Strang et al.

BMC Pulm Med (2021) 21:170

https://doi.org/10.1186/s12890-021-01533-3

RESEARCH Open Access

Chronic obstructive pulmonary disease

and lung cancer: access to palliative care,
emergency room visits and hospital deaths
Peter Strang1,2* , Per Fürst2,3, Christel Hedman2,4 , Jenny Bergqvist5,6, Helena Adlitzer7 and Torbjörn Schultz2

Abstract
Background: Despite the severe symptoms experienced by dying COPD patients, specialized palliative care (SPC)
services focus mainly on cancer patients. We aimed to study the access to SPC that COPD and lung cancer (LC)
patients receive and how that access affects the need for acute hospital care.
Methods: A descriptive regional registry study using data acquired through VAL, the Stockholm Regional Council’s
central data warehouse, which covers nearly all healthcare use in the county of Stockholm. All the patients who died
of COPD or LC from 2015 to 2019 were included. T-tests, chi-2 tests, and univariable and multivariable logistic regres-
sion analyses were performed on the accumulated data.
Results: In total, 6479 patients, (2917 with COPD and 3562 with LC) were studied. The patients with LC had more
access to SPC during the last three months of life than did those with COPD (77% vs. 18%, respectively; p < .0001),
whereas patients with COPD were more likely to be residents of nursing homes than those with LC (32% vs. 9%,
respectively; p < .0001). Higher socioeconomic status (SES) (p < .01) and patient age < 80 years (p < .001) were asso-
ciated with increased access to SPC for LC patients. Access to SPC correlated with fewer emergency room visits
(p < .0001 for both COPD and LC patients) and fewer admissions to acute hospitals during the last month of life
(p < .0001 for both groups). More COPD patients died in acute hospitals than lung cancer patients, (39% vs. 20%;
χ2 = 287, p < .0001), with significantly lower figures for those who had access to SPC (p < .0001).
Conclusions: Compared to dying COPD patients, LC patients have more access to SPC. Access to SPC reduces the
need for emergency room visits and admissions to acute hospitals.
Keywords: Chronic obstructive pulmonary disease, Lung cancer, Palliative care services, Place of death

Background disease, later on, comorbidities, such as ischemic heart

Although mild forms of chronic obstructive pulmonary disease, heart failure and arrythmias, thromboses, lung
disease (COPD) may remain undiagnosed, severe forms embolisms, osteoporosis, and lung cancer (LC) may con-
cause distressing and complex symptoms, especially in stitute a clinical burden [1]. As concluded in a review,
the end-of-life (EOL) [1]. While respiratory symptoms, comorbidities actually account for two-thirds of the
such as cough, sputum production, and dyspnea can deaths of COPD patients, whereas respiratory failure is
be disturbing for the patient early in the course of the the main cause in the others [2].
Today, COPD is a major cause of death in Sweden,
*Correspondence: peter.strang@ki.se
with almost 3000 deaths annually, compared to approxi-
1
Department of Oncology‑Pathology, Karolinska Institutet, Regional mately 3500 deaths from LC [3]. Despite successive dete-
Cancer Centre in Stockholm, Gotland, Sweden rioration, the typical COPD trajectory is an illness with
Full list of author information is available at the end of the article

© The Author(s) 2021. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which
permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the
original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or
other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line
to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory
regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this
licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/. The Creative Commons Public Domain Dedication waiver (http://​creat​iveco​
mmons.​org/​publi​cdoma​in/​zero/1.​0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.
Strang et al. BMC Pulm Med (2021) 21:170 Page 2 of 7

