Cookies on this website
We use cookies to ensure that we give you the best experience on our website. If you click 'Continue' we'll assume that you are happy to receive all cookies and you won't see this message again. Click 'Find out more' for information on how to change your cookie settings.

Researchers and clinicians from Oxford and Portsmouth have been awarded £1.8m by the Health Innovation Challenge Fund to develop a new electronic system to recognise deteriorating patients in hospitals.

The amount of data gathered about patients during even routine screening is increasing rapidly as new technologies become available. Greater computing power enables more data to be captured and stored quickly and inexpensively.

The Health Innovation Challenge Fund is a parallel funding partnership between the Wellcome Trust and the Department of Health to stimulate the creation of innovative healthcare products, technologies and interventions and to facilitate their development for the benefit of patients in the NHS and beyond. It has attracted proposals that will harness the potential of patient data and make it clinically accessible and useful.

At the moment, hospital patient information is reviewed on average every four to six hours and is limited to vital signs such as heart rate and temperature. This new system will include other hospital information such as blood test results and will update every time new information is received. Using this system will make it quicker and easier to identify patients who might need specialist treatment than is currently possible.

- Duncan Young, Professor of Intensive Care Medicine

The current process relies on nurses recognising ‘out of range’ values for the vital signs (including blood pressure, heart rate, and temperature) which are measured regularly on the wards. The new approach will allow other values, such as blood test results or previous medical history, to be included in the assessment. This will create a better overall ‘picture’ of the patient.

Previous work has shown that specific ‘at risk’ patients, such as those who have had recent abdominal surgery, can be identified up to 24 hours earlier using this method. The new challenge is to develop a system which will work across all patients admitted to hospital.

The research team includes doctors, nurses, researchers, and computer scientists who will work together to combine patient information recorded during a hospital stay to develop computer algorithms that describe an evidence based model of normality. From these algorithms it will be possible to identify how far from ‘normal’ any individual patient is at any one time, and those patients most at risk of deterioration can be flagged for more urgent review.

This will be an important decision-making tool which will help hospital staff effectively plan their workload, and provide hospital managers with the information to direct resources to areas of most need.

Similar stories

New machine learning system developed to identify deteriorating patients in hospital

Researchers have developed a machine learning algorithm that could improve clinicians’ ability to identify hospitalised patients who need intensive care.

Accidental awareness in obstetric surgery under general anaesthesia more frequent than expected

The largest ever study of awareness during obstetric general anaesthesia shows around 1 in 250 women may be affected, and some may experience long-term psychological harm.

Low-cost ventilator wins at E&T Innovation Awards

The OxVent is a rapidly deployable and scalable low-cost mechanical ventilator specially designed for COVID-19, which has now been recognised as one of the best innovations of the year by the Institute of Engineering and Technology.

Preventing opiod-related addiction harm after surgery

An international group of experts have published guidance on how to prevent opioid-related harm in adult surgical patients.

New book on global anaesthesia

Hilary Edgcombe, with two colleagues in other centres, has co-edited a new Oxford Handbook addressing safe anaesthesia in resource-limited settings