Hospital is the most common place of death across the UK, with 43% of deaths occurring in hospital in England and Wales in 2023. Although the proportion of people dying in hospital has declined over time, many people will attend hospital in the months preceding death, even if they die elsewhere. As such, care for people at the end of life accounts for a significant proportion of hospital activity.
Having to go to A&E or being admitted to hospital in an emergency close to the end of life is often not the best outcome for individuals (as described in this recent Guardian article and reader responses), while for hospitals discharging patients to appropriate care can be challenging. There has been a policy goal for some time to reduce emergency admissions at the end of life. Understanding how hospital services are used at the end of life could help to identify how patients’ needs might be better met at home or in community settings.
In this blog, we examine how hospital services are used by people at the end of life – covering from the year before the Covid pandemic until August 2023. As with our earlier blogs about the care of people who died at home and in care homes, we use data covering a population of 24 million people linking GP clinical records, hospital data and death registrations.
Which hospital services are used by people in the last month of life?
In the most recent year of the study, September 2022 to August 2023, an emergency admission occurred in the month prior to death for 80% of people who died in hospital (with the remaining patients being in hospital for more than a month before death, dying following a planned admission, or dying in A&E before being admitted). The proportion of people having an A&E visit in the last month of life was similarly high (81%). A fifth of people who died in other locations also attended A&E or had an emergency admission in the last month of life, highlighting the important role that emergency care has in the care of people at the end of life.
The proportion of people receiving planned care – such as an outpatient clinic appointment or an elective admission (including day cases) for treatment or tests – was more similar for people who died in hospital compared with people who died in other locations.
Proportion of people using hospital services in the last month of life (September 2022 to August 2023)
ChartHow did the use of hospital services for people at the end of life change over the Covid-19 pandemic?
The pandemic caused huge disruption to health services. Hospital services geared up to accept Covid patients, but emergency admissions in fact fell sharply overall during the first Covid wave, and have only recently recovered to pre-pandemic levels. In contrast, emergency admissions of people in the last month of life increased sharply, and were highest during the second Covid wave that peaked in January 2021.
We can approximate the proportion of emergency admissions (and other types of hospital activity) where the patient was in the last month, by taking admissions where the patient died (from our study population) as a percentage of all admissions, also adjusting for the proportion of the England population included in the study (see chart below).
This indicates that the proportion of people at the end of life increased by a factor of three during the first Covid wave – from around 8% before the pandemic to 22% in April 2020. The proportion of A&E visits where the patient was in the last month of life similarly increased and, as with emergency admissions, also increased dramatically during the second Covid wave. The pattern for emergency care since the second Covid wave shows much less variation – although there is a little more fluctuation than before the pandemic.
The proportion of people receiving planned care at the end of life is much lower than for emergency care, with people in the last month of life making up approximately 1% of outpatient appointments and elective admissions. These proportions increased during the first wave, but were then more stable, reflecting efforts to continue as much planned work as possible during the second wave and during recovery.
How does the use of services vary by cause of death, and how has this changed over time?
Use of hospital services in the last month of life varies substantially between causes of death. The number of events per person in the last month of life are shown below, for four common causes of death, which between them account for nearly two-thirds of deaths over the study period.
Cancer: people who died from cancer received substantially more planned care in the last month of life than other patients, reflecting more treatment for people with cancer. At the start of the pandemic, there was an increase in outpatient activity and a fall in elective care, as services moved to provide more care for vulnerable patients without hospital admission, to reduce risk of Covid-19 infection. Outpatient contacts for cancer patients in the last month of life have continued to increase since the pandemic. Elective admissions for cancer patients have largely returned to pre-pandemic levels, while emergency admissions for cancer patients have largely been below pandemic levels over the study period.
Circulatory disease: A&E visits and emergency admissions for people who died from circulatory disease dropped at the start of the pandemic, but have broadly recovered to pre-pandemic levels. Outpatient attendances have increased slightly since the start of the pandemic, but not to the same extent as for cancer patients.
People who died from flu or pneumonia are the group with the highest levels of emergency admissions and A&E visits, reflecting acute illness at the end of life and requiring hospital treatment.
