Innovation in data modelling improved COVID-19 care at NUH | Latest news

A ground-breaking project enabled healthcare staff at Nottingham University Hospitals NHS Trust (NUH) to use real-time, anonymised data to make critical decisions in managing the care of patients hospitalised with COVID-19 throughout the Pandemic. This was facilitated by the wish throughout the hospital, wider NHS and regulators to facilitate the use of data to help in managing this unprecedented challenge.

A vital component of the hospital’s approach to COVID-19, the project brought together front-line clinical teams at NUH joined forces with data scientists, public health experts, radiologists, researchers and statisticians from the University of Nottingham. The result was the development of a real-time model for managing and predicting the rapid rise in COVID-19 patients during the first wave of the Pandemic in 2020. A range of different data were combined and analysed to provide a complete picture of the impact of COVID-19 on patients being treated in the hospitals. 

Part of this was the analysis of data from 800 chest X-rays. Scoring systems were developed to estimate the severity of lung involvement in COVID and these were used in combination with clinical scoring systems to predict the severity and deterioration of COVID-19 patients and their outcomes. This enabled clinical teams to not only treat the patients already admitted, but to predict the numbers of patients likely to need hospital care and provide more information on the seriousness of their condition.

Now the team in Nottingham, whose research has been published in the journal Radiology*, believe the model they developed could have wider application for the NHS.

Dr Iain Au-Yong, a Consultant Radiologist at NUH who was involved in the project, and a co-author of the Radiology research paper, said: “It was a real privilege to be involved in this work; the collaborations between the various teams worked very well during the Pandemic. We’re proud that the work of our radiology consultants who analysed the chest X-ray data made such a useful contribution to the project overall, in terms of researching patient deterioration and outcomes from COVID-19.”

Dr Andrew Fogarty, chest physician at Nottingham City Hospital, one of the team of experts who created the model, said:

“This was a historical moment. In World War II, radar was developed rapidly when we needed it, and this project was our local health service equivalent, leading to the development of a database that was previously considered impossible.

“During the early phase of the Pandemic in March and April 2020, admissions went from a trickle to a tsunami of patients needing help. Collectively, NUH did cope with COVID-19, but it was extremely challenging for all of those in a patient-facing role.”

Dr Fogarty added: “We realised we have expertise in using data in Nottingham across the NHS and University sectors. So, by marrying these together, we were better placed to guide decisions to manage the Pandemic.”

Dr Mark Simmonds, Divisional Director for the Medicine Division at NUH and a Critical Care Consultant, was a driving force behind the project. He said the benefits for patients were seen very rapidly:

“This project was absolutely vital to us. We used the insight from the data to plan which wards needed to be opened and the staff resources required during the first wave of COVID-19, as well into the second wave in the autumn.”

Dr Simmonds added: “The accurate, timely data analysis we had about our COVID-19 patients helped us cope with demands in real-time but also to plan for the future. 

“At the beginning of the Pandemic, that planning was hour-to-hour and then day-to-day, but having confidence in the data meant that we could supply the care needed in the face of that demand as best as we could.

“This helped ensure that we had the space and staff available for COVID-19 patients. As a result of this and other measures, we minimised delays in transferring COVID-19 patients to where they needed to be.”

Professor Joe West, epidemiologist at the University of Nottingham, who led the project, said: “The COVID-19 pandemic created a sense of urgency that enabled things previously regarded as impossible to become a reality. 

“This close collaboration between colleagues at the University of Nottingham and NUH, including colleagues working in radiology, the NUH analyst team and our experts in public health, data science and epidemiology, made the project possible.

“There were several challenges to overcome, but this is health informatics success story that has been a real benefit to patients and their families in Nottingham and beyond. We’re proud we made it happen here in Nottingham and made a significant contribution to the national effort against COVID-19.”

The project combined carefully anonymised patient data from a number of previously separate data sources in NUH. It was combined to provide a “big picture” including patients’ health status and their movement through the various wards and departments in the hospital – including Critical Care - while receiving treatment.

This data provided a real time flow of information during the early Pandemic phases that allowed hospital managers to manage bed capacity and forthcoming demand for NHS acute services in Nottingham.

Whilst the project proved the power of data and how this could help inform real-time patient care during the early phases of Pandemic, the project team believes that their work has wider application in the NHS, helping to model and manage patient care in other areas, such as elective surgery.