Nosocomial transmission of SARS-CoV-2

SARS-CoV-2 spreads in many different types of environments including within healthcare settings. Understanding how much transmission happens in hospitals or in the community will aid our ability to control ongoing spread, as well as to improve interventions. By knowing what factors make certain hospitals hotspots of transmission we can limit spread by targeting these factors in particular. Within this project we will use mathematical and statistical techniques to harness the UK wide data on hospital cases to understand how important transission of SARS-CoV-2 in hospitals was to the first "wave" of COVID-19 in the UK and what we could do in the face of any future resurgence.

There is increasing evidence that the novel coronavirus (SARS-CoV-2) can spread within healthcare settings. Quantifying the relative contribution of this environment to wider transmission will aid in intervention design and control of SARS-CoV-2. In this project we aim to combine statistical and mathematical modelling, with UK wide situational reports, an individual patient level database, and Trust level information, to determine the likely contribution of nosocomial transmission to the UK COVID-19 epidemic.

Our project consists of three parts. Firstly, we will determine the contribution of nosocomial transmission by using combinations of samples from data-driven distributions to estimate the number of missed SARS-CoV-2 infections, adjusting total hospitalised COVID-19 case numbers to account for "nosocomial acquired, community onset" cases. We will also quantify the number of COVID-19 cases due to onward transmission from undetected nosocomial transmission. Secondly, we will use multilevel Poisson regression models to determine the important factors that contribute to the observed heterogeneity in nosocomial transmission between Trusts. Thirdly, we will combine the above insights to estimate the impact of interventions: from the number of undetected cases and knowledge of the natural history of COVID-19 we can estimate the likely effect of quarantine for discharged patients or screening on (re-) admission. Such information will provide key evidence to inform the use of additional interventions in the event of future resurgences.