Part 2: The vulnerability of care homes to COVID-19 outbreaks


In this section we focus on care homes. There has been debate as to the root causes of the mortality seen among residents during the pandemic. The vulnerability of a care home to an outbreak of the virus will depend on the risk of introduction, transmission and the susceptibility of residents to severe infection. Here we consider what the available data tell us about the likely contribution of each of these.

Unfortunately, available data do not allow us to assess potential COVID-19 infection risks within domiciliary care. Further research and analysis are needed, given the specific problems that the domiciliary care sector experienced in accessing COVID-19 tests and PPE.

The risk of introducing COVID-19 into care homes from hospital or the community

COVID-19 may be introduced into care homes by staff, visitors and residents. Evidence is still emerging as to the relative importance of each of these potential routes. Directors of adult social care services have raised major concerns that hospital discharges may be a source of introducing COVID-19 into care homes. Evidence to date has highlighted that staff may have been an unwitting source of infection, particularly when working as bank or agency staff, or where sick pay was not available.

To assess the scale of any possible introductions from hospitals to care homes due to discharges, we carried out novel analysis of individual patient data from all hospitals in England up to 30 April 2020. Due to poor recording, hospital data alone will underestimate care home residency by as much as 70% (from our estimates). Therefore, an additional algorithm developed by the Health Foundation was used to identify patients who were residents of care homes. This matched the address registered with a patient’s GP to care homes registered with CQC. Full details of the methods, which have been previously validated, are described in the technical appendix.

In February 2020, this captured data from 355,907 people living in care homes in England. Of these, 80.2% were older than 65 years and 44.5% were identified as living in nursing homes, as opposed to residential care homes. Using this linked dataset, we estimated the number of hospital discharges of new and permanent residents to care homes in England.

Hospital admissions and discharges were higher than historical averages for residents of care homes and the general population in January and February 2020. During February, discharges to care homes were 9% higher than the historical 5-year average across 2015–2019 for this period – see Figure 4. There were 46,700 discharges to care homes between 17 March and 30 April, the period after NHS England and NHS Improvement had issued measures stating that medically-fit patients should be discharged and elective procedures limited., However, this is 86% of the average number of discharges usually seen at that time of year (as shown in Figure 4). As can be seen from the pattern of admissions in the same figure, this drop may be due to the smaller number of patients being admitted to hospitals (following the measures already mentioned). Hospital admissions for care home residents fell to around 70% of the historical average across this period. But the pattern varied between care home types. While discharges to residential care homes fell to 75% of their historical average (12,400 discharges), discharges to nursing homes actually increased to 120% of the historical average (17,000 discharges).

Without additional data on testing and the spread of infection among individuals, it is not possible to robustly quantify whether and how much these discharges contributed to the spread of the virus. However, some of these discharges could have represented an infection risk, particularly in March and early April. At this time, testing for COVID-19 was not recommended prior to discharge from hospital. Without testing it is unlikely that asymptomatic or presymptomatic patients would have been identified, and appropriate isolation measures adopted, to prevent onward transmission in the care home. In any case, there is evidence that care homes struggled to access PPE and implement isolation measures, which will have increased the risk of spreading infection from any COVID-19 positive patients discharged to their care. There is evidence that where a care home had at least one detected case, higher staff levels and more severe PPE shortages were most linked to higher case counts. Measures introduced in the Coronavirus (COVID-19): adult social care action plan, to test all residents prior to care home admission, should reduce the risk of infection from patients being discharged from hospital.

Infections can also be introduced into care homes from visitors or care staff, who without testing may unknowingly transmit infection as they move from the community into and across care settings. 7% of care home staff have tested positive for COVID-19 since the start of the pandemic up to 19 June. In previous Health Foundation analysis, we found that in many areas of the country the scale of the outbreaks in care homes mirror that in the general population. International comparisons have also found that the total proportion of care home residents who have died due to COVID-19 closely follows the deaths attributed to COVID-19 per 100 million.

Risk of transmission within care homes

Due to care needs, residents require close personal contact with others on a frequent basis. During the early stages of the pandemic in England, evidence emerged from other countries (such as the USA) about the importance of asymptomatic infection, and that COVID-19 may present atypically in patients who are older or have dementia. In the UK, the initial guidance was to test and isolate only those who were symptomatic (with ‘conventional’ symptoms). As a result, some positive cases in care homes may not have been correctly identified and isolated, increasing the risk of transmission.

It was not until 3 July that the government committed to testing all staff in care homes regularly. National figures on the number of cases among care home workers only began to be reported on 16 July. However, the important role that staff play in transmitting COVID-19 has been observed in several research studies, with both symptomatic and asymptomatic transmission taking place. This includes early findings from a large national survey of care homes (the Vivaldi study), which showed a significant correlation between infection rates among staff and infection rates among care home residents. Regular testing, now recommended by government guidance for all care home staff, allows symptomatic and asymptomatic staff to self-isolate and prevent transmission to residents.

The susceptibility of residents to severe infection

In our analysis of the care home population, we found that in February 2020 permanent care home residents were of advanced age (80.2% were older than 65 years), and unsurprisingly living with chronic conditions. 40% of residential home residents and 48% of nursing home residents were living with dementia; 13% of care home residents and 17% of nursing home residents had diabetes; and 13% of care home residents and 15% of nursing home residents had chronic obstructive pulmonary disease. These figures are likely to underestimate the true prevalence of these conditions, as the information was taken from hospital records for residents, which can incompletely record chronic conditions on admission. If a resident did not have an inpatient admission, it was not possible to determine if they had any of these conditions. Nevertheless, these numbers highlight the fact that residents are admitted to care homes because of advanced age and increased need for additional care and assistance.

This makes residents particularly vulnerable to COVID-19 when infected. The risk of severe illness and death due to COVID-19 has been closely linked to frailty and underlying illness such as dementia, diabetes, chronic respiratory disease and a range of other chronic conditions. Therefore, if there is an outbreak of COVID-19 in a care or nursing home it is likely that residents would be at high risk of severe illness or mortality. Findings from the limited number of studies with detailed information on outbreaks in care homes support this, reinforcing the need to shield and protect residents from infection.

‡‡ NHS England has published data on hospital discharges between 30 January and 16 April, using a slightly different dataset, and comparing 2020 trends to 2019 trends. While the broad findings are in agreement, the absolute numbers of discharges differ. Please see: and for this analysis. NHS England has since revised its methodology and the original methodology is available via GitHub:

§§ This underestimates the true population of care home residents.

¶¶ Note that 17,300 discharges were to care homes of unknown type.

*** From a study of 248 care homes in Norfolk:

††† Our analysis may underestimate the number of hospital admissions towards the end of April. This is because data for patients who are still in the hospital by the end of the reporting period might not be included. Sensitivity analysis that limits the analysis to hospital spells with a duration of less than 45 days can be accessed via GitHub

‡‡‡ Dementia was measured as defined under the frailty index and uncomplicated diabetes and chronic obstructive pulmonary disorder were measured as defined by code lists under the Elixhauser Score. All long-term conditions were measured using diagnosis codes from hospital admissions over the previous 3 years.

Previous Next