International data on deaths attributed to COVID-19 among people living in care homes

Adelina Comas-Herrera1, Disha Patel1, Greg Arling2, Joel Mossong3, Andrea Schmidt4 (1Care Policy and Evaluation Centre, London School of Economics, 2Purdue University, School of Nursing, College of Health and Human Sciences, 3Health Inspection, Health Directorate, Luxembourg, 4Austrian National Public Health Institute (GÖG))

In order to facilitate updating of the previous reports on mortality linked to COVID-19 among people who lived in care homes, instead of compiling the data using a pdf, we are now sharing the international data using Tableau Public and the national data and sources in question 2.02. of the International Living Report on COVID-19 and Long-Term Care. This post aims to provide an overview of the international situation and will be updated regularly. If you can help us add data for more countries, please email a.comas@lse.ac.uk

Key findings:

  • There seem to be fewer countries regularly reporting data on COVID-19 related deaths among care home residents publicly than earlier in the pandemic, but we have been able to find data for 21 countries, including some we had not found in previous reports such as Luxembourg (we had found data for 22 countries in previous reports).
  • The share of all COVID-19 related deaths who were care home residents has decreased in most countries since the middle of 2021, reflecting that in many countries people living in care homes were given priority for COVID-19 vaccination and, later on, booster doses, and efforts to improve infection prevention and control in care homes.
  • The strong correlation between the number of COVID-19 deaths among people living in the community and among care home residents remains, highlighting the importance of community transmission despite all the measures to protect care homes.
  • Comparing the cumulative number of deaths linked to COVID-19 among care home residents and the numbers of people who were living in care homes close to the beginning of the pandemic gives an idea of the scale of the mortality impact of COVID-19 among the care home population. So far the ratio between cumulative deaths of care home residents linked to COVID-19 and the number of people living in care home is over 1 in 10 in Belgium, Slovenia, England, Scotland and the United States.
  • So far the data gathered in our reports, for the few countries we can cover, show that at least 429,265 people who lived in care homes died linked to COVID-19 so far. This is equivalent to the entire population of cities like San Juan (Puerto Rico), Bratislava (Slovakia), Auckland (New Zealand) and Oakland (United States).

Overview:

People who use long-term care services and rely on care and support from others have been at increased risk of the severe impacts of COVID-19, particularly those living in group settings such as care homes.

Due to the alarm caused by the large number of deaths among care home residents some countries started reporting data on the numbers of COVID-19 infections and associated deaths among care home residents.

We have been working with members of the LTCcovid initiative and the International Long-Term Care Policy Network (ILPN) to collect and collate these data to provide an overview of mortality amongst residents of long-term care institutions across different countries. In this post, we present a compilation of official data for countries for which we have been able to collect updated mortality numbers within the last quarter of 2021 or until April 2022, which, in many countries, captures the mortality impact of the Omicron wave on residents of long-term care facilities. We found data for 21 countries this time. The previous LTCcovid report on COVID-19 related deaths of care home residents had data for 22 countries.

Very few countries have published data on COVID-19-related deaths among people who use community-based care (see Australia and Sweden entries in question 2.02. of the International Living Report on COVID-19 and Long-Term Care).

Caution when comparing across countries

There are a number of difficulties measuring and comparing the numbers of deaths linked to COVID-19 among care home residents across countries, as there are important differences in way in which the data is collected. Key difficulties in interpreting and comparing data from different countries include:

  • Different definitions of “care homes”.
  • Inclusion or not of younger people living in care homes.
  • Including only deaths of people who had been tested vs also including deaths with suspected COVID (particularly in the first wave when testing was constrained).
  • Including all deaths of care home residents irrespective of where the death occurred vs including only deaths that took place in the care home.

The LTCcovid International report on mortality associated with COVID-19 in care homes provides a detailed discussion of the difficulties involved in interpreting and comparing this data. In section 2.02 of the International Living Report we provide the sources and definitions used in each country.

