Evolution and impact of COVID-19 outbreaks in care homes: population analysis in 189 care homes in one geographic region

Background

Internationally, institutional care settings for older adults have experienced high rates of COVID-19 infection and mortality among residents and staff.¹ In the UK, such settings are usually called care-homes, providing 24-hour nursing and/or residential care for those whose complex needs cannot be accommodated at home or in other settings.² In the UK and other countries including the US, care-home funding is often a mix of self-pay and state funding, and the care sector has been under increasing financial and capacity strain because of population ageing and constraints on public funding.³ Robust national data on the care-home population are lacking and data sources are fragmented, meaning our understanding of the needs and outcomes of residents is poor.^4,5

Early epidemiological data identified very high mortality from COVID-19 in some care-home settings, for example affecting 33% of residents in a single US facility.⁶. Atypical presentation of COVID-19 is prevalent, with delirium, postural instability and diarrhoea common presenting symptoms without fever or cough.^7,8 The role of pre-symptomatic and asymptomatic transmission has become clearer over time, with one US study finding that 56% of positive residents asymptomatic when tested and the majority of asymptomatic residents then developing symptoms in the subsequent four days.⁹ In four London care-homes with large outbreaks where all residents and staff were tested twice one week apart, 40% of residents tested positive, of whom 43% were asymptomatic at the time of testing, and 18% had atypical symptoms. Four percent of staff were positive, all of whom were asymptomatic, and viral genome sequencing found evidence of multiple introductions of infection.¹⁰

Deaths in care-homes from COVID-19 account for 47% of all COVID-19 mortality in Scotland,¹¹ but residents dying in hospital have been separately reported to date. National data are aggregated which means our understanding of variation between care-homes is limited. The aim of this study is to describe the evolution of outbreaks of COVID-19 in all care-homes in one large health board in Scotland, including timing of outbreaks, number of confirmed cases in residents, care-home characteristics associated with the presence of an outbreak, and deaths of residents in both care-homes and hospitals.

Methods

Evolution of the epidemic at resident and care-home level

The first resident test was carried out on 10^th March 2020, and the first positive test was in week 12 (beginning 16/03/20) (Figure 1, panel A). There were a total of 55 outbreaks in the five subsequent weeks to 19/04/20. Over the next six weeks to 31/05/20, there a further 15 more sporadic outbreaks, with no outbreaks in the final two weeks of observation.

• Download figure
• Open in new tab

Figure 1:

Evolution of outbreaks over time

The number of residents tested per day rose rapidly, peaking at 50-60 in late April (when testing policy changed from ‘first few symptomatic cases in each care-home’ to ‘all symptomatic cases’) to late May, then fell, with a later rise when testing policy changed to ‘test all residents in care-homes with ongoing outbreaks’. The seven day moving average of numbers of residents with a confirmed positive case rose rapidly, peaking at 28 per day on 23^rd April 2020 (Figure 1, panel C). Daily averages have fallen continually since then. Mirroring this, the Estimated Dissemination Rate was 7.9 on 03/04/20 falling to 0.8 on 16/04/20. There was a second peak of 2.8 on 23/04/20 and a steady decline to below one from 28/04/20 with values below one persistently since then (Supplementary Figure 1).

Fifty-one (72.9%, 95%CI 60.7 to 82.5) care-homes with an outbreak had one or more residents with negative tests before their first confirmed case (Table 1, Figure 2). The number of positive cases per care-home ranged from 1 to 65 (median 7, IQR 2 to 17). Twelve (17.1%) care-homes only had a single case, and 14 (20.3%) had 2-4 cases. A quarter (26.6%) of all cases were in just five care-homes, and almost half (47.8%) were in 12 care-homes.

