Access to this full-text is provided by Taylor & Francis.
Content available from Cancer Management and Research
This content is subject to copyright. Terms and conditions apply.
© 2018 Bähler et al. This work is published and licensed by Dove Medical Press Limited. The full terms of this license are available at https://www.dovepress.com/terms.
php and incorporate the Creative Commons Attribution – Non Commercial (unported, v3.0) License (http://creativecommons.org/licenses/by-nc/3.0/). By accessing the work
you hereby accept the Terms. Non-commercial uses of the work are permitted without any further permission from Dove Medical Press Limited, provided the work is properly attributed. For
permission for commercial use of this work, please see paragraphs 4.2 and 5 of our Terms (https://www.dovepress.com/terms.php).
Cancer Management and Research 2018:10 481–491
Cancer Management and Research Dovepress
submit your manuscript | www.dovepress.com
Dovepress 481
ORIGINAL RESEARCH
open access to scientific and medical research
Open Access Full Text Article
http://dx.doi.org/10.2147/CMAR.S156566
Intensity of treatment in Swiss cancer patients at
the end-of-life
Caroline Bähler1
Andri Signorell1
Eva Blozik1,2
Oliver Reich1
1Department of Health Sciences,
Helsana Insurance Group, Zürich,
Switzerland; 2Department of Medicine,
University Medical Centre Freiburg,
Freiburg im Breisgau, Germany
Purpose: Current evidence on the care-delivering process and the intensity of treatment at
the end-of-life of cancer patients is limited and remains unclear. Our objective was to examine
the care-delivering processes in health care during the last months of life with real-life data of
Swiss cancer patients.
Patients and methods: The study population consisted of adult decedents in 2014 who
were insured at Helsana Group. Data on the final cause of death were provided additionally by
the Swiss Federal Statistical Office. Of the 10,275 decedents, 2,710 (26.4%) died of cancer.
Intensity of treatment and health care utilization (including transitions) at their end-of-life were
examined. Intensity measures included the following: last dose of chemotherapy within 14 days
of death, a new chemotherapy regimen starting <30 days before death, more than one hospital
admission or spending >14 days in hospital in the last month, death in an acute care hospital,
more than one emergency visit and ≥1 intensive care unit admission in the last month of life.
Results: In the last 6 months of life, 89.5% of cancer patients had ≥1 transition, with 87.2%
being hospitalized. Within 30 days before death, 64.2% of the decedents had ≥1 intensive treat-
ment, whereby 8.9% started a new chemotherapy. In the multinomial logistic regression model,
older age, higher density of nursing home beds and home care nurses were associated with a
decrease, while living in the Italian- or French-speaking part of Switzerland was associated
with an increase in intensive care.
Conclusion: Swiss cancer patients insured by Helsana Group experience a considerable number
of transitions and intensive treatments at the end-of-life, whereby treatment intensity declines
with increasing age. Among others, increased home care nursing might be helpful to reduce
unwarranted treatments and transitions, therefore leading to better care at the end-of-life.
Keywords: cancer, end-of-life care, health care costs, transitions, intensity of treatment, health
insurance, regional variation
Introduction
Cancer is the second most common cause of death in Switzerland. In 2014, >16,000
(26%) of all decedents died of cancer.1 In 12 European Union countries, cancer was
shown to have overtaken cardiovascular diseases as the leading cause of death.2 Nowa-
days, modern medicine provides many treatment options for patients with cancer and
it is often believed that cancer patients undergo numerous treatments up until old
age at the end-of-life. Findings by Earle et al revealed that 30% of elderly decedents
had at least one indicator of intensive care (i.e., chemotherapy) at the end-of-life
when measured by means of claims-based data.3 Previous studies have also found an
increase in the use of intensive cancer care at the end-of-life.3,4 According to a more
Correspondence: Caroline Bähler
Department of Health Sciences, Helsana
Insurance Group, PO Box, 8081 Zürich,
Switzerland
Tel +41 58 340 52 01
Fax +41 58 340 04 34
Email caroline.baehler-baumgartner@
helsana.ch
Journal name: Cancer Management and Research
Article Designation: ORIGINAL RESEARCH
Year: 2018
Volume: 10
Running head verso: Bähler et al
Running head recto: Intensity of cancer care at the end-of-life
DOI: http://dx.doi.org/10.2147/CMAR.S156566
This article was published in the following Dove Press journal:
Cancer Management and Research
Cancer Management and Research 2018:10
submit your manuscript | www.dovepress.com
Dovepress
Dovepress
482
Bähler et al
recent review by Langton et al, almost 40% of mainly elderly
patients received chemotherapy or life-sustaining interven-
tions in their last month of life.5 But some of these treatment
options might not be advisable for cancer patients at their
end-of-life. Physicians, patients as well as their relatives are,
therefore, confronted with difficult decisions whether or not
to provide (further) treatments at the end-of-life.6 Thereby,
more intensive care was not always regarded as better care.7
In addition, studies have shown that early use of palliative
care may, among others, result in prolonged life expectancy
and less pain in dying patients.8–10
Earlier studies have found considerable variations con-
cerning chemotherapy and radiotherapy treatment at the
end-of-life in cancer patients.3,11 The number of transitions
between care settings also differed considerably between
regions in Switzerland in the last 6 months before death.12
While some differences in the intensity of treatment at the
end-of-life certainly are justified, others might indicate over-
use or underuse of treatment options in particular regions or
subgroups.13,14 The availability of health care services may
play an important role: decedents who resided in regions
with a higher availability of hospice services were shown to
receive less intensive care at the end-of-life.3 However, little
is known in Switzerland about the care of cancer patients at
the end-of-life. Previous research mainly concentrated on
chemotherapy, radiotherapy and hospitalizations in the last
30 days of life.11,15,16 A large study of four European coun-
tries regarding the health care utilization and the care setting
transitions in the last 3 months of life of cancer patients did
not include Swiss patients.17 Knowledge in the field of health
care utilization at the end-of-life is decisive not only for health
care providers but also for cancer patients and their relatives.
Study aims
This study aimed to examine the care-delivering processes in
health care during the last months of life with real-life data
of Swiss cancer patients. Age differences in the intensity of
treatment, using established criteria, as well as in health care
utilization (including transitions) were examined, thereby
accounting for a variety of individual and regional factors.3
Our goal was to establish clarity on whether differences in the
intensity of care between regions or patient groups exist, and
if so, to what extent. The indicators for intensity cannot judge
the quality of care provided to cancer patients, but can be
used as red flags to identify regions or groups with potential
challenges concerning the coordination of care at the end-
of-life. The identification of such differences is the first step
in addressing possible overuse or underuse. Former research
has shown that it is feasible to assess intensity of care at the
end-of-life using routine claims and further administrative
databases.3,7,18–20
Patients and methods
Study population
Helsana is one of the largest health insurers in Switzerland
and covered almost one-fifth of all decedents in Switzerland
in 2014. The study population consists of a retrospective
cohort of decedents in 2014 who were insured by Helsana
Insurance Group, and of whom data on the last year of life
were available. Further, data on the final cause of death of
the decedents were provided by the Swiss Federal Statistical
Office and merged with insurance claims data.