Definition of care homes

The definition of care homes used varies between countries, in some both nursing homes and other forms of residential care, whereas in other countries only nursing homes are included. This has sometimes also changed over time, for example in the United States the data currently available only covers certified nursing facilities, whereas for the previous reports, there was data available for other residential facilities.

Counting deaths confirmed by testing vs suspected due to compatible symptoms

A lack of testing capabilities, differences in the presentation of the disease amongst this population and the exclusion of deaths indirectly linked COVID-19, means it is likely that those countries only reporting deaths confirmed through testing are under estimating the true burden. However, reporting suspected cases may increase the risk of miss-attribution of deaths that could lead to an overestimation of fatalities attributable to COVID-19.

Deaths of people living in care homes vs deaths that took place in care homes

The first graph below aims to illustrate the importance of being aware of the impact of differences in definitions of deaths, showing how important it is to be cautious when comparing countries and to be careful before making assertions about the ability of these different healthcare systems to tackle the pandemic based on this type of data.

Below we plot the data for those countries that record and publish both the number of COVID-19 deaths of care home residents (including those who may have died in hospital) and the number of COVID-19 deaths occurring within care homes.  In this figure, COVID-19 deaths are based on confirmed and suspected cases. As might be expected, the share of deaths accounted for by care home residents is considerably higher than the share of deaths occurring within care homes. The differences between the countries range from 4 percentage points in Ireland to 14 percentage points in Luxembourg. These differences may reflect differences in hospital capacity across the different waves of the pandemic, differences in policy regarding the care of care home residents infected with COVID-19, and differences in the extent of medicalisation of care homes. Where countries only provide data of deaths in care homes, these figures are likely to be an underestimate of the deaths of care home residents. In many countries, there are no information systems that allow the identification of care home residents in death certificates or these numbers have stopped being reported.

In the following figures the data from Norway is plotted alongside the other countries for ease of comparison, but we understand that the data represent the number of deaths IN care homes (that is, place of death), rather than the number of deaths OF care home residents. We must acknowledge the likelihood that the data for Norway may underestimate the number of care home residents who died as some may have been transferred to hospitals and that may have been registered as the place of death.

The scale of COVID-19 related mortality among care home residents

A useful way to illustrate the scale of COVID-19 related deaths among care residents is to compare the total number of deaths so far with the numbers of people who were living in care homes at the beginning of the pandemic. This relationship can be expressed as the cumulative number of COVID-related resident deaths as a ratio to the number of residents in care homes at the beginning of the pandemic. It is important to note that, while giving an indication of the scale of COVID-19 related mortality in care home, comparing the cumulative number of deaths over the pandemic to the numbers of people living in care homes at the beginning of the pandemic is not equivalent to the incidence of COVID mortality among the care home population.

We can see there are substantial differences between countries. In Scotland, on the 24th of April 2022, the cumulative number of COVID-linked deaths among nursing home residents so far was equivalent to a ratio of 13.16 to 100 residents living in nursing homes at the beginning of the pandemic, compared with 0.12 to 100 in South Korea by the 23rd October 2021. These may represent differences in the incidence of COVID-19 among the care home resident population, differences in the management of COVID-19 in care homes, differences due to the period covered by the data, or other factors. However, as highlighted above caution should be taken in attempting to compare performance across countries.

In this graph we are comparing the cumulative numbers of deaths linked to COVID-19 who were care home residents to the number of residents living in care homes as close to the time when the pandemic started as possible (for most countries end of 2019), as the occupancy rate is known to have decreased during the pandemic and could potentially exaggerate the cumulative estimates. We also tried to make sure to match the definition of care home used in the mortality data with that used for the estimated number of residents, but we cannot be sure that this is accurate in all countries. It must be noted that we utilised the number of care home residents where we had access to that figure, and in the case that we did not, we used an estimate for the number of beds instead, therefore these data need to be treated with caution and are not suitable for detailed comparisons between countries.

The strong link between COVID-19 infections in the community and deaths among care home residents

The next graph compares how heavy the burden of mortality has been amongst residents of care homes was, as compared with mortality among the population not living in care homes. We use deaths among people not living in a care home (i.e. in the community) to approximate extent of transmission in the general population, as differences in testing capacity between countries and across time mean that comparing infection rates is very problematic.

