This indicator measures the percentage of deaths due to prostate cancer that occurred in a hospital versus in a private home or at some other location. The findings are based on information contained in the national vital statistics database for 2011 (by province). National results showing the location of deaths due to prostate, breast, lung and colorectal cancer in percentages are also provided.
Findings from several Canadian surveys have shown that, if given a choice, many cancer patients would prefer to die at home or in a hospice rather than in a hospital.1,2 However, a lack of awareness about the availability of appropriate palliative care resources or services at home (e.g., to ensure effective symptom management) may make death at home less achievable.3
Knowing where prostate cancer patients die is important for two reasons: it enables a better understanding of health system resource allocation and it could give us clues on the extent to which health care system use at the end of life aligns with patient preferences. Policy enablers to dying at home may include provincial health plan coverage of pain control medication delivery in the home.
While the results presented here provide a relatively crude measure, they can highlight potential opportunities for end-of-life planning and show how better data collection and analysis can be used to support health system improvements.
What are the results?
There was much variation across provinces in terms of the location of death for men who died from prostate cancer: in Manitoba, the data suggest that 90.0% of deaths occurred in hospitals compared to 33.3% of deaths in Prince Edward Island (Figure 5.5). The percentage of deaths reported as occurring in private homes also varied considerably—from 30.4% of men in Nova Scotia to 0.0% in Saskatchewan and Prince Edward Island. It is important to note that provincial variations exist in how location of death is categorized on death certificates, as well as in how different settings (i.e., designation of hospital-based hospices or palliative care units) are classified. Manitoba has indicated, for example, that many of the in-hospital deaths recorded in the province’s vital statistics data actually occurred in hospital-based hospices or palliative care units as opposed to in acute care hospital beds.
In the data reported for 2011, a slightly lower percentage of men who succumbed to prostate cancer died in hospital compared to patients who died from breast, lung and colorectal cancers (Figure 5.6).
What do the results mean?
Findings that some provinces had a lower proportion of deaths in hospital may reflect the existence of more options for community-based care (e.g., home care, hospice) in those jurisdictions. These provinces may also offer more resources that support home-based end-of-life care. For example, some provinces cover the cost of pain medication delivered in-home while others do not. Certain provinces may also have developed more strategic initiatives for advanced care planning.
Patient and family preferences may also play a role in influencing where patients die. While patient surveys have indicated that the hospital is the least preferred setting for end-of-life care, other factors such as the availability of health services and resources may influence what actually happens. In reality, symptom management resources, emotional support from caregivers or loved ones, and/or financial resources needed to support dying at home are not always available or realistically achievable at home. It is important to be aware that patient and caregiver preferences or needs may also change over time due to clinical, psychological or practical challenges. For some people, the preferred end-of-life setting may eventually be in the hospital or in a hospital-like setting.
Prostate cancer tends to have a longer disease course than other cancers, which suggests that health care teams may have more opportunities to introduce palliative care discussions and arrange hospice care earlier on. In the United States, reported use of hospices by men with prostate cancer varied from 18% to 53%.4,5 A recent study found that the use of these services increased over time during the final weeks or months of life.5
A study of palliative care services in Wales showed that prostate cancer patients had a lower average number of inpatient admissions per year for palliative services compared to patients with breast, lung and colorectal cancers.6 However, once admitted to hospital, prostate cancer patients had the longest length-of-stay. More investigation is needed into the reasons for the initial admission to hospital and why patients remain there.
An analysis by the Canadian Institute for Health Information showed that 53% of cancer patients (all types) who died in acute care hospitals had palliative care documented as the main reason for hospitalization.7 While some of these patients may have been cared for in designated palliative care units, most of them were not.
Canadian Cancer Society’s Steering Committee. Canadian Cancer Statistics 2010. Toronto (ON): Canadian Cancer Society; 2010. 124 p.
Murray MA, Fiset V, Young S, Kryworuchko J. Where the dying live: a systematic review of determinants of place of end-of-life cancer care. Oncol Nurs Forum. 2009 Jan;36(1):69-77.
Canadian Hospice Palliative Care Association. The Pan-Canadian Gold Standard for Palliative Home Care: Toward Equitable Access to High Quality Hospice Palliative and End-of-Life Care at Home. Ottawa (ON): Canadian Hospice Palliative Care Association; 2006. 20 p.
Bergman J, Kwan L, Fink A, Connor SE, Litwin MS. Hospice and emergency room use by disadvantaged men dying of prostate cancer. J Urol. 2009 May;181(5):2084-9.
Bergman J, Saigal CS, Lorenz KA, Hanley J, Miller DC, Gore JL, et al. Hospice use and high-intensity care in men dying of prostate cancer. Arch Intern Med. 2011 Feb 14;171(3):204-10.
Green JS. An investigation into the use of palliative care services by patients with prostate cancer. Am J Hosp Palliat Care. 2002 Jul-Aug;19(4):259-62.
Canadian Institute for Health Information. End-of-Life Hospital Care for Cancer Patients. Ottawa (ON): Canadian Institute for Health Information; 2013. 26 p.
Data for this indicator were collected by the provinces based on information recorded on the official registration of death. Data were then submitted to Statistics Canada to be included in the Vital Statistics Database. This database contains data elements that identify cause of death and location of death. “Location” is grouped into the following categories: hospital, other health care facility (e.g., long-term care or chronic care facility), private home or any other specified locality, or “unknown.” Depending on the province, a hospice can be categorized as “other health care facility” or “other specified locality or unknown.” (For this analysis we grouped the “other” categories together.)
The definition of “hospital” varied across provinces. In Quebec, this category included residential and long-term care centres. In Manitoba, designated palliative care units were included as part of the hospital category in their data collection, while in other provinces this type of bed might have been considered part of long-term care, which puts it in the “other” category in the charts presented here.642 As a result, percentages of hospital deaths for Quebec and Manitoba may appear higher relative to other provinces but do not necessarily indicate any actual differences in the delivery of services.
There is also variation in the way palliative care beds are designated in hospitals across the provinces. The impact of this on reported variations in deaths that occurred in hospital is not known.642 Further investigation is needed to determine the true influence on the results presented here.
Coding on death certificates also varies by province. In Saskatchewan and Prince Edward Island, a very small proportion of deaths are recorded at home, which suggests that most in-home deaths may likely be recorded in the “Other” category (Statistics Canada, personal communication).
This indicator only examines the location of the patient at the time of death and does not take into account time spent in other settings during the weeks leading up to death. As such, it does not fully reflect the patient’s use of health system resources during the end of life.
Place of death
Definition: Percentage of cancer deaths occurring in hospital, private home, or other
By province: Number of prostate cancer deaths in hospital, private home, or other;
Canada: Number of prostate, breast, lung, and colorectal cancer deaths in hospital, private home, or other
Number of prostate cancer deaths in the given province;
Number of prostate, breast, lung, or colorectal cancer deaths in Canada
Data source: Vital Statistics death database
Measurement timeframe: 2011
Stratification variables: Province, disease site (prostate, breast, lung, colorectal)
“Other” includes other specified locality and other health care facilities. Unknown localities are excluded.