stable periods, abrupted by exacerbations, often accom- recent studies, only two included 1000 or more patients
panied by infections but also by concomitant heart fail- [18, 19]. For this reason, we aimed to conduct a large
ure [1, 4]. The severe exacerbations may end in recovery study using recent data, as inclusion criteria for admis-
or, conversely, death. For this reason, prognostication sion to palliative care services for non-malignant condi-
is difficult. Efforts have been made to establish reli- tions such as COPD are changing rapidly.
able prognostic factors, both in the form of single fac-
tors and as prognostic indices, with some consensus on
group level; however, difficulties in establishing a prog- Aims
nosis for individual patients remain [4, 5]. Respiratory In this study, we aimed to retrospectively compare the
factors, such as dyspnea, FEV1%, COPD exacerbations, access to SPC of COPD and LC patients during their last
exercise capacity, inspiratory fraction and inspiratory three months of life and analyze to what extent age, sex,
capacity are frequently suggested as predictors of death, or socioeconomic status (SES) influences the opportuni-
along with comorbidities and older age [6]. To improve ties to receive SPC. A further aim was to study whether
the predictive capacity of these variables, constructs have access to SPC reduced acute emergency visits and admis-
been used, such as BODE, comprising Body Mass Index, sions to acute hospitals during the last month of life.
Obstruction, Dyspnea, and Exercise capacity [7], ADO
(Age, Dyspnea, and Obstruction) [8, 9] and DOSE (Dysp-
nea, Obstruction, Smoking, and Exercise capacity) [10]. Patients and methods
Nevertheless, strong single factors such as dyspnea The Methods and Results sections are, when possible,
[6] and composites, such as BODE, DOSE, and ADO reported based on the Strengthening the Reporting of
mainly predict survival prospects at the group level, but Observational Studies in Epidemiology (STROBE) crite-
to a lesser degree in individual cases [5, 6, 11, 12]. For this ria [20].
reason, even experienced pulmonary specialists in Swe-
den are reluctant to refer severely ill COPD patients for
Study design
specialized palliative care (SPC) services, although, they
We conducted a descriptive regional registry data study
do not hesitate to refer LC patients with a fairly predict-
using VAL, the Stockholm Region’s central data ware-
able disease trajectory [13]. This phenomenon is not iso-
house. There are separate registers within VAL for out-
lated to Sweden, as similar observations have been made
patient visits to hospitals and hospital stays. Complete
internationally, as shown in a recent systematic review
data were retrieved for patients who died between 2015
[4]. Consequently, COPD is more often seen as a chronic
and 2019, and various aspects of healthcare consump-
disease, even in EOL situations. Therefore, COPD
tion were compared between those who died from either
patients are more likely than LC patients to receive life-
COPD or LC. For each patient included in the analysis,
sustaining measures, such as non-invasive and invasive
data were collected for the 12 months preceding the date
ventilation and cardiopulmonary resuscitation [4, 14, 15].
of death. The data were further analyzed according to
In Sweden, efforts have been made to increase aware-
age, sex, living arrangements (residents in nursing homes
ness about the benefits of a palliative care approach in
versus all others), and SES using Mosaic [21–23]. Mosaic
EOL situations, regardless of the place of care. Therefore,
provides socioeconomic information and allows the
general palliative care is consistently offered in all health-
county council (Stockholm Region) to define and allocate
care facilities, including nursing homes. SPC is offered for
different areas of residence within the county of Stock-
those with complex symptoms and greater needs, mainly
holm to one of three different socioeconomic classes
in the form of advanced palliative home care or hospital
(Mosaic 1–3), mainly based on income and education,
palliative care units. Both types of care are staffed 24 h
but also factoring in other elements, such as cultural
a day, 7 days a week with physicians, registered nurses,
aspects, lifestyle, and living arrangements. The county
physiotherapists, occupational therapists, dieticians,
of Stockholm is divided into 1300 small areas, and each
assistant nurses, and other medical professionals [16].
area is classified as Mosaic 1, 2, or 3. The three groups
Considering the effectiveness of SPC for symptom con-
are approximately equal in size, but Mosaic group 1 is the
trol and generalized support [16] and the documented
most affluent.
positive outcomes for patients with COPD [17], it is
reasonable to assume that both LC and COPD patients
would be candidates for SPC. Butler et al. recently Population
reviewed comparative studies on COPD and LC [4], how- All the patients over the age of 18 who died during the
ever, 10 of the 20 studies reviewed were based on data years 2015 to 2019 with a main diagnosis of COPD (J 44
collected more than 10 years earlier and, among the more in ICD-10) or LC (C34) were included in this study.
Strang et al. BMC Pulm Med (2021) 21:170 Page 3 of 7

