AI and Evidence

Our journey pioneering predictive health technologies began in 2016 with a 12,000-patient randomised clinical trial, which was published in the European Journal of Emergency Medicine. This first study demonstrated that tele-coaching reduced emergency admissions for high-risk patients.

Building on this initial study, we incorporated sophisticated AI and machine learning approaches to unlock insights within healthcare data at scale. Our research includes several randomised controlled trials, real world evidence studies, health economic studies and patient & public & engagement assessments, and represents global leadership in applying advanced analytics on routine health records.

We're proud that robust evidence and research is at the heart of our work. Our flagship trial is an extensive NHS trial of 1,800 patients, with guidance from the Nuffield Trust.

Published in the British Journal of General Practice and Emergency Medicine Journal, this trial validated our AI platform's ability to identify and predict risks, enabling targeted nursing outreach to improve outcomes.


We continue to invest in research partnerships and projects advancing equity, fairness, and transparency in AI-enabled care.

Our journey

Our work began by clinically proving the value of tele-coaching, progressed to honing AI predictive capabilities and now we spearhead research expanding the real-world impact of AI. Our publications, partnerships and products demonstrate our commitment to rigorous, ethical innovation.

Publications

See our latest publications, reports and posters to find out how we use artificial intelligent and machine learning algorithms to make a difference. Click on a box below to view and download our work. Questions? Just let us know by emailing [email protected]

(2023) Impact Summary Report: Key findings on population health from an AI prediction and health coaching model

A brief report produced by HN summarising the main highlights from our RCT published by Emergency Medicine Journal. Features a patient story, foreword by Joachim and infographics

Source: Emergency Medical Journal

(2022)-Peer-reviewed findings from our randomised controlled trial: mortality impact of AI intervention

To investigate how an AI case-finding and clinical coaching intervention impacted mortality and how this impact varied by age, gender, and deprivation status.

Source: Emergency Medicine Journal (EMJ / BMJ)

(2022)- The impact on primary care of a case-management intervention for reducing emergency attendance: randomised control trial

The impact on primary care workload of case management interventions to reduce emergency department (ED) attendances is unknown. The aim of this trail is to examine the impact of a telephone-based case-management intervention targeting people with high ED attendance on primary care use.

Source: British Journal of General Practice

The AI-guided clinical coaching intervention that prevents 34% of A&E attendances and significantly reduces health inequalities

A case study developed with Paddy Hannigan, Clinical Chair Stafford & Surrounds CCG, SRO Staffordshire & Stoke-on-Trent ICS Digital Programme (2022)

Patient feedback on the benefits of artificial intelligence guided clinical coaching

A report documenting how HN’s approach helps improve patients’ mental and physical wellbeing, reducing the need for unplanned emergency care and hospitalisation. (2022)

Reducing the high healthcare demand of a few individuals

Professor Matthew Cooke talks about how the demands on healthcare systems can be reduced by focusing on a select cohort of the population (2021)

Intensity and dynamics of the top 1% and 5% highest users of primary care

A poster of our analysis from Attenborough Surgery in Hertfordshire (2021)

Ensuring fairness and inclusion in AI-guided patient screening

A Q&A with Sara Reis –PhD Medical Imaging & Computing, Data Scientist at HN exploring how machine-learning approaches and data can be used to identify those most in need of care but often under-served.

NHS at Home: Are we going to implement a new model of care or populate the car boots of the future?

An in-depth analysis of how NHS at Home can be a practical approach to rapidly creating additional capacity and assist a significant number of patients at high-risk, long-term conditions with HN's CEO Mark England.

Personalised Care Group system

A case study of our York deployment, written by the NHS England Personalised Care Group. (2020)

Hospital at Home by the HN Advisory Board

Professor Matthew Cooke, Emergency Department Clinician & Dr Steven Laitner, GP and Public Health Consultant take a comprehensive look at how NHS care can be delivered at home, whether by hospital outreach, or by increasing capacity and capability in the community. Whilst establishing a safe and effective ‘early supported discharge’ process and the challenge with which the NHS has wrestled for decades.

Putting patients first: Using AI and clinically led health coaching to manage unplanned demand in the NHS

Produced to explain the benefits of clinical health coaching. (2020)

Predicting and preventing avoidable urgent and emergency care: measuring impact across activity, mortality and cost benefits

Poster produced by East Kent NHS analytics team – winners of Association of Healthcare Analysts’ Team of the Year award. (2019)

Artificial intelligence enabled tele-health coaching changes patients’ self-reported patient activation and health outcomes

(2019) M. Wieske Thesis on PROMS, 2019 UCL master’s thesis analysing the PROMS impact of Clinical Coaching.

(2016)- A case management intervention targeted to reduce healthcare consumption for frequent Emergency Department visitors: results from an adaptive randomized trial

To investigate whether a telephone-based, nurse-led case management intervention can reduce healthcare consumption for frequent Emergency Department visitors in a large-scale setup.

Source: European Journal of Emergency Medicine, 23(5), 344–350

(2013) - A telephone-based case-management intervention reduces healthcare utilization for frequent emergency department visitors

To determine whether a nurse-managed telephone-based case-management intervention can reduce healthcare utilization and improve self-assessed health status in frequent emergency department users.

Source: European Journal of Emergency Medicine, 20(5), 327–334.

Working in partnership never stops

Since 2019, we've partnered with UCL's Health Data Science MSc to supervise student research projects. Students work with our data scientists, gaining real-world experience. We also invest in ongoing R&D with partners. This guides development through independent evaluation and peer review.