A total of 11,356 adult decedents were included in the
Helsana population. Of these, 599 (5.3%) were excluded
due to lack of detailed information on health care utiliza-
tion (i.e., decedents living abroad, or lump sums used for
reimbursement in nursing home residents), and 482 (4.2%)
were excluded due to unknown cause of death or inability to
merge the two datasets. Finally, 10,275 decedents of the year
2014 were included, of whom 2,710 (26.4%) died of cancer
(or its complications at an early stage).
A previous study examining the representativeness of
the database found that the Helsana population was slightly
older than the average Swiss population.21 But due to the
geographical diversity covered by the Helsana database, the
study contains information of a variety of cultural, social,
regional and otherwise divergent populations.
As this study is retrospective and based on anonymized
routine administrative health care claims data, a patient
informed consent was not needed, according to the national
ethical and legal regulation (article 22 of the Swiss data
protection law). Furthermore, a formal request was sent to
the Ethics committee Kantonale Ethikkommission Zürich
in the Canton of Zurich. According to this committee, no
further ethics approval was needed as the study falls outside
the scope of the Swiss Federal Act on Research involving
Human Beings (Human Research Act).
Measures
Cancer patients were defined as those decedents whose under-
lying cause of death was cancer. Based on the study by Earle
et al,3 intensity of care at the end-of-life was defined as the
occurrence of any of the following indicators: 1) last dose
of chemotherapy within 14 days of death, 2) a new chemo-
therapy regimen starting <30 days before death, 3) more than
one emergency visit in the last month of life, 4) more than
Cancer Management and Research 2018:10 submit your manuscript | www.dovepress.com
Dovepress
Dovepress
483
Intensity of cancer care at the end-of-life
one hospital admission or spending >14 days in hospital in
the last month of life, 5) death in an acute care hospital and
6) at least one intensive care unit (ICU) admission in the last
month of life. These indicators of the intensity of end-of-life
cancer care were derived from administrative data such as
claims data, and were identified by focus groups with cancer
patients and family members as well as with expert panels.22
An emergency visit was counted only if the patient spent at
least 1 h in the ICU. For the purpose of multivariate analysis,
the number of intensive treatments at the end-of-life was
divided into three groups: 0, 1 and 2+ treatment(s).
Additionally, transitions (defined as a change in the health
care setting as identified by the claims data23) and the place
of death (hospital, nursing home, home/others) were ana-
lyzed. The number of transitions between health care settings
(hospital, nursing home, home/others) and the proportion of
decedents with burdensome transitions, as defined by Teno et
al,23 were calculated. A burdensome transition was defined as
three or more hospitalizations in the last 90 days of life, or
at least one transition in the last 3 days of life.23 Health care
utilization also comprised the proportion of decedents with
at least one hospitalization and the median number of days in
hospital, as well as the proportion of decedents with at least
one nursing home admission with the median length of stay.
Furthermore, the number of physician visits by primary care
physicians and specialists is shown.
The following variables concerning patient characteristics
were considered: age, sex, cause of death, number of chronic
conditions, health insurance plan, as well as regional covari-
ates: language region, community characters (urban vs. rural),
the cantonal density of hospital beds (ranging from 1.8 to 11.5
per 1,000 inhabitants), the cantonal density of nursing home
beds (ranging from 48.3 to 120.9 per 1,000 inhabitants aged
65 years and older), as well as the density of home care nurses
(ranging from 1.4 to 3.8 per 1,000 inhabitants) and ambulatory
care physicians on the cantonal level (ranging from 92 to 425
per 100,000 inhabitants). Data on the final cause of death of
the decedents were provided by the Swiss Federal Statistical
Office. The merging was performed on the premises of the
Swiss Federal Statistical Office using only few predefined key
variables, excluding the name and address. Chronic conditions
were identified based on the Anatomical Therapeutic Chemi-
cal classification system, using an updated measure of the
Pharmacy-based Cost Group model by Huber et al.24 The fol-
lowing 22 chronic conditions were distinguished: acid-related
disorders, bone diseases (osteoporosis), cancer, cardiovascular
diseases (including hypertension), dementia, diabetes melli-
tus, epilepsy, glaucoma, gout/hyperuricemia, human immu-
nodeficiency virus, hyperlipidemia, intestinal inflammatory
diseases, iron-deficiency anemia, migraines, pain, Parkinson’s
disease, psychological disorders (sleep disorders, depression),
psychoses, respiratory illness (asthma, COPD), rheumatologic
conditions, thyroid disorders, tuberculosis.
Statistical analysis
Population characteristics, health care utilization, as well as
transitions between health care settings in cancer patients in
the last 6 months of life were assessed and compared to those
dying of other causes by means of descriptive statistics. Data
are presented as percentages for categorical variables and as
medians and interquartile ranges (IQRs; presented as the first
and the third quartiles) for continuous variables.
Intensive treatments for the end-of-life in patients who died
of cancer were assessed. Age differences regarding the intensity
of cancer treatment are presented using boxplots. Multinomial
logistic regression analysis was performed to determine the
impact of individual and regional variables on the odds of inten-
sive cancer treatments at the end-of-life.25 Odds ratios (ORs)
are shown for the three groups of intensive treatments: 0, 1 and
2 or more treatments. A threshold for statistical significance of
p=0.05 (two sided) was utilized. Because hospice use cannot be
identified by means of the present data, a sensitivity analysis
was carried out to examine health care utilization and treatment
intensity in cancer decedents, thereby excluding the cancer
decedents who lived in 4 out of 26 cantons with hospice avail-
ability (Aargau, Luzern, Zürich and Schwyz).26 All analyses
were carried out using R statistics, version 3.2.0 (R Foundation
for Statistical Computing, Vienna, Austria).
Results
The median (IQR) age of the total analytic sample of 10,275
decedents was 84 (75–90) years, and almost 53% were
women. The distribution of the decedents in different age
groups was as follows: 453 (16.7%) of the cancer decedents
were aged 18–64 years, 716 (26.4%) were aged 65–74 years,
908 (33.5%) were aged 75–84 years and 633 (23.4%) of the
cancer decedents were aged 85 years and older. In non-cancer
decedents, the distribution was as follows: 595 (7.9%) were
aged 18–64 years, 710 (9.4%) were aged 65–74 years, 2,025
(26.8%) were aged 75–84 years and 4,235 (56.0%) of the
decedents were aged 85 years and older. The characteristics
of the study population are presented in Table 1. Cancer
decedents were on average younger and they more often lived
in the Italian part of Switzerland and in rural areas compared
with decedents dying of other causes. Furthermore, they
more often had supplementary hospital insurance and higher
deductibles, although differences in the latter were rather
small. The main causes of death of those not dying of cancer
Cancer Management and Research 2018:10
submit your manuscript | www.dovepress.com
Dovepress
Dovepress
484
Bähler et al
were diseases of the circulatory system (29.0%), followed
by mental, behavioral and neurodevelopmental disorders
(8.2%), stroke (6.3%), diseases of the respiratory system
(6.0%) and diseases of the nervous system (5.3%). Cancer
decedents most frequently died of lung cancer (18.5%), fol-
lowed by colorectal cancer (9.6%) and hematologic cancer
(8.9%). The frequencies of cancer types as the main cause of
death, separately for men and women, are shown in Table 2.