Variables Table 1 Characteristics and care utilization for 6479 patients

Access to SPC and the need for emergency room visits who died in COPD (n = 2917) or in LC (n = 3562)
and/or admission to acute hospitals were used as out- Characteristics and care utilization COPD LC p-value
come measures. Age, sex, living arrangements (nursing
homes versus all others), and Mosaic groups as a meas- Deaths 2917 3562
ure of SES were used as explanatory variables. Age, all years (sd) 80.2 (9.1) 72.7 (9.9) < .0001
Women 80.9 (9.0) 72.9 (10.0) < .0001
Selection bias Men 79.3 (9.0) 72.6 (9.9) < .0001
Dropouts: As reporting data to VAL is an obligation for Age groups
each clinic/care unit in the region, the data are complete, 20–39 years, n (%) 0 (0.0) 9 (0.3) < .01
with few missing values. This means that any individual 40–59 years, n (%) 58 (2.0) 313 (8.8) < .0001
who has used public healthcare during a given year is 60–69 years, n (%) 293 (10.0) 859 (24.1) < .0001
included in the VAL databases; this also applies to most 70–79 years, n (%) 928 (31.8) 1522 (42.7) < .0001
forms of private care, as private healthcare suppliers have 80 years or older, n (%) 1638 (56.2) 859 (24.1) < .0001
economic agreements with the regional council. Sex—women (%) 59.0 52.4 < .0001
Nursing home residents: Nursing home residents were Access to SPC (%) 17.7 76.7 < .0001
identified through registrations of medical interven- Care in nursing homes (%) 31.5 9.4 < .0001
tions by physicians, as such care is exclusive to nursing Age, nursing home residents (years) 84 79 < .0001
home residents and has a unique, identifiable code. It Emergency room visits, all (%) 64.4 45.1 < .0001
is unlikely that a nursing home resident does not have Patients with access to SPC (%) 522 391
2
a single registration during their last year of life. If so, Patients without access to SPC (%) 67 661
they were not included in the analysis. Hospital as place of death (%) 39 20 < .0001
Patients with access to SPC (%) 11.82 8.61
Study size Patients without access to SPC (%) 45.22 58.01
This study covered all cohorts, i.e., all deaths (all causes) T-test was used for comparison of age. Chi-2 test was used for comparison of
proportions
during the years 2015–2019. Therefore, no power cal- 1
The difference between those LC patients with and without access to SPC
culations were performed. (vertical comparison in the Table) was significant (p < .0001) both as regards
emergency room visits and hospital as a place of death
2
Statistical methods, missing data The difference between those COPD patients with and without access to SPC
(vertical comparison in the Table) was significant (p < .0001) both as regards
T-tests and chi-2 tests were used to compare the pro- emergency room visits and hospital as a place of death
portions. The few missing data were not substituted.
Univariable logistic regression analyses were performed
and variables with a p-value of < 0.20 were then entered during the last 3 months of life was significantly higher
in multivariable logistic regression models. The SAS for men than women, for persons belonging to higher
version 9.4 was used for statistics. socioeconomic classes (Mosaic 1 and 2), and for
younger patients. The biggest difference was related
Results to diagnosis, patients with LC were much more likely
During the years 2015–2019, there were 2917 deaths to have access to specialized palliative care than those
from COPD and 3562 deaths from LC, for a total of with COPD (Table 2).
6489 patients. The patients who died from COPD
were older (80.2 years) than those with LC (72.7 years) Multivariable logistic regression (both groups)
(p < 0.0001). The proportions of deaths in different age We tested three different models for the likelihood of
groups are shown in Table 1. A larger proportion of being admitted to specialized palliative care services
COPD than LC patients were female, (59% vs. 52.4% during the last three months of life, with similar results.
χ2 = 28.9, p < 0.0001). During the last 3 months of life, Diagnosis (COPD and LC), sex, and SES (Mosaic 1–3)
17.7% of the patients with COPD and 76.7% of the were included in the first model. In the second model,
patients with LC had access to specialized palliative age was entered as a continuous variable, whereas age
care (χ2 = 2235, p < 0.0001) (Table 1). was included as age categories (18–39, 40–59, 60–79,
and ≥ 80 years) in the third model. The third model is
Univariable logistic regression (both groups) shown in Table 2. Sex differences were attenuated in all
In the initial univariable analyses, the likelihood of three models. The odds ratio for women was 0.90 and
being admitted to specialized palliative care services significant in the univariable model; however, it was 1.1,
but not significant, in the multivariable model.
Strang et al. BMC Pulm Med (2021) 21:170 Page 4 of 7