View this table:

• View inline
• View popup
• Download powerpoint

Table 1:

Outbreaks in 70 care-homes

• Download figure
• Open in new tab

Figure 2:

Patterns of outbreak for each care-home with an outbreak

The detected incidence (positive cases per bed) in care-homes ranged from 1.0% to 66.7% (median 20.7%, IQR 6.4% to 34.1%) but there was no relationship between care-home number of beds and detected incidence (Figure 1, panel C; Pearson-R 0.02, p=0.90). There was considerable heterogeneity in patterns of diagnosed infection, with 13 (18.6%) care-homes only having cases diagnosed on a single day, and a further 10 (14.3%) only having cases over 2-14 days (Figure 2). Although there was evidence of sustained infection over several weeks in many care homes, in 18 (25.7%) there were gaps between diagnosed cases of 14-27 days, and in 5 (7.1%) gaps of ≥28 days, consistent with new introductions of infection (Figure 2, care-homes 3, 21, 31, 35, 53). However, 43 (61.4%) care-homes had not had a new case for at least 28 days consistent with outbreaks ceasing (and a further 18 (25.7%) had not had a new case for 14-27 days).

Deaths of care-home residents in care-homes and in hospital

There were 419 deaths of care-home residents where COVID-19 was recorded on the death certificate, 403 (96.2%) occurring in care-homes and 16 (3.8%) occurring in hospital. The number of COVID-19 deaths rose rapidly to peak at 63-68 per week in weeks 16 to 19 (weeks beginning 13/04 to 04/05).

Almost all COVID-19 related deaths occurring in care-homes were in those with an outbreak (401 deaths, representing 99.5% of COVID-19 related deaths occurring in care-homes). In the 70 care-homes with an outbreak, there were 472 excess deaths in weeks 13-20 (67.6% of all deaths in these care-homes in that time), of which 399 (84.5%) were COVID-19 related. In care-homes with an outbreak, the peak ratio of observed to expected deaths was 5.3 in week 17 (figure 3, panel A).

• Download figure
• Open in new tab

Figure 3:

deaths in care-homes

Source: National Records Scotland (NRS) death registration. Data is by week of registration (which is later than date of death). NRS data is provisional and may be subject to change, particularly later data if registration is delayed

In comparison, in the 119 care-homes without an outbreak, there was no clear excess mortality and only two COVID-19 related deaths (one stated as confirmed and one as suspected on the death certificate, both untested and not known to public health) (figure 3, panel B).

In 2014-2019, 10.2% of deaths of care-home residents were in hospital. Sixteen care-home residents died of COVID-19 in hospital in weeks 13-20 (3.8% of all care-home resident COVID deaths), with 14 non-COVID deaths, compared to an average in the same period of 39.2 over the previous five years.

Discussion

Just over one-third of care-homes in the health board had experienced a COVID-19 outbreak, but with wide variation in the size, duration and pattern of outbreaks. Almost all outbreaks were in care-homes for older people but the only care-home characteristics statistically significantly associated with the presence of an outbreak was number of beds. Many care-homes experienced single cases or short outbreaks, but sustained or repeated outbreaks were common. In the period examined, there were 852 confirmed cases and 399 COVID-19 related deaths recorded in 70 care-homes with 3451 beds, plus an estimated additional 73 (15.5%) excess deaths not recorded as COVID-19 related. A quarter of confirmed cases and deaths were in just five care-homes, and approximately half in twelve care-homes. In the remaining 119 care-homes, there were no outbreaks, no excess deaths, and only two COVID-19 related deaths reported neither of whom were tested.

There were only 16 (3.8%) COVID-19 related deaths of care-home residents in hospital, and approximately one-third the expected non-COVID-19 deaths in hospitals during the epidemic, consistent with some of the observed non-COVID-19 excess deaths in care homes being deaths that would otherwise have occurred in hospital (although numbers are small).