Overall, 89.5% of the cancer decedents were transferred at
least once in the last 6 months of life, compared with 58.7%
of the decedents dying from other causes, whereby 87.2% of
cancer decedents and 54.3% of non-cancer decedents were
hospitalized (Table 3). Nearly one-fifth of patients dying of
cancer and only about 1 in 10 patients not dying of cancer
had at least one burdensome transition (defined as three or
more hospitalizations in the last 90 days of life, or at least
one transition in the last 3 days of life). The median number
of acute hospital admissions was twice as high in cancer
decedents (2 vs. 1; p<0.001). The number of consultations
by specialists was substantially higher in cancer decedents,
whereas no differences were found in the number of primary
care physician consultations. Almost 60% of cancer dece-
dents died in the hospital, whereas decedents dying of other
causes most frequently died in nursing homes.
Looking at cancer decedents, almost 65% had at least one
intensive treatment (as defined by Earle et al3) at the end-of-
life, whereby the number of intensive treatments decreased
with increasing age (Figure 1). Altogether, 35.8% of cancer
decedents had none, 23.8% had one and 40.4% had two and
more intensive cancer treatment(s). The median (IQR) number
of intensive treatments was 2.0 (1.0–2.0) in the youngest and
0.0 (0.0–1.0) in the oldest age group (p<0.001). In contrast,
no statistically significant sex-specific differences were found.
The proportions of patients with a specific intensive treat-
ment varied markedly (Table 4). The proportion of decedents
was highest in the youngest age group and lowest in the oldest
age group for all intensive treatments. Variations between
language areas were found for all intensive treatments except
for starting a new chemotherapy regimen in the last month of
life and more than one emergency visit in the last month of
life. Receiving a last dose of chemotherapy within 14 days
of death, in-hospital death and ICU admission(s) more fre-
quently occurred in the Italian-speaking part of Switzerland,
while a higher proportion of patients with more than one
hospital admission or spending >14 days in hospital in the
last month of life was found in the French-speaking regions.
In the multinomial regression model, decedents in the
French or Italian part of Switzerland had a 2.4 times higher
odds for 2+ intensive treatments compared with cancer dece-
dents in the German part (Table 5). Moreover, each increase
in the density of hospital beds was associated with a 1.14
higher odds for 2+ treatments. In contrast, each increase in
the density of home care nurses or nursing home beds was
associated with a 17.4 and a 1.3 lower odds for one intensive
Table 1 Characteristics of the study population (N=10,275)
Characteristics of study
population
Total Decedents
dying of cancer
Decedents dying
of other causes
p-value
n 10,275 2,710 (26.4%) 7,565 (73.6%)
Female sex 5,412 (52.7%) 1,228 (45.3%) 4,184 (55.3%) <0.001
Age, years, median (IQR) 84.0 (75–90) 77.0 (68–84) 86.0 (79–91) <0.001
Language area <0.001
German 8,199 (79.8%) 2,114 (78.0%) 6,085 (80.4%)
French 1,598 (15.6%) 423 (15.6%) 1,175 (15.5%)
Italian 478 (4.7%) 173 (6.4%) 305 (4.0%)
Type of residence (urban area) 3,312 (32.2%) 797 (29.4%) 2,515 (33.2%) <0.001
Chronic conditions, median (IQR) 4.0 (3.0–6.0) 5.0 (3.0–6.0) 4.0 (2.0–5.0) <0.001
Managed care 3,190 (31.0%) 1,017 (37.5%) 2,173 (28.7%) <0.001
Supplementary hospital insurance 2,031 (19.8%) 617 (22.8%) 1,414 (18.7%) <0.001
Higher deductible (>CHF 500) 599 (5.8%) 191 (7.0%) 408 (5.4%) 0.002
Notes: p-values, assigning the differences between the decedents dying of cancer vs. dying of other causes, were calculated using Fisher’s exact test for dichotomous variables,
Wilcoxon rank sum test for continuous variables and chi-squared test for categorical variables. Rhaeto-Romanic area is assigned to the German area.
Abbreviation: IQR, interquartile range (presented as the rst and the third quartiles).
Table 2 Cause of death by cancer type and sex (n=2,710)
Type of cancer Men, n (%) Women, n (%) Total, n (%)
Lung cancer 318 (21.5) 184 (15.0) 502 (18.5)
Colorectal cancer 138 (9.3) 121 (9.9) 259 (9.6)
Hematologic cancer 125 (8.4) 115 (9.4) 240 (8.9)
Prostate cancer 236 (15.9) 236 (8.7)
Breast cancer 212 (17.3) 212 (7.8)
Pancreas cancer 85 (5.7) 103 (8.4) 188 (6.9)
Other cancer types 580 (39.1) 493 (40.1) 1,073 (39.6)
Cancer Management and Research 2018:10 submit your manuscript | www.dovepress.com
Dovepress
Dovepress
485
Intensity of cancer care at the end-of-life
Table 3 Health care utilization of decedents dying of cancer vs. dying of other causes in their last 6 months of life (N=10,275)
Health care use Total Decedents dying
of cancer
Decedents dying
of other causes
p-value
n (%) 10,275 2,710 (26.4) 7,565 (73.6)
Individuals with at least one transition 6,867 (66.8) 2,425 (89.5) 4,442 (58.7) <0.001
Number of transitions,a median (IQR) 2.0 (1–4) 3.0 (1–5) 2.0 (1–3) <0.001
Individuals with burdensome transition(s) 1,341 (13.1) 450 (16.6) 891 (11.8) <0.001
Individuals with hospitalization(s) 6,472 (63.0) 2,364 (87.2) 4,108 (54.3) <0.001
Length of stay,a median (IQR) 35.0 (13–91) 31.0 (16–61) 38.0 (11–137) <0.001
Individuals with nursing home admission(s) 4,686 (45.6) 743 (27.4) 3,943 (52.1) <0.001
Length of stay,a median (IQR) 176.0 (80–181) 66.0 (20–170) 179.0 (127–181) <0.001
Individuals with consultation(s) 9,816 (95.5) 2,670 (98.5) 7,146 (94.5) <0.001
Number of consultations,a median (IQR) 12.0 (6–20) 19.0 (11–29) 10.0 (5–16) <0.001
By primary care physicians 6.0 (2–11) 6.0 (2–11) 6.0 (2–10) NS
By specialists 3.0 (1–10) 11.0 (3–21) 2.0 (0–6) <0.001
Place of death <0.001
Home 2,269 (22.1) 559 (20.6) 1,710 (22.6)
Hospital 4,133 (40.2) 1,546 (57.0) 2,587 (34.2)
Nursing home 3,873 (37.7) 605 (22.3) 3,268 (43.2)
Notes: p-values, assigning the differences between the decedents dying of cancer vs. of other causes, were calculated using Fisher’s exact test for dichotomous variables,
Wilcoxon rank sum test for continuous variables and chi-squared test for categorical variables. aIn decedents with at least one admission, transition or consultation,
respectively.