Table 2 Access to SPC. Odds ratio (OR) for different variables

Variable Univariable analysis Multivariable analysis
OR (95% CI) p-value p-value OR (95% CI) p-value p-value

Sex
Women 0.90 (0.82–.995) .04 1.10 (0.97–1.24) .14 (ns)
Men Ref Ref
Socio-economic status
Mosaic group 1 1.32 (1.16–1.50) < .0001 1.33 (1.14–1.56) .0004
Mosaic group 2 1.26 (1.13–1.41) < .0001 1.16 (1.01–1.33) .04 .04
Mosaic group 3 Ref Ref
Age groups
18–39 years 14.64 (1.83–116.94) .01 3.09 (0.38–24.79) .29 (ns)
40–59 years 3.98 (3.15–5.03) < .0001 1.38 (1.05–1.82) .02
60–79 years 2.57 (2.32–2.86) < .0001 1.37 (1.19–1.56) < .0001
80 years or older Ref Ref
Diagnosis
COPD 0.06 (0.06–0.07) < .0001 0.07 (0.06–0.08) < .0001
Lung cancer Ref Ref

Table 3 COPD (only): Access to SPC. Odds ratio (OR) for different Table 4 Lung cancer (only): Access to SPC. Odds ratio (OR) for
variables different variables
Variable Univariable analysis Multivariable analysis Variable Univariable analysis Multivariable analysis
OR (95% CI) p-value OR (95% CI) p-value OR (95% CI) p-value OR (95% CI) p-value

Sex Sex
Women 1.02 (0.84–1.24) .85 (ns) 1.01 (0.83–1.22) .94 (ns) Women 1.14 (0.97–1.32) .11 (ns) 1.14 (0.98–1.34) .10 (ns)
Men Ref Ref Men Ref Ref
Socio-economic status Socio-economic status
Mosaic group 1 1.21 (0.95–1.54) < .13 (ns) 1.19 (0.93–1.52) .16 (ns) Mosaic group 1 1.39 (1.13–1.72) .002 1.40 (1.13–1.73) .002
Mosaic group 2 0.998 (0.80– < .98 (ns) 1.00 (0.80–1.24) .97 (ns) Mosaic group 2 1.27 (1.06–1.51) .009 1.25 (1.05–1.49) .01
1.24) Mosaic group 3 Ref Ref
Mosaic group 3 Ref Ref Age groups
Age groups 18–39 years 3.38 (0.42– .25 (ns) 3.49 (0.43– .24 (ns)
40–59 years 0.36 (0.13–1.01)) .05 (ns) 0.38 (0.13–1.05) .06 (ns) 27.14) 28.16)
60–69 years 0.75 (0.53–1.08) .13 (ns) 0.79 (0.55–1.13) .19 (ns) 40–59 years 1.68 (1.22–2.29) .001 1.78 (1.29–2.45) .0004
70–79 years 1.31 (1.01–1.60) .01 1.31 (1.07–1.61) < .001 60–69 years 1.48 (1.19–1.84) .0004 1.49 (1.19–1.85) .0004
80 years or Ref Ref 70–79 years 1.58 (1.30–1.91) < .0001 1.59 (1.31–1.93) < .0001
older 80 years or Ref Ref
older

COPD (separate analysis)

