Optimising the management of patients with respiratory illness including influenza and COVID-19 in emergency and acute pathways
Reader in Medical Statistics
School of Population Health and Environmental Sciences, King’s College London
Dr Vasa Curcin, Professor Jonathan Edgeworth
King’s College London
Over 300,000 individuals have been exposed to SARS-CoV-2 in the UK, contributing to over 40,000 deaths (1). GSTT has diagnosed over 1400 patients since March 2020, with over 1000 of these diagnosed through emergency attendances or acute admission pathways (preliminary data). Approximately 20% of those admitted to GSTT with confirmed SARS-CoV-2 infection died during their admission.
In addition to COVID-19, other season respiratory infections have been identified as one of the priorities in the NHS long term plan. Nationally, patients with influenza may occupy 4000 beds per day. (2) Every year GSTT diagnoses around 1000 cases of influenza. Last year this included 62 patients spending 893 days on intensive care. Last winter, a moderate season, 58% of admissions lasted <48 hours suggesting improved patient pathways may prevent many admissions altogether.
There is therefore an urgent need for an evidence-based approach to aid clinicians in decision-making in acute and emergency settings when dealing with respiratory infections. Diagnosis of COVID-19 in the emergency department relies on clinical judgement and routinely available tests. Unlike influenza, no point of care test for the diagnosis of SARS-COV-2 infection has yet received favourable appraisal from the NHS technology, nor is any in routine clinical use (ref, ref). This necessitates that clinicians rely on clinical judgement when diagnosing SARS-CoV-2 infection, using routinely available information and tests.
Once a diagnosis has been made, no clinical score has been validated with sufficient sensitivity or specificity to aid in decisions on patient management for COVID-19(3) or influenza (ref). Identifying those patients most at risk of developing severe disease from respiratory illness, such as requirement for oxygen and ventilation, is necessary to inform patient management and flow through the hospital. To this end the Guy’s and St. Thomas’ Charity awarded our lab an £100,000 grant to work on predictive models, severity scoring and redesigning hospital pathways. Risk factors associated with severe disease and death from COVID-19 have been reported by groups from the UK (4) and beyond (5). Other parameters that have been associated with severe disease include laboratory tests such as lymphocyte count, C-reactive protein and ferritin (6, 7). Such associations with severe disease have yet to be validated in our own cohort. Knowledge of risk factors for severe disease will help aid clinicians in decision-making, such as who requires admission or higher levels of care.
There is a need to ensure patient management and flow minimises risk of transmission of respiratory illness between patients. Evidence from the first wave of the pandemic suggest transmission of COVID-19 in hospitals and care homes is a major cause of disease and death (ref). The experience from GSTT suggests there may have been up to 100 cases of hospital acquired COVID-19 (preliminary data), suggesting patient management and pathways could be improved. Recent data from GSTT also suggests the burden of influenza transmission is underappreciated. Our pilot data showed 15% of influenza cases were hospital-acquired, with an in-hospital mortality of 18%, an attributable relative risk of 4.0 compared to those who did not acquire influenza. Using an evidence-based approach to identify those most at risk of both hospital acquisition of COVID-19 and influenza, and subsequent severe disease, is important to ensure the most vulnerable are protected.
There is also an opportunity to trial and include novel biomarkers to provide extra data to inform patient management. Collaborators have identified immune correlates of COVID-19 disease through the COVID-IP study (8), identifying IP-10 as a reliable marker or severity. Furthermore, our department has collected serum samples from over 500 patients, and partnered with the Faculty of Life Sciences at KCL are analysing these to define the immune response to COVID-19 and validate point of care tests.
There is an urgent need to optimise management patient pathways and management in preparation for further cases of COVID-19 and before the next seasonal epidemic of respiratory illness. This proposal is a collaboration with KCL’s School of Population Science, who are actively recruiting data scientists to collaborate in these areas through the DRIVE-Health sponsorships.
Aims and Objectives
We aim to optimise patient pathways, improve patient management and protect those vulnerable from hospital acquisition. This will involve analysis of data gathered from patients managed at GSTT and from application of the available evidence from other centres.
We have identified key questions that may inform these improvements:
1. Can rapid diagnostics, including point of care testing, be deployed to aid diagnosis of influenza and COVID-19?
2. Of those with confirmed influenza or COVID-19 disease, which patients require admission to hospital?
3. Of those admitted with influenza or COVID-19, who is likely to have adverse outcomes?
4. Of those admitted for reasons other than influenza or COVID-19, who is most at risk of acquisition whilst in hospital, and the adverse outcomes associated with hospital acquisition?
To address who requires admission to hospital investigators will consider routinely collected parameters available at 6 hours and 48 hours after admission. These parameters will be used to predict those who are at risk of further deterioration, as measured by later readmission after discharge, requirement for ventilation and in-hospital mortality. A severity score derived from these parameters will be derived, with the goal of aiding clinicans with decisions on discharge.
Parameters to be used to help address these questions include:
· Patient demographic information
· Physiological data including observations taken during the first four hours of attendance to the emergency department
· Routine laboratory parameters tested during the hospital stay
· Radiological information related to COVID-19 diagnosis
· Prescription medicine information