Progress update on addressing NHS winter pressures through health data research

Across the funded projects, researchers sought to better understand who is most at risk of hospitalisation during the winter months, as well as predicting pressures on hospitals, GPs and other healthcare systems. Identifying what interventions may work to reduce poor health outcomes was also a key focus. Research teams were matched with policy analysts, who played an active role in providing advice and feedback on work to maximise policy relevance.

New insights

Led by Professor Liz Sapey, Director of the Institute of Inflammation and Ageing at the University of Birmingham, researchers have developed a prediction tool to improve the flow of patients through Same Day Emergency Care (SDEC) – a way of treating emergency patients who would otherwise be admitted to hospital. The project highlighted that previously used tools, developed in specific populations, are less effective in a diverse urban centre. The team are now working with trusts to look at the workforce implications of using their new tool and aim to publish their findings in the future.

In another project, Dr Darren Green at Northern Care Alliance NHS Trust and his team explored how machine learning methods could be used to predict risk deterioration in patients presenting in A&E, based on their first recorded vital signs and other early tests upon arriving to hospital. Their initial findings, published in Nature Scientific Reports, show that introducing this modelling method has the potential to reduce alert fatigue and identify high-risk patients with a lower National Early Warning Score 2 (NEWS2) – a way to assess illness severity– that might be missed currently.

Researchers led by Professor Sir Aziz Sheikh at the University of Edinburgh investigated the risk of being admitted to hospital with acute respiratory infections. Through analysing health records on 5.4 million residents of Scotland from 1 September 2022 to 31 January 2023, it was found that younger children, older adults, those from more deprived backgrounds and individuals with health conditions are at increased risk for winter hospitalisations. The study, published in the Royal Society of Medicine journal, builds on the work of EAVE II which was supported through the HDR UK-led Data and Connectivity COVID-19 National Core Study.

Dr Rhos Walker, Director of Strategy at HDR UK, said: “Winter is an especially difficult time for the NHS each year, and this programme has continued to demonstrate how harnessing the power of routinely collected healthcare data can help to better understand and find solutions to the underlying causes of pressures on the health service.

“Building on the progress seen during the pandemic – where initiatives such as the HDR UK-led Data and Connectivity COVID-19 National Core Study made a diverse range of data more accessible to researchers, so policymakers could be provided with the most up-to date evidence– a unique aspect of this initiative was the inclusion of policy analysts working alongside the research teams, supporting the delivery of data insights at pace to inform policy and practice. Much was learnt from undertaking this innovative approach that will be taken forward as HDR UK continues in our commitment to delivering data insights at a pace to inform policy and practice.”

Professor Danny McAuley, Scientific Director for NIHR Programmes, said: “These projects illustrate the value of using cutting-edge research and data to identify the most common winter pressures faced by the NHS and find solutions to help address them.

“The evidence all of the research teams have collected will be crucial in helping to tackle a range of challenges such as preventing illness, freeing up vital time and resources for NHS staff, and reducing hospital admissions and stays.”

Building links between policy analysts and research teams

This programme tested new methods of funding and managing research, learning from how research insights were rapidly delivered and saved lives during the pandemic. Each project was ‘buddied up’ with government analysts working in the DHSC, the Office for National Statistics (ONS) or the UK Health Security Agency (UKHSA).

This new approach aimed to strengthen the links between the research and policymakers, supporting the translation of the findings into impact. George Ball, a policy analyst at DHSC who was paired with the PHLOSARIP project, said:

“I’m a data scientist and it was interesting to see how these researchers were applying machine learning to health data. The academics were very willing to engage, and I had discussions about some of the theory of what they were doing, I was given early drafts of their research results/findings and my feedback was considered.”

The buddying process implemented for this programme was used to act as a proof of principle, and the lessons learnt from this process support closer collaboration with research teams, government analysts, and policy makers going forward.

Although the projects are at different points in the research process and not all outputs have been peer reviewed at present, findings to date demonstrate the potential to help address winter pressures in the future.

Find out more about all 16 funded projects