Health Data Research UK (HDR) and the National Institute for Health and Care Research (NIHR) are launching research projects to examine how winter pressures can be eased by the NHS.
By using data-driven approaches, the research aims to identify pressures, understand the causes and develop ways to overcome and avoid them.
16 projects are to be carried out in total and are expected to produce results by the end of March, with findings set to be published later this year.
In a project led by Professor Darren Green at Northern Care Alliance NHS Foundation Trust, the focus is on early and safe identification of the right bed for the right patient on exiting emergency departments. The study will use artificial intelligence models to analyse if certain groups of patients can be prioritised for admission to specific wards after their arrival; which patients are at risk of imminent deterioration; if there any unsafe discharges and which vulnerable patient groups are more affected by winter pressures.
Dr Dan Todkill at University of Warwick is leading a project on machine learning to forecast the peak and magnitude of winter healthcare pressures due to respiratory syncytial virus (RSV). This project is using machine learning techniques to predict the peaks in RSV demand, how big the peaks will be and how quickly they will reduce.
A project led by Professor Elizabeth Sapey at University Hospitals Birmingham NHS Foundation Trust will focus on improving patient flows through acute medical departments through better patient selection for Same Day Emergency Care (SDEC). A scoring system is to be developed based around diverse patient population to identify patients who could benefit from SDEC. It will use data which will be available to clinical staff within four hours of the patient arriving to hospital.
Dr Honghan Wu at University College London is leading a project focusing on using rare disease phenotype models to identify people at risk of COVID-19 related adverse outcome. It will tackle challenges by bringing together knowledge about rare diseases, using new data science technologies to explore COVID impact datasets, and create an accurate identification system for people living with rare diseases.
In a project led by Professor Suzanne Mason at University of Sheffield, the focus is on understanding demand for emergency care using regional routine data from emergency department and acute hospital admissions. It will examine anonymised routine data from NHS emergency departments throughout the country. The project will also analyse emergency admissions to understand who is being admitted to hospital and what happens to them.
Dr Nazrul Islam and Dr Hajira Dambha-Miller at University of Southampton are conducting a study on which combinations of multiple long-term conditions are associated with the greater risk of hospital admission over the winter season, and to what extent the COVID-19 or influenza vaccinations modify that risk. The study will use an anonymised electronic database of patient records to establish statistical tests with mathematical methods, and machine learning approaches to highlight which combinations of disease are the most likely to have a high risk and if it can be reduced by vaccination.
A project by Professor Amitava Banerjee at University College London will centre on improving characterisation, prediction and intervention for COVID- and influenza- related morbidity and mortality. Using national electronic health records, the aim is to explore how COVID-19 and influenza affect hospitalisations, death and long COVID.
Professor Michael Boniface at the University of Southampton is looking at “DS4SmartDischarge: Data Science Informing Complex Discharge Winter Policy.” By using computer algorithms, the aim of the study is to understand what causes patients to be delayed. This includes health status, procedures, treatments, and care needs assessment.
Professor Sir Aziz Sheikh at the University of Edinburgh is conducting a study on describing, characterising and predicting winter respiratory accident and emergency attendances, hospital and intensive care unit admissions and deaths. The aim of the project is to extend Early Pandemic Evaluation and Enhanced Surveillance of COVID-19 (EAVE II) to the most at risk of serious outcomes when developing an infection within their airways.
A project focusing on predicting hospital length of stay in acute respiratory infections patients is being led by Dr Dan Burns and Prof Matthew Inada-Kim at the University of Southampton NHS. The study aims to identify severely ill respiratory patients at higher risk of longer hospital stays and find ways to predict whether a patient will experience the same when they have a similar infection, with researchers to explore relationships between length of stay and age, gender, ethnicity, diagnosed diseases and staffing levels.
Dr Alice Lee, Ian Sinha and Ian Buchan at the University of Liverpool are conducting a study on data-intensive action on winter pressures through healthcare resourcing and access in Cheshire and Merseyside, with focus on children, social prescribing and telecare. The project will focus on identifying children who are at risk of needing emergency healthcare for respiratory conditions, and measure the impact of existing pilot projects that address respiratory health inequalities in Cheshire and Mersey.
Another study will focus on linking individual-level clinical records with the 2021 Census data, in order to identify patients at high risk of hospitalisation and death during winter and examine the relationship between poverty and clinical vulnerabilities.
A study focusing on comparison of risk factors for hospitalisations and death from winter infections will see researchers learn more about the factors which make a person more likely to experience an infection in winter, and more likely to become severely ill as a result. The study also aims to expand understanding of which risk factors are associated with just one infection and which are linked with multiple infections.
The next study is to focus on identifying early warning signs that an organisation is under pressure by investigating potential patterns in primary care data, and how those measures of pressure may vary between practices and over time. Additionally, it will focus on whether patients experience worse outcomes if a practice looks to be overwhelmed.
Researchers from King’s College London are to use artificial intelligence to understand how the cost-of-living crisis impacts on children’s health, with Dr Martin Chapman using AI to develop understanding of how preventative interventions can improve the health of children and young people in the UK.
Finally, the SIREN study, previously focused on developing understanding of the immunity to SARS-CoV-2 conferred to healthcare workers from infection and vaccination, will support further research. The dataset is accessible to others via the Co-Connect platform and data collection in the form of active testing is funded until March this year. The new study will expand testing to influenza, respiratory syncytial virus and SARS-CoV-2 to create a new surveillance tool integrating incidence and healthcare worker sickness data, pertaining these pathogens.
The Department of Health and Social Care (DHSC) is sponsoring the projects. Dr Mary De Silva, Deputy Chief Scientific Advisor at DHSC, said that “these projects aim to harness the power of routinely collected healthcare data to understand what is causing the pressures, and crucially to provide new solutions that can be swiftly turned into working practice.
“We’re also testing new methods of funding and managing research, learning from how we rapidly delivered research results which saved lives during the pandemic. For these projects, it was just three weeks from launching the call for proposals to awarding the grants. Every project is ‘buddied up’ with government analysts working in the DHSC, the Office for National Statistics and the UK Health Security Agency, to ensure that the results feed immediately into policy and practice.”