In a separate univariable analysis of COPD patients,
neither sex nor SES (Mosaic groups) was significantly
associated with the probability of receiving specialized
palliative care, whereas there was an association with
age groups; i.e., patients aged 70–79 years had an odds Lung cancer (separate analysis)
ratio of 1.31, compared to the patients aged ≥ 80 years In contrast to the COPD patients, the LC patients from
(p = 0.01) (Table 3). The findings remained unchanged in more affluent Mosaic groups were more likely to be
the multivariable model that included the same explana- admitted to specialized palliative care, as were patients
tory variables (Table 3). from all the age groups between 40–79 years, compared
with the oldest group (≥ 80 years old) (Table 4).
Strang et al. BMC Pulm Med (2021) 21:170 Page 5 of 7

Care in nursing homes a delay. In the 1970s, women had a much lower risk of
In total, 31.5% of the dying COPD patients, but only 9.4% dying from LC (approximately 10–15 deaths/100,000
of the dying LC patients (χ2 = 498, p < 0.0001), were cared compared to 40–55 deaths/100,000 for men) [24]. Since
for in nursing homes during the EOL. The nursing home then, female deaths have increased, and male deaths have
residents with COPD were older than those with LC, decreased. Today, both groups have incidences between
(84 years vs. 79 years, respectively; p < 0.0001) (Table 1). 30 and 40/100,000 and, as shown in our data, deaths from
LC are now somewhat more common among females, at
Emergency room visits during the last month of life least in the Stockholm region.
During the last month of life, 64.4% and 45.1% of the Equal access to SPC for patients with similar needs,
patients with COPD and LC, respectively, at least one regardless of sex, age, or SES, is a common goal in health-
emergency room visit (χ2 = 240.2, p < 0.0001). The fig- care. However, inequalities still exist. The likelihood of
ure was 67% among the COPD patients without access receiving hospice care differs depending on the health-
to SPC, but the percentage was significantly lower for care system. In a US Medicare study, cancer patients
those with access (52%; χ2 = 48.2, p < 0.0001). The cor- who were younger, male, black, unmarried, and those
responding figures for the LC patients with and without with lower incomes were less likely to receive hospice
access to SPC were 66% and 39%, respectively (χ2 = 183.9, care [25]. In a recent review, higher SES and being female
p < 0.0001) (Table 1). were variables associated with access to palliative care
[26].
Admissions to acute hospitals during the last month of life In this study, we found that access to palliative care was
The figures for admissions to acute hospitals during the not related to sex, in contrast to a recent study on COPD
last month of life were similar to those of emergency and LC by Kendzerska et al., who observed that women
room visits. In total, 62% and 52% of the COPD and LC received more palliative home care than men [19]. More-
patients, respectively, were admitted to acute hospitals over, for the COPD patients in our study, access to SPC
(χ2 = 52.9, p < 0.0001). For the COPD patients, 52% with was not correlated with SES, according to the Mosaic
access to palliative care were admitted to acute hospi- groups, but was so for the patients with LC; this was
tals, whereas the figure was 64% for those without pallia- also the case in the multivariable regression models that
tive care (χ2 = 25.6, p < 0.0001). For the LC patients, the included age and sex. The reason for this remains unclear.
figures were 48% with and 69% without access to SPC, Similarly, age was a less significant factor for the COPD
respectively (χ2 = 116, p < 0.0001) (data not shown). patients, but it was of great significance to those with LC;
i.e., the oldest patients (≥ 80 years) had much less access
Hospitals as the place of death to SPC, which is in line with previous studies [26].
In total, more COPD patients than LC patients died Access to SPC has been associated with better symp-
in acute hospitals (39% vs. 20%, respectively; χ2 = 287, tom control and social support [17]. Obviously, SPC ser-
p < 0.0001). Those patients who had access to SPC died vices are successful in the management of symptoms and
less frequently in acute hospitals than those without acute conditions, which was indirectly reflected in this
access (11.8% vs. 45.2% for COPD patients; χ2 = 198, study; i.e., patients who were enrolled in SPC services
p < 0.0001 and 8.6% vs. 58.0% for LC patients; χ2 = 965, made significantly fewer emergency room visits and were
p < 0.0001) (Table 1). admitted to acute hospitals to a lesser extent during the
last month of life than those not enrolled in SPC services,
Discussion which is in good agreement with the study by Kendzerska
In summary, deaths from COPD and LC were more fre- et al. [19]. Moreover, only 12% of the patients with COPD
quently seen for women than men, and those dying of and 9% of those with LC who had access to SPC died at
COPD were older than those dying of LC. Access to SPC acute hospitals, compared to 45% and 58%, respectively,
was more likely for persons with LC, and it was partially of those who did not have access, implying that SPC
affected by age and SES, although not by sex. Access to unburdens acute hospital care. This is a question of qual-
SPC significantly reduced the need for emergency room ity, as many patients prefer home as their place of death
visits, admissions to acute hospitals, and hospital deaths. [27].
Although smoking was much more common among Finally, neither all cancer nor COPD patients are in
Swedish men in the 1950s and 1960s, tobacco consump- need of specialized palliative care at the end of life,
tion increased among female Swedes in the 1970s, in par- although most of them would probably benefit from an
allel with a reduced use in men. As the time lag between earlier integration of palliative care with the disease-
exposure and a diagnosed illness may be a question specific treatments, which was shown in a seminal ran-
of decades, changes in the illness panorama also have domized study by Temel et al., in newly diagnosed,
Strang et al. BMC Pulm Med (2021) 21:170 Page 6 of 7