Abbreviations: IQR, interquartile range (presented as the rst and the third quartiles); NS, not signicant.
18–64 65–74
Age group
Number of intensive
treatments
75–84
Male
Female
85+
0.0
0.5
1.0
1.5
2.0
Figure 1 Mean number of intensive treatments at the end-of-life, divided by sex and age group (n=2,710).
Table 4 Proportions of cancer decedents with intensive treatments at the end-of-life by age group (n=2,710)
Intensive treatment Age in years p-value
Total
(n=2,710)
18–64
(n=453)
65–74
(n=716)
75–84
(n=908)
85+
(n=633)
Last dose of chemotherapy within 14 days of death 199 (7.3%) 63 (13.9%) 75 (10.5%) 52 (5.7%) 9 (1.4%) <0.001
Starting a new chemotherapy regimen ≤30 days before
death
242 (8.9%) 73 (16.1%) 88 (12.3%) 70 (7.7%) 11 (1.7%) <0.001
More than one emergency visit in the last month of life 56 (2.1%) 18 (4.0%) 12 (1.7%) 21 (2.3%) 5 (0.8%) 0.003
More than one hospital admission or spending >14 days
in hospital in the last month of life
1,053 (38.9%) 237 (52.3%) 321 (44.8%) 345 (38.0%) 150 (23.7%) <0.001
Spending >14 days in hospital in the last month of life 976 (36.0%) 213 (47.0%) 295 (41.2%) 330 (36.3%) 138 (21.8%) <0.001
More than one hospital admission in the last month of
life
174 (6.4%) 47 (10.4%) 58 (8.1%) 47 (5.2%) 22 (3.5%) <0.001
Death in an acute care hospital 1,522 (56.2%) 323 (71.3%) 464 (64.8%) 504 (55.5%) 231 (36.5%) <0.001
At least one ICU admission in the last month of life 187 (6.9%) 46 (10.2%) 48 (6.7%) 68 (7.5%) 25 (3.9%) <0.001
Note: p-values, assigning the differences between the different age groups, were calculated using chi-squared test.
Abbreviation: ICU, intensive care unit.
Cancer Management and Research 2018:10
submit your manuscript | www.dovepress.com
Dovepress
Dovepress
486
Bähler et al
treatment, and a 15.5 and a 1.1 lower odds for 2+ intensive
treatments in cancer patients, respectively. As found previ-
ously, older age was substantially associated with lower odds
of one or two or more intensive treatments compared with
their younger counterparts. Patients with a higher number of
chronic conditions were also less likely to undergo two or
more treatments. The type of cancer also had a significant
impact on the intensity of treatment: intensive care was
generally more likely in patients with lung cancer and less
likely in patients with prostate cancer.
In the subpopulation of cancer decedents who lived in
a canton without hospice availability, 1,159 (69.1%) of the
1,677 cancer patients had at least one intensive treatment
at the end-of-life, 90.2% were transferred at least once in
the last 6 months of life, whereby 88.7% were hospitalized.
Sixty-one percent of patients died in an acute care hospi-
tal. In contrast, 64.2% of the cancer decedents from the
total study population had at least one intensive treatment
and 89.5% had at least one transition, with 87.2% being
hospitalized during the corresponding time span. Also,
56.2% of the cancer patients died in an acute care hospital.
Table 6 shows the proportions of cancer decedents of the
subpopulation with intensive treatments at the end-of-life
by age group.
Table 5 Multinomial logistic regression model on intensive cancer treatment groupsa
Predictors Intensive cancer treatment group
Group 1, OR 95% CI Group 2, OR 95% CI
Age (in years) 0.973 0.964–0.982 0.945 0.937–0.954
Sex (females) 0.939 0.748–1.180 0.922 0.751–1.130
Language areab
German 1.000 1.000
French 1.224 0.835–1.795 2.419 1.733–3.377
Italian 1.522 0.888–2.607 2.351 1.461–3.783
Density of nursing home beds 0.987 0.975–0.998 0.989 0.979–0.999
Density of hospital beds 0.993 0.880–1.120 1.139 1.027–1.264
Density of home care nurses 0.826 0.685–0.996 0.845 0.714–0.999
Density of ambulatory care physicians 1.000 0.998–1.003 0.999 0.996–1.001
Cancer group
Colorectal cancer 1.000 1.000
Hematologic cancer 1.178 0.723–1.917 1.431 0.938–2.182
Lung cancer 1.640 1.085–2.478 1.314 0.907–1.903
Breast cancer 1.216 0.746–1.982 0.674 0.423–1.073
Pancreas cancer 0.947 0.562–1.595 1.045 0.667–1.636
Prostate cancer 0.756 0.467–1.224 0.569 0.365–0.885
Other cancer types 1.098 0.759–1.590 0.991 0.715–1.373
Number of chronic conditions 0.983 0.936–1.032 0.930 0.890–0.972
Notes: Statistically signicant differences are presented in bold. aGroup 1 = one intensive treatment (n=644); Group 2 = two or more intensive treatments (n=1,059).
bRhaeto-Romanic area is assigned to the German area.
Abbreviation: OR, odds ratio.
Table 6 Proportions of cancer decedents with intensive treatments at the end-of-life by age group, whereby cancer decedents who lived in one of
the four cantons with hospice availability were excluded (n=1,677)
Intensive treatment Age in years p-value
Total
(n=1,677)
18–64
(n=280)
65–74
(n=464)
75–84
(n=562)
85+
(n=371)
Last dose of chemotherapy within 14 days of death 128 (7.6%) 35 (12.5%) 52 (11.2%) 35 (6.2%) 6 (1.6%) <0.001
Starting a new chemotherapy regimen ≤30 days before
death
154 (9.2%) 43 (15.4%) 56 (12.1%) 48 (8.5%) 7 (1.9%) <0.001
More than one emergency visit in the last month of life 35 (2.1%) 11 (3.9%) 9 (1.9%) 14 (2.5%) 1 (0.3%) 0.010
More than one hospital admission or spending >14 days
in hospital in the last month of life
724 (43.2%) 155 (55.4%) 227 (48.9%) 240 (42.7%) 102 (27.5%) <0.001
Spending >14 days in hospital in the last month of life 675 (40.3%) 138 (49.3%) 211 (45.5%) 231 (41.1%) 95 (25.6%) <0.001
More than one hospital admission in the last month of life 112 (6.7%) 30 (10.7%) 39 (8.4%) 31 (5.5%) 12 (3.2%) <0.001
Death in an acute care hospital 1,023 (61.0%) 211 (75.4%) 324 (69.8%) 331 (58.9%) 157 (42.3%) <0.001
At least one ICU admission in the last month of life 116 (6.9%) 32 (11.4%) 32 (6.9%) 40 (7.1%) 12 (3.2%) <0.001
Note: p-values, assigning the differences between the different age groups, were calculated using chi-squared test.