Strengths include the inclusion of all care-homes in a geographical area using linked data, and examination of resident deaths in care-homes and in hospitals. Availability of resident testing was higher and earlier than in many other areas because the regional virology laboratory had capacity, and a public health outreach team supported rapid testing as part of an active outbreak response programme. Limitations include that testing policy and therefore case ascertainment changed over time. Before 17/03/20, testing typically ceased after two to five cases were positive (except where outbreaks were large when additional testing was sometimes done). Subsequently all suspected cases were tested, with asymptomatic resident testing implemented from the end of May in care-homes with ongoing outbreaks. Case numbers are likely to be underestimated, particularly early in the epidemic.¹⁰ However, the absence of excess mortality in care-homes without an outbreak makes it unlikely that any large outbreaks were missed. Finally, the study examines 189 care-homes and so is relatively underpowered to examine associations with care-home characteristics.

Data from the Care Quality Commission shows that 36% of English care-homes report an outbreak compared to 37% in this study.¹⁶ A larger proportion of all COVID-19 deaths are reported to have occurred in Scottish compared to English care-homes (47% vs 28%),^11,17 although both figures are in the range reported internationally (from 24% in Hungary to 82% in Canada).¹ This may reflect differences in admission practices in England and Scotland, since approximately one-eighth of COVID-19 related deaths of care home residents in England are in hospital¹⁸ versus only 3.8% in this study. Data from a London point-prevalence study of four nursing homes for ∼400 residents identified a 26% mortality rate across the homes, higher than that observed here, likely reflecting selection of care-homes with large outbreaks in London.¹⁰ US data has also shown significant associations between the presence of outbreaks and larger care-home size, and lack of association with regulatory quality ratings.¹⁹

The impact of COVID-19 on care-homes has been very large, but this study highlights that the direct and indirect impact on mortality is concentrated in homes with an outbreak. In this study, almost all excess deaths were in the one-third of care-homes with outbreaks, with further concentration in this group since approximately half of confirmed cases and deaths were in just 7% of care-homes. Equally, the observed pattern of outbreaks highlights that any increase in community transmission has the potential to drive further outbreaks, because the majority of older people’s care homes have not had an outbreak, or rapidly controlled their outbreak, meaning that large numbers of residents likely remain susceptible. Predictive modelling of risk must be specific to the care-home population, rather than based on community data.²⁰

Approximately one in six excess deaths were not attributed to COVID-19 on the death certificate, but these deaths were almost exclusively in care-homes with an outbreak, and there was no evidence of significant numbers of COVID-19 or other deaths in hospital. This suggests that excess deaths in care-home residents in this health board are not related to wider changes to the system of care (e.g. inappropriate withdrawal of primary care or inappropriate reductions in hospital admission affecting all care-home residents). Whether excess deaths not attributed to COVID-19 in care-homes with an outbreak are a direct effect (undiagnosed COVID-19 in residents presenting with atypical symptoms) or an indirect effect (e.g. related to changes in care for uninfected residents when staff are overwhelmed and/or short-staffed) needs investigation because at least some may be preventable.

The analysis highlights that univariate associations with care-home characteristics will often be misleading because care-home type (the population served), ownership, regulatory quality ratings and a history of multiple previous outbreaks of other infectious diseases are all strongly associated with care-home size, which (unsurprisingly) dominates associations with the presence of a COVID-19 outbreak.¹⁹ Although care-home size cannot be altered without losing places for existing residents, there may be potential to create discrete units within care-homes where smaller numbers of staff and residents are effectively cohorted to create self-contained units. Such efforts will be complicated by individual care-home built environment, and will be difficult to sustain without rapid outside support during any large COVID-19 outbreak when staff illness and absence may risk compromising safe care. Additional measures to respond to new outbreaks of COVID-19 will also be required, including maintaining high provision of adequate Personal Protective Equipment (PPE), better support for infection control, ensuring self-isolation and active surveillance of residents and staff to ensure early detection of outbreaks and ongoing transmission, and staffing support for care-homes with many staff absent (box 2). However, COVID-19 has high infectivity, which is reflected in high rates of nosocomial infection of other patients and staff in hospital settings despite better established PPE access and infection control training than social care settings.²¹ Infection control is therefore intrinsically difficult in care-home settings where a priority is to maintain social and cognitive function through interaction. Shielding residents and care-homes from further outbreaks will be essential but poses difficult problems, for example in relation to the impact of preventing or minimising family and friends visiting which has its own impact on quality of life.²²

