had suggested only 3% of patients faced such lengthy waits.
The implication of long A&E waits on patient care is clear and harrowing, with around 500 people dying each week because of delays and problems with urgent and emergency care.
Modern healthcare systems across Europe are buckling under the strain of post-Covid challenges characterised by dramatic increases in patient demand, a lengthy care backlog and a weary workforce.
As the NHS prepares to mark its 75th anniversary in July, it’s important not only to reflect and look back at the marvel of the health service, but also to consider what the future may hold. We’ve reached a watershed moment for healthcare delivery. 2023 must be the year for an AI-driven breakthrough in the way we deliver healthcare. How else should we tackle the devastating reality that one in ten people attending A&E will wait over 12 hours for care?
Clinically evidenced and highly effective
To help the NHS step off the precipice of disaster, a radical new way of thinking is needed.
The use of AI in healthcare – specifically AI-powered predictive analytics – offers an incredible and sustainable approach that can improve the allocation of preventative care resources, reduce healthcare costs and deliver better patient outcomes.
At HN, we’ve taken important first steps in establishing a strong evidence base for our work. Through a lengthy and robust randomised controlled trial – delivered in partnership with the Nuffield Trust think tank – we have shown huge impact for patients and health systems.
New approaches using technology and AI can make a difference in the short and medium term. A notable example is our continuing partnership with Staffordshire and Stoke-on-Trent Integrated Care System (ICS). In 2016, when we first began our trial with Staffs ICS, the aim was to look at how existing patient data could be used to predict those most likely to attend A&E or need hospital care in the near future, and intervene with targeted clinical coaching to reduce their dependence on A&E and GP services.
Using HN Predict, Staffordshire was able to identify and prioritise people with risk of worsening health conditions in real-time by analysing their patient records, before they need hospital care. The algorithm targeted individuals who were likely to consume three or more acute hospital bed days in the next six months or require an increase in GP-led care. These patients were contacted and supported by remote scalable tele-coaching.
The results of the trial have been overwhelmingly positive: a 35% reduction in A&E attendances on average and a 30% reduction per patient in the average total hospital care cost. The success of the trial has been down to the relationship between Staffordshire and HN and the ability to co-design the solution together. Only by opening a strong and honest two-way dialogue, and truly understanding the challenges our partners face, can we solve the perennial problems of unrelenting care demand.
Taking predictive care to the next level
Over the years we have demonstrated that we can find, engage and impact the top 5% of hospital users, who take up 30% of all hospital bed days. With this information, we can positively impact every ICS in the country, freeing hospital capacity, avoiding elective and non-elective admissions, and ultimately improve patient safety and health system efficiency.
Following a successful wave 1 trial, our new NHS recovery offer will see HN offer up to £900,000 in total to systems who are interested in being wave 2 partners.
NHS leaders and managers face complex and numerous demands. They need a high-impact, low-disruption solution that is truly a game-changer in how unplanned care is viewed. HN’s predictive engine uses routine healthcare data and does not require extra data collection. Its implementation does not disrupt existing clinical pathways and can be smoothly integrated among existing providers.
Ultimately, our approach boils down to a commitment to deliver personalised health and care services to the right people, at the right time.
It’s time to focus on prevention-based healthcare to ensure services are being used effectively, and to empower patients to become effective self-managers.
By shifting towards a predictive model – identifying high-intensity users of health services and working with them so they can self-manage their conditions – we can create a high-quality, sustainable healthcare system fit for the future.
Joachim Werr is the founder and executive chair of HN
This is indeed the future, but to unlock its full potential especially using previously unrecognised patterns of co-morbidities requires a full longitudinal patient record, collected at scale. Enter the ICS…
Join the discussion - comment on this article