Abbreviation: ICU, intensive care unit.
Cancer Management and Research 2018:10 submit your manuscript | www.dovepress.com
Dovepress
Dovepress
487
Intensity of cancer care at the end-of-life
Discussion
This is the first study exploring detailed patient-relevant
dimensions of treatment intensity for cancer patients at the
end-of-life in a real-life Swiss population. It reveals that the
intensity of treatment and the number of transitions are high
in cancer patients.
According to our findings, almost 65% of Swiss cancer
decedents had at least one intensive treatment at the end-of-
life. This proportion is higher compared with the findings
from studies conducted considerably earlier in other health
care settings and cultural contexts. Only 30% of all decedents
had at least one indicator of intensive care in the US study by
Earle et al3 conducted in 1993–1996. However, they found
an increase in the intensity of treatment within the observed
time span. Ho et al4 stated that 22.4% of Canadian adults
with a cancer cause of death experienced at least one inci-
dent of potentially aggressive cancer care at the end-of-life.
However, some of the treatments such as death in an acute
hospital, which was shown to be frequent in our cohort, were
not considered in their definition of potentially aggressive
cancer care and, thus, explains the higher rate of intensive
treatment in this study. Including only these four indicators
for potentially aggressive cancer care, as used by Ho et al,4
in our analysis would result in a proportion of 18.0% of
cancer patients with at least one intensive treatment at the
end-of-life.
In our study, 56% of the cancer patients died in an acute
hospital. This is widely in line with the literature. In an
international cohort study using administrative and registry
data of 2010 on health care utilization at the end-of-life,
high proportions of cancer patients dying in acute hospitals
were found in Belgium (51.2%), Canada (52.1%), England
(41.7%), Germany (38.3%) and Norway (44.7%), when com-
pared with the USA (22.2%) and the Netherlands (29.4%).27
The low rate in the USA might be due to the high pressure
on costs forcing patients to move from hospitals to hospices
or nursing homes, as well as to the high level of hospice
capacity. In Taiwan, almost two-thirds of the cancer patients
died in an acute care hospital.28 According to a mortality
follow-back study in London, dying at home was associated
with more peace for decedents in their last week of life and
with less grief in their relatives, but preceding discussions
about patients’ preferences are needed.29 Moreover, this high
proportion of hospital deaths is contrary to what the Swiss
policy was seeking and to what patients prefer according to
a representative Swiss survey.30
Our findings revealed that about 7% of the decedents
received a last dose of chemotherapy within 14 days of death
and nearly 9% started a new chemotherapy regimen within
30 days of death. In the review by Langton et al,5 chemo-
therapy was delivered to 1%–19% of patients in the last 14
days. In the retrospective Canadian study, chemotherapy
was administered to 2.4%–4.8% of decedents in the last 2
weeks of life.31 However, in this latter study, only intravenous
administration of chemotherapy could be considered. In the
study by Earle et al,3 5.7% of the decedents received a new
chemotherapy within 30 days of death. This is comparable
to 7.5% of elderly patients (65 years and older) receiving a
new chemotherapy in our cohort. In Portuguese patients with
advanced solid tumors, 14% started a new chemotherapy
regimen within 30 days of death.32 In a prospective US study,
the use of chemotherapy at the end-of-life was associated
with an increased risk of cardiopulmonary resuscitation,
mechanical ventilation and dying in an ICU rather than in
their preferred place of death.33
Almost 40% of the patients in our cohort spent >14 days
in the hospital and roughly 6% had multiple hospital admis-
sions in the last month of life. These results correspond to
findings from Taiwan, where 46.2% spent >14 days in the
hospital and 13.6% had multiple hospital admissions dur-
ing their last month of life.28 In the study by Earle et al,3
only 11.6% spent >14 days in the hospital in the last month
of life, but the proportion of patients with more than one
acute hospital admission is comparable (9.1%). Due to long
hospital stays in many patients of our cohort, the odds for
readmissions were lower. According to our analyses, 6.9%
had at least one ICU admission and 2.1% had more than one
emergency visit in the last month. ICU admission during the
same time span was seen in 3.5%–27.2% of cancer patients,
and emergency visits were found in 9.2%–57.6% of patients
in previous findings.3,27,28,34 Patients who spent <1 h in the
ICU were regarded as not having an admission to the ICU
in our study. This might have led to an underestimation of
the number of decedents with at least one ICU admission.
In sum, intensive care at the end-of-life assessed by
means of claims data is frequent in Switzerland, but varies
broadly across regions because of different health care sys-
tems, health care professionals involved in the care of those
patients as well as due to different cultural settings. While
some of these treatments are certainly fully appropriate and
in accordance with patient preferences, others might be the
result of a lack of coordination or good communication, or
the density parameters of health care providers.
We identified important differences according to the fac-
tors associated with treatment intensity that are assumingly
not directly linked with the patients’ characteristics, such as
Cancer Management and Research 2018:10
submit your manuscript | www.dovepress.com
Dovepress
Dovepress
488
Bähler et al
diagnosis. Specifically, the density of ambulatory health care
providers as well as the density of the health care institutions
seemed to have an important impact on the cancer treatment
at the end-of-life. A high density of nursing home beds and a
high density of home care nurses were both associated with
a lower likelihood of having one and two or more intensive
treatments at the end-of-life, respectively. In contrast, a high
density of hospital beds was related to more intensive treat-
ment. Albeit, it needs to be considered that the CI in the latter
association between the density of nursing home beds and
intensive treatments is relatively wide and must therefore
be interpreted with caution. Similar associations between
treatment intensity and density of health care providers and
infrastructure at the end-of-life were found in an earlier
study conducted in Switzerland,12 as well as in international
studies.4,5 In a recently published study, the use of home-
visit nursing was associated with a significantly lower risk
of hospitalizations related to pressure ulcers.35
Furthermore, we found considerable variations between
the different language areas, whereby Italian- and French-
speaking patients had more intensive treatments compared
with German-speaking patients. Similarly, patients residing
in the canton of Ticino (Italian-speaking part of Switzerland)
were hospitalized more frequently and more often received
anticancer drug therapy in the last month(s) of life.15,16 In a
recently published Swiss study regarding health care costs at
the end-of-life, considerable differences were found between
the different language areas even after extensive analysis of
a variety of individual and further regional characteristics.36
We, therefore, suggest that cultural aspects have an important
influence on the treatment intensity in Switzerland due to
dissimilar expectations and preferences of patients, family
members and health care providers.
Considering patient-related characteristics, the intensity
of treatment decreased almost linearly with increasing age
in our cohort even after adjustment for comorbidities. A
decrease in intensive cancer treatment with increasing age
at the end-of-life was documented previously.3–5 The same
was also found in a previous Swiss and US study regarding
the use of chemotherapy.16,37 Levinsky et al38 have found a
decrease in Medicare expenditures with increasing age in
the last year of life, after adjustment for several influencing
factors such as sex, place and cause of death, mainly due
to lower intensity of medical care with increasing age. In
contrast to foregoing analysis, we did not find a sex-specific
difference in the intensity of treatment.3–5
In a multinomial regression analysis, intensive treatment
was more likely in patients with lung cancer and less likely in
patients with prostate cancer, which is in line with Canadian
findings.4 However, CIs for all cancer types are rather wide
and need to be interpreted with caution. Further research
is, therefore, needed to clarify the impact of cancer type on
health care utilization measures such as transitions.