metastatic non-small-cell lung cancer (NSCLC) [28]. approved the final version of the manuscript. All the authors meet the criteria
for authorship.
Those who were randomized to palliative care inter-
ventions, in parallel with their cancer-specific treat- Funding
ment showed improved quality of life, improved mood Open access funding provided by Karolinska Institute. Dr. Strang received
grants from the Regional Cancer Centre Stockholm-Gotland (Dnr VKN 2019-
and even an improved overall survival. Patients with far 0070), The Cancer Research Funds of Radiumhemmet (no. 201241), Region
advanced COPD also have a heavy symptom burden [1] Stockholm (ALF Project 20200472), and Stockholm’s Sjukhem Foundation’s
and should be candidates for palliative care interventions Jubilee Fund during the conduct of the study.
earlier than today. There are guidelines to facilitate earlier Availability of data and materials
identification of COPD patients to support earlier recog- The datasets generated, used and analyzed during the current study available
nition of patients nearing the end of life [29]. from the corresponding author on reasonable request.
Access to palliative care for an individual patient
depends both on the availability of services and on actu- Declarations
ally being referred. The latter aspect is important when Ethics approval and consent to participate
comparing lung cancer and COPD patients: while pul- The patients included in this study were deceased and registered in the
monary specialists have routines for referring lung cancer Stockholm Region’s administrative healthcare databases (VAL databases). The
study was conducted in compliance with Good Clinical Practices protocol and
patients, they are more hesitant to refer COPD patients also in compliance with the Declaration of Helsinki principles, when applica-
[13], due to the more unpredictable trajectory in COPD ble (deceased persons with encrypted id-numbers). The working procedure
patients, who more often die from acute events after a and study design were examined by the Swedish Ethical Review Authority
(Etikprövningsmyndigheten, Dnr 2020-02186), who had no ethical objections
stable period.” to the study. They gave permission to use the information and healthcare
data of the deceased and required no informed consent from next of kin, in
Strengths and limitations accordance with Swedish law (2003:460), fourth paragraph (2003:615). The
same law also waives the requirement to obtain informal consent.Consent for
As all healthcare in Sweden, with few exceptions, is publication.Not applicable.
financed by taxes, and reporting to the VAL databases is
mandatory, the data have very few missing values. Competing interests
None of the authors declare any potential conflicts of interest with respect to
A possible limitation to this study is that the diagnosis the research, authorship, and/or publication of this article.
for each patient was not based on the death certificate,
but on the primary diagnosis during the last episode of Author details
1
Department of Oncology‑Pathology, Karolinska Institutet, Regional Cancer
care, which, in the case of LC, was often strengthened Centre in Stockholm, Gotland, Sweden. 2 R & D Department, Stockholms
by a diagnosis of secondary tumors (metastases). There- Sjukhem Foundation, P.O. Box 12230, 102 26 Stockholm, Sweden. 3 Depart-
fore, we cannot exclude the possibility that the immedi- ment of Oncology‑Pathology, Karolinska Institutet, Stockholm, Sweden.
4
Department of Molecular Medicine and Surgery, Karolinska Institutet,
ate cause of death for some of the patients in this study Stockholm, Sweden. 5 Department of Medical Epidemiology and Biostatistics,
may have been something other than COPD or LC, such Karolinska Institutet, Stockholm, Sweden. 6 Department of Surgery, Capio St
as cardiac arrest. Görans Sjukhus, Stockholm, Sweden. 7 Regional Cancer Centre in Stockholm,
Gotland, Sweden.

