Oxford researchers, including Prof Lionel Tarassenko and Prof David Clifton, have developed a machine learning algorithm that significantly outperforms other currently used early-warning score systems to identify hospitalised patients who need intensive care. The algorithm was developed for the HAVEN Project, which aims to produce a hospital-wide IT system that enables a continuous risk assessment in all hospital patients and predicts those at risk of deterioration.
Every year, more than 60,000 patients deteriorate on UK hospital wards to the extent they require admission to an ICU. Late recognition of patient deterioration in hospital is associated with worse outcomes, including higher mortality.
Despite the widespread introduction of early-warning score (EWS) systems and electronic health records, deterioration still goes unrecognised. Most of these systems are designed to give an early warning score based on abnormalities in patients’ vital signs, and clinicians are alerted when the EWS rises above a given threshold. However, they are unable to account for trends over time, patients with chronically abnormal physiology or other indicators of deterioration.
The HAVEN (Hospital-wide Alerting Via Electronic Noticeboard) system combines patients’ vital signs – such as blood pressure, heart rate and temperature – with their blood test results, comorbidities and frailty into a single risk score. The HAVEN score gives a more precise indication of which patients are deteriorating when compared with previously published scores. It detected up to 48 hours in advance nearly twice as many patients who suffered a cardiac arrest or needed intensive care, as those identified by the next best system.
Tarassenko, Clifton and their co-authors recently published the results of their study in which they were able to retrospectively validate HAVEN on 266,295 admissions to four hospitals. The detailed findings have been published in the American Journal of Respiratory and Critical Care Medicine.
The HAVEN system was developed as part of a collaboration between the University of Oxford’s Institute of Biomedical Engineering and the Nuffield Department of Clinical Neurosciences (NDCN), with support from the NIHR Oxford Biomedical Research Centre.
As well as the Oxford BRC, this independent research was funded by the Health Innovation Challenge Fund, a funding partnership between the Department of Health and Wellcome Trust.