In the last 6 months of life, 90% of cancer patients of our
cohort had at least one transition with 87% being hospital-
ized, and >13% even had a burdensome transition. These
are among the highest proportions published so far. In a
large cohort study using administrative and registry data on
the utilization and costs at the end-of-life, between 69.9%
(Germany) and 88.7% (Belgium) of elderly patients with
cancer were hospitalized at least once in the last 6 months
of life.27 The proportion of cancer patients with at least one
transition between care settings in the last 3 months of death
amounted to 53%–69% in four European countries.17 In Aus-
tralia, the median number of hospital admissions (5 vs. 3)
and the median number of days in hospital (34 vs. 30) were
higher in cancer decedents when compared with non-cancer
decedents in their last year of life.39 This is comparable with
our analysis resulting in a median of two (acute only) hospital
admissions in the last 6 months of life with a median length
of stay of 31 days. A previous Swiss study on the treatment
of adult cancer patients from 2006 to 2008 with at least
one hospitalization in their final month of life revealed that
patients spent a median of 22 days in an acute hospital in the
last 3 months of life, with a median of two hospitalizations.15
Transitions, especially hospitalizations at the end-of-life,
can be very distressing. They may lead to discontinuity of
care and to a loss of information between patient and health
care providers, ending up in decreased quality of care at the
end-of-life.23,40 Therefore, transitions have previously been
proposed as indicators for quality of end-of-life care.41
In the subanalysis, excluding the cancer decedents who
lived in one of the four cantons with hospice availability, the
treatment intensity as well as further health care utilization
measures were slightly higher compared with the total study
population of cancer decedents. Similarly, cancer decedents
who resided in regions with a higher availability of hospice
services were shown to receive less intensive care at the end-
of-life in the USA.3 This is also in line with a previous study
that found a decreased risk of in-hospital death in regions with
greater hospice availability and use.42 However, differences
between our cohorts are rather small and further research is,
therefore, needed.
In this study, 2,710 (26.4%) of all decedents died of cancer
in the year 2014. Lung cancer was the leading cause of cancer
death. Our data correspond with previous findings on cancer
Cancer Management and Research 2018:10 submit your manuscript | www.dovepress.com
Dovepress
Dovepress
489
Intensity of cancer care at the end-of-life
research. According to the Organization for Economic Co-
operation and Development,43 cancer accounted for 25% of
all deaths in 2013, whereby lung cancer was the main cause
of cancer mortality. Cancer accounted for 26.2% of all deaths
in Switzerland according to the Swiss Federal Statistical
Office.1 Hence, our data can be regarded as representative
for the whole Swiss population.
Strengths and limitations
The analysis of this study is based on claims and census
data of a sizable cohort, representing almost one-fifth of
all Swiss decedents. The data cover a broad range of highly
reliable and comprehensive information on end-of-life care.
However, this is a retrospective study looking at patients with
a known date of death. In most cases, however, the date of
death cannot be foreknown, even though cancer has one of
the most predictable courses of disease. Nevertheless, two
studies comparing health care utilization and costs showed
very similar results when comparing their prospectively and
retrospectively collected data.34,44
Unfortunately, we were not able to distinguish patients
whose death was preceded by a clinical state characterized by
metastatic cancer and slow deterioration before death from
those who died from complications of cancer treatment at
an early (but not late) stage.45
Although our intensity measures have not been formally
validated, they were repeatedly used in other studies as well
as by the American Society of Clinical Oncology. The quality
of this set of end-of-life intensity measures was, therefore,
regarded as good.20 Furthermore, among others, ICU admis-
sion has been found to be relatively robust and stable in
terms of hospital end-of-life treatment intensity measure.46
Further research is, however, needed regarding the defini-
tion of quality end-of-life care indicators as well as the time
trends of treatment intensity. This would enable researchers
to better quantify the ratio between appropriate and inap-
propriate cancer treatments in future studies. Moreover, a
more detailed analysis of the association between treatment
intensity and the density of different types of specialists and
primary care physicians in the ambulatory setting might be
useful in explaining the regional differences in the treatment
patterns of cancer patients.47,48
This study cannot judge the quality of care provided to
patients at the end-of-life. The indicators for intensity can
only be used as red flags to identify regions or groups of
patients with potential challenges concerning the coordina-
tion of care at the end-of-life.22 Since the study looks at health
care utilization in the last months of life – independently of
the reason for use – the utilization might not be related to
cancer treatment. Additionally, influencing variables, such as
the stage, histology of the disease or patients’ preferences,
could not be taken into account.
Implications
This study has important implications related to the health
care system. A higher density of home care nurses was associ-
ated with a decrease in treatment intensity at the end-of-life
in our analysis. In Switzerland, considerable regional varia-
tions regarding the health services supply at the end-of-life
exist, ranging from regions with a comprehensive supply of
palliative care, including mobile palliative care teams, to
regions with hardly any offering. According to the National
Palliative Care Strategy 2013–2015,26 palliative care is not
yet available for all patients in need. But palliative care was
recently found to decrease the risk of hospital admission
(OR=0.21, 95% CI: 0.18–0.26) and in-hospital mortality
(OR=0.03, 95% CI: 0.02–0.04) in Italy.49 In addition, stud-
ies have shown that early use of palliative care may, result in
prolonged life expectancy and less pain in dying patients.8–10
Last but not least, better coordination and discharge planning
may help reduce end-of-life health care intensity.50 Besides
the strengthening of advanced care planning conversations,
a closer link between curative therapy and palliative care ser-
vices is warranted.51,52 Additionally, linking patient–provider
communication and outcomes might be helpful in developing
evidence-based interventions in cancer care.53 International
evidence demonstrated that the costs of introduction of com-
prehensive palliative care services may pay off.54 In light of
our results as well as of international study findings, a denser
supply of nursing services and a better coordination of care
are urgently needed.
There is an ongoing discussion regarding the intensity
of treatment at the end-of-life in highly industrialized
countries such as Switzerland. Many people are concerned
that particularly older cancer patients are overtreated with
chemotherapy. This is especially true for a country such as
Switzerland with one of the highest life expectancy world-
wide.55 Our data show that treatment intensity substantially
decreases with increasing age. Transparency concerning
health care utilization is important in order to learn from each
other to improve care at the end-of-life in cancer patients.
There is a huge variety of health care systems and no two
are alike. However, as many notably industrialized countries
aim at reducing hospitalizations and intensive treatments at
the end-of-life, comparisons between countries may help to
identify similarities and differences in treatment intensity
Cancer Management and Research 2018:10
submit your manuscript | www.dovepress.com
Dovepress
Dovepress
490
Bähler et al
and its underlying causes. The use of similar methodologies
and definitions (i.e., regarding the definition of treatment
intensity) to evaluate different health care delivery systems
across geographic areas and time periods is thus needed.
