Chart of the week: The proportion of Covid-19 infections occurring within hospitals is growing as cases rise

Each week we present analysis of data in chart form to illustrate some key issues and invite discussion. This week, Sarah Scobie looks at the thorny problem of Covid-19 infections occuring when patients are already in hospital for other reasons. How significant a problem is this, and should we be worried?

Data story

Published: 18/11/2020

Daily data on new hospital cases brings together information on people admitted with Covid-19 and cases where Covid-19 was diagnosed in hospital. These numbers, which are reported daily, have been climbing steadily since early September.

During the first wave, it was estimated that 20% of hospital cases were acquired in hospital – defined as cases diagnosed more than a week after admission. A similar proportion of hospital cases were acquired in hospital between 1 September and 14 November: the proportion has grown with the number of cases, from 10% in September to 19% in the first two weeks of November.

An investigation into hospital-acquired Covid-19 during the first wave identified a complex range of factors which contributed to Covid-19 infections in hospital, from the challenges of separating Covid and non-Covid patients; the impact of building design; high levels of bed occupancy; and staffing levels, training and morale. Some of the factors which were relevant in the first wave are less significant now, such as lack of testing and PPE, but hospitals are also busier as they continue to undertake more treatment of non-Covid patients. Other factors, such as ward layout, may harder to address without affecting overall capacity.

Hospital-acquired Covid-19 can be disastrous for patients already unwell with other conditions.  It is also very challenging for staff, who are themselves at risk. And new infections in hospital make the job of restoring services across the NHS even more difficult.

The number of Covid-19 cases in hospital is likely to continue to climb, while cases in the population have not yet stabilised in many areas. Evidence from other countries indicates that it is not inevitable that hospital-acquired infections must also rise. We need to be careful of the complacency that marked the initial response to hospital-acquired infections in the 1990s and early 2000s. Learning lessons from elsewhere will be important to inform how we address hospital-acquired Covid-19 here. 

New hospital patients with Covid-19 18/11/2020

Chart

Note:  


1. The data source is the Daily publication of Covid-19 admissions and bed occupancy.

2. Hospital acquired cases have been calculated by subtracting measure 3 (Estimate of new admissions to hospital from the community) from measure 2 (Estimated new hospital cases) 

3. A rolling 7-day average of both hospital and community acquired cases has been calculated, so, for example, the value for 7 September is the average for 1-7 September.

Source:  

Nuffield Trust analysis of NHS England, Daily publication of Covid-19 admissions and bed occupancy data, downloaded 16th November 2020.

Read more

Comments