There are a number of unanswered questions. First, SARS-CoV-2 transmission dynamics in this context are not well understood, including the extent to which outbreaks are sustained by ongoing transmission within care-homes and/or by introduction of new infections, and the relative importance of staff introducing infection vs newly admitted residents (or visitors if care-homes open up to visiting). More systematic testing of both residents and staff with whole genome sequencing to trace transmission chains will be of great value, and likely inform public health response (both are underway in this health board).

Second, the care-home characteristics examined in this study are high-level. Research is needed to understand the relative importance of the built environment, the intensity and nature of staff-resident and resident-resident interactions (which will vary considerably depending on the client group), variations in use of agency staff and investment in staff training, access to and effective use of Personal Protective Equipment, and infection control procedures. Some of this data can be collected by survey, but some may require direct observation of care which is challenging during an epidemic. Third, research is needed to understand excess deaths, including exploring variation in the attribution of death to be COVID-19 related, and the circumstances and likely causes of other deaths.

Finally, COVID-19 has highlighted that care-home residents are relatively invisible in routine data, because we cannot reliably identify care-home residents in routine data.⁴ This has seriously hampered our understanding of the epidemic and its impact, but reflects longer-term neglect of the care sector despite its key role in looking after many of our most vulnerable citizens. In the medium term, we need care-home residence to be accurately recorded to support systematic understanding of the needs of a highly vulnerable population. Similar issues apply to people receiving social care in their own homes, who are also vulnerable and largely invisible in routine data.

Conclusion

The impact of COVID-19 is concentrated in a small number of care-homes with repeated or sustained outbreaks where the impact has been very large in terms of the numbers of residents, family and staff affected, either directly or by the loss of loved ones. Many care-homes for older people and virtually all other care homes do not yet appear to have had an outbreak. There is therefore considerable risk of further outbreaks with large number of deaths in care-homes if community COVID-19 incidence increases again. Allowing families and friends to visit residents again is important for quality of life, but needs to be balanced against the need to shield residents of care homes in areas where community incidence is high or increasing. Early detection of outbreaks through regular testing, reliable PPE supply, support for infection control, and measures to ensure safe staffing are all likely to be needed to contain the size of established outbreaks.

Role of funding source

The analysis was not externally funded.

Patient and public involvement

There was no patient or public involvement in the analysis.

Data sharing

Data may be available on request to the corresponding author, subject to NHS Scotland disclosure controls to prevent identification of individuals.

Ethics

Analysis was of public health data at care-home level (with no individual or identifiable patient data) in order to understand the evolution of COVID-19 outbreaks in one health board. Separate research ethics review was not required as this work was undertaken under generic approval for the Lothian Research Safe Haven and the NHS Lothian/University of Edinburgh Dataloch partnership agreement. Care Inspectorate data is publicly available at https://www.careinspectorate.com/index.php/statistics-and-analysis/data-and-analysis and is public sector information licensed under the Open Government Licence v3.0.

Contributorship

The study was conceived by BG, CE, FG, NH, JES, LW and JB. GB, NH, SS, and MT carried out data collection and management, BG, NH, SS, MT, JB and BG carried out data analysis. JB and BG led the drafting of the paper, and all authors contributed to drafting and revision, and approved the submitted paper.

Competing interests

All authors have completed the ICMJE uniform disclosure form and declare: no support from any organisation for the submitted work; no financial relationships with any organisations that might have an interest in the submitted work in the previous three years, and no other relationships or activities that could appear to have influenced the submitted work.