Quantification of cross-country variations in end-of-life
care can help guide policymakers with the implementation
of culturally and linguistically adapted strategies regarding
prevention, early detection or cancer treatment in regions
with highest cancer type-specific mortality.56
Conclusion
Swiss cancer patients in the last months of life experience a
considerable number of transitions and intensive treatments.
There is reason to fear that these care processes may not
be in accordance with patient preferences and current best
practice in end-of-life care. This study contributes to the
ongoing discussion on coordination of medical care at the
end-of-life in Swiss cancer patients. It indicates that factors
such as density per type of health care provider, the cultural
context and other nonclinical patient characteristics may
contribute to disparities in the management of those patients.
Increased care coordination and timely communication
between patients, families and health care professionals about
the goals of care might be helpful to reduce unwarranted
transitions and intensity of care at the end-of-life.
Acknowledgments
The authors are grateful to the Swiss Federal Statistical Office
for providing them the opportunity to use data on the final
cause of death. The authors also thank Sonja Wehrle and
Mikaël Thomas for their helpful support in coding in-patient
treatments and Annette Jamieson for her critical review of
the manuscript.
Disclosure
The authors report no conflicts of interest in this work.
References
1. Bundesamt für Statistik [Swiss Federal Statistical Office]. Todesur-
sachen des Jahres; 2014 [Causes of death in 2014]. https://www.bfs.
admin.ch/bfs/de/home/statistiken/gesundheit.assetdetail.40496.html.
Accessed November 15, 2016. German.
2. Townsend N, Wilson L, Bhatnagar P, Wickramasinghe K, Rayner M,
Nichols M. Cardiovascular disease in Europe: epidemiological update
2016. Eur Heart J. 2016;37(42):3232–3245.
3. Earle CC, Neville BA, Landrum MB, Ayanian JZ, Block SD, Weeks
JC. Trends in the aggressiveness of cancer care near the end of life. J
Clin Oncol. 2004;22(2):315–321.
4. Ho TH, Barbera L, Saskin R, Lu H, Neville BA, Earle CC. Trends
in the aggressiveness of end-of-life cancer care in the universal
health care system of Ontario, Canada. J Clin Oncol. 2011;29(12):
1587–1591.
5. Langton JM, Blanch B, Drew AK, Haas M, Ingham JM, Pearson S-A.
Retrospective studies of end-of-life resource utilization and costs in
cancer care using health administrative data: a systematic review. Palliat
Med. 2014;28(10):1167–1196.
6. Bosshard G, Nilstun T, Bilsen J, et al. Forgoing treatment at the end of
life in 6 European countries. Arch Intern Med. 2005;165(4):401–407.
7. Wright AA, Keating NL, Ayanian JZ, et al. Family perspectives on
aggressive cancer care near the end of life. JAMA. 2016;315(3):284–292.
8. Connor SR, Pyenson B, Fitch K, Spence C, Iwasaki K. Comparing
hospice and nonhospice patient survival among patients who die within
a three-year window. J Pain Symptom Manage. 2007;33(3):238–246.
9. Hamano J, Yamaguchi T, Maeda I, et al. Multicenter cohort study on
the survival time of cancer patients dying at home or in a hospital: does
place matter? Cancer. 2016;122(9):1453–1460.
10. Temel JS, Greer JA, Muzikansky A, et al. Early palliative care for
patients with metastatic non-small-cell lung cancer. N Engl J Med.
2010;363(8):733–742.
11. Matter-Walstra KW, Achermann R, Rapold R, et al. Delivery of health
care at the end of life in cancer patients of four Swiss cantons: a retro-
spective database study (SAKK 89/09). BMC Cancer. 2014;14:306.
12. Bähler C, Signorell A, Reich O. Health care utilisation and transitions
between health care settings in the last 6 months of life in Switzerland.
PLoS One. 2016;11(9):e0160932.
13. Glasziou P, Straus S, Brownlee S, et al. Evidence for underuse of effective
medical services around the world. Lancet. 2017;390(10090):169–177.
14. Brownlee S, Chalkidou K, Doust J, et al. Evidence for overuse of medi-
cal services around the world. Lancet. 2017;390(10090):156–168.
15. Matter-Walstra KW, Achermann R, Rapold R, et al. Days spent in
acute care hospitals at the end of life of cancer patients in four Swiss
cantons: a retrospective database study (SAKK 89/09). Eur J Cancer
Care (Engl). Epub 2016 Feb 9.
16. Matter-Walstra KW, Achermann R, Rapold R, et al. Cancer-related
therapies at the end of life in hospitalized cancer patients from four
Swiss cantons: SAKK 89/09. Oncology. 2015;88(1):18–27.
17. Ko W, Deliens L, Miccinesi G, et al. Care provided and care setting
transitions in the last three months of life of cancer patients: a nationwide
monitoring study in four European countries. BMC Cancer. 2014;14:960.
18. Alturki A, Gagnon B, Petrecca K, Scott SC, Nadeau L, Mayo N. Pat-
terns of care at end of life for people with primary intracranial tumors:
lessons learned. J Neurooncol. 2014;117(1):103–115.
19. Gagnon B, Mayo NE, Hanley J, MacDonald N. Pattern of care at the
end of life: does age make a difference in what happens to women with
breast cancer? J Clin Oncol. 2004;22(17):3458–3465.
20. Luta X, Maessen M, Egger M, Stuck AE, Goodman D, Clough-Gorr
KM. Measuring intensity of end of life care: a systematic review. PLoS
One. 2015;10(4):e0123764.
21. Huber CA, Diem P, Schwenkglenks M, Rapold R, Reich O. Estimating
the prevalence of comorbid conditions and their effect on health care
costs in patients with diabetes mellitus in Switzerland. Diabetes Metab
Syndr Obes. 2014;7:455–465.
22. Earle CC, Park ER, Lai B, Weeks JC, Ayanian JZ, Block SD. Identify-
ing potential indicators of the quality of end-of-life cancer care from
administrative data. J Clin Oncol. 2003;21(6):1133–1138.
23. Teno JM, Gozalo PL, Bynum JP, et al. Change in end-of-life care for
Medicare beneficiaries: site of death, place of care, and health care
transitions in 2000, 2005, and 2009. JAMA. 2013;309(5):470–477.
24. Huber CA, Szucs TD, Rapold R, Reich O. Identifying patients with
chronic conditions using pharmacy data in Switzerland: an updated
mapping approach to the classification of medications. BMC Public
Health. 2013;13:1030.
25. Hosmer DW, Lemeshow S. Applied Logistic Regression. 2nd ed. New
York, NY; Chichester: Wiley; 2000. Wilet series in probability and
statistics: texts and references section.
26. Bundesamt für Gesundheit und Schweizerische Konferenz der kan-
tonalen Gesundheitsdirektorinnen und -direktoren [Federal Office
of Public Health (FOPH) and the Swiss Conference of the Cantonal
Ministers of Public Health (CMPH)]. Nationale Strategie Palliative
Care 2013–2015; 2012. German.
