TY - JOUR
T1 - PLACE OF CANCER DEATHS IN ENGLAND, 2001-2010: TIME TRENDS AND DETERMINANTS
AU - Gao, W.
AU - Ho, Y.
AU - Verne, J.
AU - Glickman, M.
AU - Higginson, I. J.
PY - 2012/9
Y1 - 2012/9
N2 -
Background Place of death is an important quality indicator for end of life care. Understanding where patients with cancer die and what factors are associated with place of death is important for improving end of life care and has significant health care cost implications. There is, however, limited up-to-date evidence to guide practice.
Methods This is a whole population-based observational study. All deaths with cancer as the underlying cause of death (ICD–10 codes: C00-C97), which occurred in the period 2001–2010, were extracted from death registrations in England. The time trends in place of death (coded as hospital, home, hospice, nursing home and elsewhere) were described and tested by using the Cochran-Armitage Trend Test. Factors associated with hospital death (age, gender, year of death, cancer site, index of multiple deprivation (IMD), region) were evaluated using the generalised linear model. The clustering effect within region was accounted for using an exchangeable working correlation structure.
Results A total of 1,270, 071 cancer deaths were recorded over the 10 years period, of which 47.4% occurred in hospitals, 24.1% in homes, 17.1% in hospices, 6.6% in nursing homes and 4.8% elsewhere. In the study period, deaths in hospital decreased from 49.7% to 42.2% (annual decrease 0.8%; Pz=59.6<0.001), and home deaths increased from 22.3% to 27.6% (annual increase 0.4%; Pz=48.7<0.0001). Independent risk factors for dying in hospitals were: being older (Adjusted OR75–84 vs 18–54 (AOR): 1.25; 95% confidence interval (95%CI): 1.24–1.26) and living in deprived areas (AORs:1.06–1.31). Leukaemia (AORref=lung: 3.66; 95%CI: 3.65–3.68), Non-Hodgkin’s Lymphoma (AORref=lung: 2.89; 95%CI: 2.88–2.90) and Bladder cancer (AORref=1.46; 95%CI: 1.45–1.47) were the top three cancer deaths most likely to occur in hospitals. Men had a slightly higher chance than women of dying in hospitals (AOR: 1.13; 95%CI: 1.13–1.14). Cancer deaths in hospitals were more likely to occur in London than in South West (AORs range: 0.84 in South East Coast to 1.27 in London).
Conclusion We found an overall reducing trend in hospital deaths and an increasing trend in home deaths in the past ten years. However, significant inequality in place of death still exists. Future research needs to explore the underlying reasons.
AB -
Background Place of death is an important quality indicator for end of life care. Understanding where patients with cancer die and what factors are associated with place of death is important for improving end of life care and has significant health care cost implications. There is, however, limited up-to-date evidence to guide practice.
Methods This is a whole population-based observational study. All deaths with cancer as the underlying cause of death (ICD–10 codes: C00-C97), which occurred in the period 2001–2010, were extracted from death registrations in England. The time trends in place of death (coded as hospital, home, hospice, nursing home and elsewhere) were described and tested by using the Cochran-Armitage Trend Test. Factors associated with hospital death (age, gender, year of death, cancer site, index of multiple deprivation (IMD), region) were evaluated using the generalised linear model. The clustering effect within region was accounted for using an exchangeable working correlation structure.
Results A total of 1,270, 071 cancer deaths were recorded over the 10 years period, of which 47.4% occurred in hospitals, 24.1% in homes, 17.1% in hospices, 6.6% in nursing homes and 4.8% elsewhere. In the study period, deaths in hospital decreased from 49.7% to 42.2% (annual decrease 0.8%; Pz=59.6<0.001), and home deaths increased from 22.3% to 27.6% (annual increase 0.4%; Pz=48.7<0.0001). Independent risk factors for dying in hospitals were: being older (Adjusted OR75–84 vs 18–54 (AOR): 1.25; 95% confidence interval (95%CI): 1.24–1.26) and living in deprived areas (AORs:1.06–1.31). Leukaemia (AORref=lung: 3.66; 95%CI: 3.65–3.68), Non-Hodgkin’s Lymphoma (AORref=lung: 2.89; 95%CI: 2.88–2.90) and Bladder cancer (AORref=1.46; 95%CI: 1.45–1.47) were the top three cancer deaths most likely to occur in hospitals. Men had a slightly higher chance than women of dying in hospitals (AOR: 1.13; 95%CI: 1.13–1.14). Cancer deaths in hospitals were more likely to occur in London than in South West (AORs range: 0.84 in South East Coast to 1.27 in London).
Conclusion We found an overall reducing trend in hospital deaths and an increasing trend in home deaths in the past ten years. However, significant inequality in place of death still exists. Future research needs to explore the underlying reasons.
U2 - 10.1136/jech-2012-201753.070
DO - 10.1136/jech-2012-201753.070
M3 - Meeting abstract
SN - 0143-005X
VL - 66
SP - A27-A27
JO - Journal of Epidemiology and Community Health
JF - Journal of Epidemiology and Community Health
ER -