Conclusions Received: 9 March 2021 Accepted: 7 May 2021

Access to SPC is greater for patients with LC than for
patients with COPD. Access is equal and fair in relation
to sex, but for LC patients, access is associated with SES
and age. SPC reduces the need for emergency visits and References
acute hospital care during the last month of life. 1. Vogelmeier CF, Criner GJ, Martinez FJ, Anzueto A, Barnes PJ, Bourbeau J,
Celli BR, Chen R, Decramer M, Fabbri LM et al. Global strategy for the diag-
nosis, management, and prevention of Chronic Obstructive Lung Disease
2017 Report: GOLD Executive Summary. Eur Respir J 2017, 49(3).
Abbreviations 2. Sin DD, Anthonisen NR, Soriano JB, Agusti AG. Mortality in COPD: role of
EOL: End-of-life; OR: Odds ratio; SES: Socioeconomic status. comorbidities. Eur Respir J. 2006;28(6):1245–57.
3. Statistics on death causes 2019 (Statistik om dödsorsaker). https://​www.​
Acknowledgements socia​lstyr​elsen.​se/​stati​stik-​och-​data/​stati​stik/​stati​stika​mnen/​dodso​rsaker/
We would like to thank Region Stockholm for generously providing us with 4. Butler SJ, Ellerton L, Gershon AS, Goldstein RS, Brooks D. Comparison of
the data for this study and acknowledge The Stockholm Sjukhem Founda- end-of-life care in people with chronic obstructive pulmonary disease or
tion for providing excellent facilities in their research and development unit. lung cancer: a systematic review. Palliat Med. 2020;34(8):1030–43.
We would also like to thank Editage (www.​edita​ge.​com) for English language 5. Almagro P, Yun S, Sangil A, Rodríguez-Carballeira M, Marine M, Landete P,
editing. Soler-Cataluña JJ, Soriano JB, Miravitlles M. Palliative care and prognosis in
COPD: a systematic review with a validation cohort. Int J Chron Obstruct
Authors’ contributions Pulmon Dis. 2017;12:1721–9.
All the authors participated in the planning of the study. TS retrieved the 6. Henoch I, Ekberg-Jansson A, Lofdahl CG, Strang P. Early Predictors of mor-
dataset. PS and TS performed the statistical analysis, and all the authors par- tality in patients with COPD, in relation to respiratory and non-respiratory
ticipated in the interpretation of the results. PS and TS wrote the first draft of causes of death: a national register study. Int J Chron Obstruct Pulmon
the manuscript, and CH, JB, PF, and HA made critical revisions. All the authors Dis. 2020;15:1495–505.
Strang et al. BMC Pulm Med (2021) 21:170 Page 7 of 7