Cancer Management and Research 2018:10 submit your manuscript | www.dovepress.com
Dovepress
Dovepress
Cancer Management and Research
Publish your work in this journal
Submit your manuscript here: https://www.dovepress.com/cancer-management-and-research-journal
Cancer Management and Research is an international, peer-reviewed
open access journal focusing on cancer research and the optimal use of
preventative and integrated treatment interventions to achieve improved
outcomes, enhanced survival and quality of life for the cancer patient.
The manuscript management system is completely online and includes
a very quick and fair peer-review system, which is all easy to use. Visit
http://www.dovepress.com/testimonials.php to read real quotes from
published authors.
Dovepress
491
Intensity of cancer care at the end-of-life
27. Bekelman JE, Halpern SD, Blankart CR, et al. Comparison of site of
death, health care utilization, and hospital expenditures for patients dying
with cancer in 7 developed countries. JAMA. 2016;315(3):272–283.
28. Tang ST, Wu S-C, Hung Y-N, Huang E-W, Chen J-S, Liu T-W. Trends
in quality of end-of-life care for Taiwanese cancer patients who died
in 2000–2006. Ann Oncol. 2009;20(2):343–348.
29. Gomes B, Calanzani N, Koffman J, Higginson IJ. Is dying in hospital
better than home in incurable cancer and what factors influence this?
A population-based study. BMC Med. 2015;13:235.
30. GfK Switzerland. Bevölkerungsbefragung Palliative Care [Swiss Survey
on Palliative Care]. Forschungsinstitut GfK: Switzerland AG; 2009.
31. Barbera L, Seow H, Sutradhar R, et al. Quality of end-of-life cancer
care in Canada: a retrospective four-province study using administrative
health care data. Curr Oncol. 2015;22(5):341–355.
32. Braga S, Miranda A, Fonseca R, et al. The aggressiveness of cancer care
in the last three months of life: a retrospective single centre analysis.
Psychooncology. 2007;16(9):863–868.
33. Wright AA, Zhang B, Keating NL, Weeks JC, Prigerson HG. Associations
between palliative chemotherapy and adult cancer patients, end of life
care and place of death: prospective cohort study. BMJ. 2014;348:g1219.
34. Setoguchi S, Earle CC, Glynn R, et al. Comparison of prospective and
retrospective indicators of the quality of end-of-life cancer care. J Clin
Oncol. 2008;26(35):5671–5678.
35. Lee HJ, Ju YJ, Park E-C, Kim J, Lee SG. Effects of home-visit nurs-
ing services on hospitalization in the elderly with pressure ulcers: a
longitudinal study. Eur J Public Health. 2017;27(5):822–826.
36. Panczak R, Luta X, Maessen M, et al. Regional variation of cost of care
in the last 12 months of life in Switzerland: small-area analysis using
insurance claims data. Med Care. 2017;55(2):155–163.
37. Emanuel EJ, Young-Xu Y, Levinsky NG, Gazelle G, Saynina O, Ash
AS. Chemotherapy use among Medicare beneficiaries at the end of life.
Ann Intern Med. 2003;138(8):639–643.
38. Levinsky NG, Yu W, Ash AS, et al. Influence of age on Medi-
care expenditures and medical care in the last year of life. JAMA.
2001;286(11):1349–1355.
39. Rosenwax LK, McNamara BA, Murray K, McCabe RJ, Aoun SM,
Currow DC. Hospital and emergency department use in the last year
of life: a baseline for future modifications to end-of-life care. Med J
Aust. 2011;194(11):570–573.
40. Morrison RS, Penrod JD, Cassel JB, et al. Cost savings associated with
US hospital palliative care consultation programs. Arch Intern Med.
2008;168(16):1783–1790.
41. Perrels AJ, Fleming J, Zhao J, et al. Place of death and end-of-life tran-
sitions experienced by very old people with differing cognitive status:
retrospective analysis of a prospective population-based cohort aged
85 and over. Palliat Med. 2014;28(3):220–233.
42. Pritchard RS, Fisher ES, Teno JM, et al. Influence of patient prefer-
ences and local health system characteristics on the place of death. J
Am Geriatr Soc. 1998;46(10):1242–1250.
43. Organisation for Economic Co-operation and Development. Health at
a Glance 2015: OECD Indicators; 2015. Available from: http://dx.doi.
org/10.1787/health_glance-2015-en. Accessed November 7, 2017.
44. Pyenson B, Connor S, Fitch K, Kinzbrunner B. Medicare cost in
matched hospice and non-hospice cohorts. J Pain Symptom Manage.
2004;28(3):200–210.
45. Alturki A. Patterns of Care at end of Life for People with Primary Intra-
cranial Tumors: Lessons Learned [thesis]. Montreal, QC: Department
of Epidemiology, Biostatistics and Occupational Health; 2013.
46. Barnato AE, Farrell MH, Chang C-CH, Lave JR, Roberts MS, Angus
DC. Development and validation of hospital “end-of-life” treatment
intensity measures. Med Care. 2009;47(10):1098–1105.
47. Engel-Nitz NM, Alemayehu B, Parry D, Nathan F. Differences in
treatment patterns among patients with castration-resistant prostate
cancer treated by oncologists versus urologists in a US managed care
population. Cancer Manag Res. 2011;3:233–245.
48. Ankuda CK, Petterson SM, Wingrove P, Bazemore AW. Regional
variation in primary care involvement at the end of life. Ann Fam Med.
2017;15(1):63–67.
49. Scaccabarozzi G, Limonta F, Amodio E. Hospital, local palliative care
network and public health: how do they involve terminally ill patients?
Eur J Public Health. 2017;27(1):25–30.
50. Brown JR, Chang C-H, Zhou W, MacKenzie TA, Malenka DJ, Good-
man DC. Health system characteristics and rates of readmission after
acute myocardial infarction in the United States. J Am Heart Assoc.
2014;3(3):e000714.
51. Gawande A. Quantity and quality of life: duties of care in life-limiting
illness. JAMA. 2016;315(3):267–269.
52. Ferris FD, Bruera E, Cherny N, et al. Palliative cancer care a decade
later: accomplishments, the need, next steps–from the American Society
of Clinical Oncology. J Clin Oncol. 2009;27(18):3052–3058.
53. Sheldon LK, Hong F, Berry D. Patient-provider communication data:
linking process and outcomes in oncology care. Cancer Manag Res.
2011;3:311–317.
54. Fassbender K, Fainsinger R, Brenneis C, Brown P, Braun T, Jacobs P.
Utilization and costs of the introduction of system-wide palliative care
in Alberta, 1993–2000. Palliat Med. 2005;19(7):513–520.
55. Organisation for Economic Co-operation and Development. Health
at a glance. 2013. Life Expectancy at Birth; 2013. Available from:
https://www.oecd.org/els/health-systems/Health-at-a-Glance-2013.pdf.
Accessed November 09, 2017.
56. Wheeler SB, Basch E. Translating cancer surveillance data into effective
public health interventions. JAMA. 2017;317(4):365–367.
Available via license: CC BY-NC 4.0
Content may be subject to copyright.
