Wei Gao and colleagues describe how location of death has changed for patients with cancer in England between 1993 and 2010.
Most patients with cancer prefer to die at home or in a hospice, but hospitals remain the most common place of death (PoD).This study aims to explore the changing time trends of PoD and the associated factors, which are essential for end-of-life care improvement.
Methods and Findings
The study analysed all cancer deaths in England collected by the Office for National Statistics during 1993–2010 (n = 2,281,223). Time trends of age- and gender-standardised proportion of deaths in individual PoDs were evaluated using weighted piecewise linear regression. Variables associated with PoD (home or hospice versus hospital) were determined using proportion ratio (PR) derived from the log-binomial regression, adjusting for clustering effects. Hospital remained the most common PoD throughout the study period (48.0%; 95% CI 47.9%–48.0%), followed by home (24.5%; 95% CI 24.4%–24.5%), and hospice (16.4%; 95% CI 16.3%–16.4%). Home and hospice deaths increased since 2005 (0.87%; 95% CI 0.74%–0.99%/year, 0.24%; 95% CI 0.17%–0.32%/year, respectively, p<0.001), while hospital deaths declined (−1.20%; 95% CI −1.41 to −0.99/year, p<0.001). Patients who died from haematological cancer (PRs 0.46–0.52), who were single, widowed, or divorced (PRs 0.75–0.88), and aged over 75 (PRs 0.81–0.84 for 75–84; 0.66–0.72 for 85+) were less likely to die in home or hospice (p<0.001; reference groups: colorectal cancer, married, age 25–54). There was little improvement in patients with lung cancer of dying in home or hospice (PRs 0.87–0.88). Marital status became the second most important factor associated with PoD, after cancer type. Patients from less deprived areas (higher quintile of the deprivation index) were more likely to die at home or in a hospice than those from more deprived areas (lower quintile of the deprivation index; PRs 1.02–1.12). The analysis is limited by a lack of data on individual patients' preferences for PoD or a clinical indication of the most appropriate PoD.
More efforts are needed to reduce hospital deaths. Health care facilities should be improved and enhanced to support the increased home and hospice deaths. People who are single, widowed, or divorced should be a focus for end-of-life care improvement, along with known at risk groups such as haematological cancer, lung cancer, older age, and deprivation.
Please see later in the article for the Editors' Summary
Death is the only certain event in our lives. Consequently, end-of-life care is an issue that is relevant to everyone, and everyone hopes for a “good death” (a death that is free from avoidable distress and suffering) for themselves and for their loved ones. Many factors contribute to a good death, including the place of death. When asked, most people say they would rather die at home or in a hospice (a care facility that gives supportive care to people in the final phase of a terminal illness and that focuses on comfort and quality of life rather than on cure) than in a hospital. Importantly, patients who die at home or in a hospice often have a better quality of life than those who die in hospital, and caring for terminally ill patients in the community is less expensive than caring for them in hospital.
Why Was This Study Done?
Many countries have introduced end-of-life care policies that are designed to enable more people to die at home or in hospices. England, for example, implemented its National End of Life Care Programme in 2004. However, to improve end-of-life care services and to enable more people to die in their preferred place, we need to understand how the patterns of place of death and the factors that affect the place of death are changing over time. In this population-based study, the researchers examine the changing pattern of place of death of people with cancer and the factors associated with place of death in England between 1993 and 2010. Cancer is a leading cause of death globally and is responsible for 8 million deaths annually. Deaths from cancer still occur most commonly in hospitals, which are the least preferred place of death for people with cancer; home and hospices are the first and second preferred places of death, respectively, for such people.
What Did the Researchers Do and Find?
The researchers used death registration data collected by the Office of National Statistics to identify all the adult cancer deaths in England between 1993 and 2010 (2.28 million deaths) and to determine where these deaths occurred, time trends in place of death, and the factors associated with place of death. Hospital was the commonest place of death throughout the study period—48% of cancer deaths occurred in hospital, 24.5% at home, and 16.4% in hospices. The proportion of home deaths increased after 2005 whereas the proportion of hospital deaths declined. The proportion of deaths in hospices also increased over the study period. The most important factor associated with place of death was cancer site. For example, patients who died from a blood (hematological) cancer were more likely to die in hospital than patients with colorectal cancer throughout the study period although the proportion dying at home or in a hospice increased over time. Being single, widowed, or divorced was associated with a higher likelihood of dying in hospital than being married. Being over 75 was associated with a higher likelihood of dying in hospital than being 25–54 although elderly people were more likely to die at home or in a hospice after 2006 than in earlier periods.
What Do These Findings Mean?
These findings show that the hospital is still the commonest place of death for patients with cancer in England. However, the increase in home and hospice deaths since 2005 suggests that the National End of Life Care Programme has enabled more people to die at their preferred place of death. These findings identify cancer site, marital status, and age as the three most important factors associated with place of death for patients with cancer. Because the study is a large-scale, population-based study, these findings are likely to be generalizable to other high-income settings. However, because the study did not include data on individual patient preferences for place of death, these findings should be applied with care to individual patients. Importantly, these findings indicate that more needs to be done to support people with cancer (and other terminal illnesses) who wish to die at home or in a hospice. Moreover, they identify groups of people—single, widowed or divorced individuals, older people, and people with specific types of cancer—who need extra help to ensure that they are able to choose where they die.
Please access these Web sites via the online version of this summary at http://dx.doi.org/10.1371/journal.pmed.1001410.
The Cicely Saunders International, a not-for-profit organization, promotes research to improve the care and treatment of all patients with terminal illnesses at home, in hospices and in hospital; its website includes information on end-of-life care and on Cicely Saunders, the founder of the hospice movement in England
This study is part of GUIDE_Care, a project initiated by the Cicely Saunders Institute to investigate patterns in place of death and the factors that affect these patterns
Information on the National End of Life Care Programme is available
The UK National Health Service Choices website provides information (including videos of personal experiences) on end-of-life issues for carers, information on end-of-life care for patients with cancer, and an end-of-life care guide for people approaching the end of their life
The US National Cancer Institute has a fact sheet on end-of-life care for people who have cancer and provides information on hospice care and home care for patients with cancer (in English and Spanish)
The not-for-profit organization HealthTalkOnline provides personal stories about living with dying
The NHS National End of Life Intelligence Network (NEoLCIN) provides information on broad issues about end-of-life care
The South West Public Health Observatory (SWPHO) aims to improve the health of the population through producing evidence to inform decision making on health issues at local, regional, and national levels. SWPHO also produces specific end-of-life care resources (e.g., specialised reports, end-of-life care profiles) and disseminates it via the NEoLCIN website