People who died from dementia have the lowest overall use of hospital services, and the most notable decline in A&E visits and emergency admissions. A&E visits fell by 14% between August 2019 and August 2023, while emergency admissions fell by 17%. These decreases are consistent with the shift in hospital use noted for people who died in care homes. People who died from dementia have very low elective admissions in the last month of life.
Implications
This blog has set out the scale of end of life care provision provided in hospital, and in particular the extent to which care in the last month of life is a core activity within emergency hospital care. 43% of people die in hospital, and of those who die elsewhere, almost a fifth will attend A&E or have an emergency admission in the last month of life. In the most recent study year, ending August 2023, 9% of emergency admissions were people in the last month of life, and this proportion increased to over 22% during the first wave of the pandemic. Comparing causes of death, people dying from infections have the most A&E visits and emergency admissions, although these are common experiences across all causes of death.
The scale of provision of end of life care in hospital points to opportunities to improve care, for the benefit of people dying, their families and also those who care for them, as well as the wider health system.
First, the National Audit for Care at the End of Life has highlighted significant opportunities to improve end of life care in hospital. In 2022 (the most recent audit available), only 60% of providers met the current standard to have face-to-face specialist palliative care advice available for eight hours a day, seven days a week. There was a documented discussion about the extent to which the patient wished to be involved in decisions about their care in only a quarter of cases audited –indicating missed opportunities to involve patients in care. There was no documented plan for meeting non-physical needs for over half of cases.
Second, policies that prioritise provision of end of life care in other settings could enable more end of life care needs to be met at home, in care homes or hospices. A better understanding of what leads to people needing to attend A&E, and be admitted in an emergency near the end of life, is likely to be vital in understanding hospital deaths and how to avoid them, where appropriate, and in accordance with the person’s preferences.
The decline in A&E attendances and emergency admissions for people who die in care homes points to the scope for a more structured approach to individual care planning to support end of life care, to enable more people to be cared for outside of hospital. It will be important to evaluate the Enhanced Health in Care Homes programme to fully understand the impact, and identify wider lessons, such as for people close to the end of life living at home.
Third, there are direct lessons from the Covid-19 vaccination programme about scope to reduce hospital deaths through preventive care. Patients dying from infections continue to be those most likely to attend A&E and be admitted to hospital in an emergency at the end of life. The flu vaccination programme reaches only 79% of people over 65, with lower rates in more deprived areas and among minority ethnic groups. More concerted action to improve vaccination coverage would have a direct impact on the need for urgent hospital care at the end of life, particularly for respiratory infections. In doing so, this could also contribute to reducing inequalities in end of life care, by reducing variation in emergency hospital admission at the end of life.
There is only one chance to get end of life care right. This should be a sufficient reason for taking action to improve quality of care. But health system leaders should also consider the wider opportunities to enable local health systems to become more sustainable, and address long-standing aims to develop services out of hospital.
Methodology
The quality of care that people receive at the end of life, as well as inequalities in experiences of care, are critical policy concerns. Please find here more information about our approach to examining changing trends in service use towards the end of life: Describing our methodology
Data notes
Hospital activity for the England population for the study period was obtained from published data and from Hospital Episode Statistics (HES) data as summarised below, separately from the OpenSAFELY data.
The study population includes 42% of deaths in England, so in order to estimate the proportion of hospital activity which occurred among people in the last month of life, all hospital activity was weighted by 0.42. The comparison is approximate, because for all hospital activity the month relates to the date of admission, while for the study population of people at the end of life the month relates to the month during which they died.
HES data sources
Activity type | Source |
A&E visits | |
Outpatient attendances | |
Emergency admissions | |
Elective admissions | Hospital Admitted Patient Care Activity, 2022-23 with data from April 2023 from Hospital Episode Statistics Admitted Patient Care data. Hospital Episode Statistics data (year 2023/24) Copyright © (2024), NHS England. Re-used with the permission of NHS England. All rights reserved. |
Suggested citation
Scobie S, Georghiou T, Julian S, Bagri S, Davies M (2024) “How are hospital services used at the end of life?”, Nuffield Trust blog
Using health and care services at the end of life
Comment series