This chart demonstrates a high correlation between the number of COVID-19 deaths among people living in the community, our proxy for community transmission, and COVID-19 deaths of care home residents. We can see that in countries where deaths of people living in the community were low, deaths of care home residents were also low. This highlights that, although the steps taken to minimise the risk of COVID-19 infections in care homes, the key driver of COVID-19 mortality in care homes is high transmission of the virus within the surrounding community.

Differences over time in the share of all COVID-19 attributed deaths who were people living in care homes

This figure shows the share of all COVID-19 deaths who were care home residents in the period from the beginning of the pandemic to the latest update as of April 25, 2022. We are distinguishing 4 different time periods. The first covers from the beginning of the pandemic to July 2020, which we are calling the first wave (we acknowledge that different countries experienced slightly different timings). The second covers from July 2020 to January 2021, we are considering this the second wave and is up to the time when at least a few countries started to give the first dose of COVID-19 vaccinations to care home residents. We then identified another time period, from January 2021 to June 2021 as the “ongoing vaccination” period. The final period is since July 2021 and up to the most recent date available (up to 25th April 2022). In this graph we only include countries for which we have updated data from the same source across all time periods. During the period we have called “ongoing vaccination” there were almost no COVID-19-related deaths among care home residents in Australia, so the column appears blank. It is important to note that there is wide variation in the speed in which countries have vaccinated care home residents and it is not always possible to find data on this (see the information available in the relevant question, (3.11, in the LTCcovid International Living report).

The figure shows that the general trends have been varied. While we can see that the share of COVID-19 deaths of care home residents has decreased over time once the vaccination period started, we can also see that, in the most recent period, the Omicron variant has resulted in increased numbers of deaths linked to COVID-19 in care home residents again in many countries. It is likely we will continue to see variation in the experiences between countries due to the differences in vaccination programmes (and their uptake) and evolving infection, protection and control measures. We can also see that different countries had varied experiences across Wave 1 and Wave 2 but must note that the recording and reporting of deaths amongst care home residents were likely to have improved since Wave 1, where there is a higher risk of underreporting as the pandemic first took hold and systems were not in place. The authors of this report are aware that some of the data sources have been revised and improved between different versions of the report, so this figure can only claim to represent the data as published in our previous and current report and should be treated with caution.

Citation

Our data visualisations rely on work from many different people associated with the LTC Covid Report. When citing this post and any of the data visualisations, please also cite the underlying data sources. All the sources for the data can be found in question 2.02 in the LTCcovid International Living Report. This can be cited as:

Comas-Herrera A, Marczak J, Byrd W, Lorenz-Dant K, Patel D, Pharoah D (eds.) and LTCcovid contributors. LTCcovid International living report on COVID-19 and Long-Term Care. LTCcovid, Care Policy & Evaluation Centre, London School of Economics and Political Science. https://doi.org/10.21953/lse.mlre15e0u6s6

We are grateful to the previous co-authors of the COVID-19 related Mortality in care homes reports: Joseba Zalakaín, Elizabeth Lemmon, David Henderson, Charles Litwin, Amy T. Hsu, Florien Kruse and Jose-Luis Fernández. We are also grateful to Sara Charlesworth, Hongsoo Kim, Esther Perez de Vargas Bonilla, Frode Fadnes Jacobsen, Pete Kinross, Klara Lorenz-Dant, Tine Rostgaard, Marta Szebehely and Bruce Guthrie for helping us identify and interpret the data. Any errors or omissions are the responsibility of the authors of the post.

We are also grateful to all the contributors to the LTCcovid International Living report, as the information they are providing on both the structure of the Long-Term Care systems in their countries, the impacts experienced and the measures adopted are extremely helpful to understand the context for these data.

1 thought on “International data on deaths attributed to COVID-19 among people living in care homes”

  1. Pingback: Mortality associated with COVID-19 outbreaks in care homes: international evidence – Resources to support community and institutional Long-Term Care responses to COVID-19

Leave a Reply