A Personalised Risk Predictor for Covid-19 Outcome: Development and Validation of a Rapid Front Door Triage Tool
Coronavirus disease 2019 (COVID-19) is a global pandemic. It is estimated that, among patients admitted to hospital with COVID-19, approximately 30% may require breathing support. The pandemic has at times nearly overwhelmed the NHS, and exceeded
critical care bed capacity many times over, even with implementation of interventions to reduce transmission of the infection.
Rapid and effective ‘triage’ (assessing who needs what interventions and where to prioritise care) at the point of hospital admission would be beneficial to ensure that patients
at higher risk of deterioration are managed and monitored appropriately. This becomes increasingly important as hospital resources become more stretched.
In this study, we propose to use routinely collected electronic health record data, available as part of DECOVID, to develop and validation a prognostic model to predict risk of (a) COVID-19; and (b) deterioration leading to requirement of ventilatory support or death. The model will be designed to be implemented at the point of hospital admission in order to rapidly 'triage' patients.
DECOVID is a Big Data project that collects electronic health records from the largest NHS
Trusts in the UK. This information has been merged into a very large data pool and we will use this data to perform detailed analyses that will help answer important questions about COVID-19.
This study has been prioritized by a senior clinical and research team (the DECOVID Scientific Steering Committee), who have identified this as being critical to improve patient care and outcomes. It has also been reviewed by a group of patients and public members (The DECOVID Data Trust Committee) who have approved data sharing to answer this important question.
The results will be shared in open access journals, so that as many people as possible can benefit from what we learn. All analytical code (the computer code we use to analyse the data) will also be made available any researchers, to ensure that our research follows “FAIR” principles (Findable, Accessible, Interoperable and Reusable).