7. Celli BR, Cote CG, Marin JM, Casanova C, MontesdeOca M, Mendez diseases: a population study comparing the dying experience between
RA, PintoPlata V, Cabral HJ. The body-mass index, airflow obstruction, those with chronic obstructive pulmonary disease and lung cancer. Int J
dyspnea, and exercise capacity index in chronic obstructive pulmonary Chron Obstruct Pulmon Dis. 2019;14:1691–701.
disease. N Engl J Med. 2004;350(10):1005–12. 20. Vandenbroucke JP, von Elm E, Altman DG, Gotzsche PC, Mulrow CD,
8. Abu Hussein N, Ter Riet G, Schoenenberger L, Bridevaux PO, Chhajed Pocock SJ, Poole C, Schlesselman JJ, Egger M, Initiative S. Strengthen-
PN, Fitting JW, Geiser T, Jochmann A, Joos Zellweger L, Kohler M, et al. ing the Reporting Of Observational Studies In Epidemiology (STROBE):
The ADO index as a predictor of two-year mortality in general practice- explanation and elaboration. Epidemiology. 2007;18(6):805–35.
based chronic obstructive pulmonary disease cohorts. Respiration. 21. InsightOne: Experia MIS ­MosaicTM Sweden. In. 2015.
2014;88(3):208–14. 22. Dahlén E, Komen J, Jonsson EW, Almqvist C, Kull I, Wettermark B. Elimi-
9. Puhan MA, Garcia-Aymerich J, Frey M, ter Riet G, Anto JM, Agusti AG, nated patient fee and changes in dispensing patterns of asthma medica-
Gomez FP, Rodriguez-Roisin R, Moons KG, Kessels AG, et al. Expansion tion in children: an interrupted time series analysis. Basic Clin Pharmacol
of the prognostic assessment of patients with chronic obstructive Toxicol. 2019;125(4):360–9.
pulmonary disease: the updated BODE index and the ADO index. Lancet. 23. Strang P, Furst P, Schultz T. Excess deaths from COVID-19 correlate with
2009;374(9691):704–11. age and socio-economic status. A database study in the Stockholm
10. Sundh J, Janson C, Lisspers K, Stallberg B, Montgomery S. The dyspnoea, region. Ups J Med Sci. 2020;125(4):297–304.
obstruction, smoking, exacerbation (DOSE) index is predictive of mortal- 24. Cancer in Sweden. Registry data on incidence and deaths 1970–2017
ity in COPD. Prim Care Respir J. 2012;21(3):295–301. (Swe: Cancer i Sverige Registerdata över förekomst och dödlighet
11. Bellou V, Belbasis L, Konstantinidis AK, Tzoulaki I, Evangelou E. Prognos- 1970–2017). https://​www.​cance​rcent​rum.​se/​globa​lasse​ts/​vara-​uppdr​ag/​
tic models for outcome prediction in patients with chronic obstruc- stati​stik/​cancer_​i_​sveri​ge_​regis​terda​ta_​forek​omst_​dodli​ghet_​1970-​2017.​
tive pulmonary disease: systematic review and critical appraisal. BMJ. pdf
2019;367:l5358. 25. Jarosek SL, Shippee TP, Virnig BA. Place of death of individuals with termi-
12. Smith LE, Moore E, Ali I, Smeeth L, Stone P, Quint JK. Prognostic variables nal cancer: new insights from medicare hospice place-of-service codes. J
and scores identifying the end of life in COPD: a systematic review. Int J Am Geriatr Soc. 2016;64(9):1815–22.
Chron Obstruct Pulmon Dis. 2017;12:2239–56. 26. Nelson KE, Wright R, Peeler A, Brockie T, Davidson PM. Sociodemographic
13. Strang S, Ekberg-Jansson A, Strang P, Larsson LO. Palliative care in COPD- disparities in access to hospice and palliative care: an integrative review.
web survey in Sweden highlights the current situation for a vulnerable Am J Hosp Palliat Care 2021:1049909120985419.
group of patients. Ups J Med Sci. 2013;118(3):181–6. 27. Cai J, Zhang L, Guerriere D, Coyte PC. Congruence between preferred and
14. Brown CE, Engelberg RA, Nielsen EL, Curtis JR. Palliative care for patients actual place of death for those in receipt of home-based palliative care. J
dying in the intensive care unit with chronic lung disease compared with Palliat Med. 2020;23(11):1460–7.
metastatic cancer. Ann Am Thorac Soc. 2016;13(5):684–9. 28. Temel JS, Greer JA, Muzikansky A, Gallagher ER, Admane S, Jackson VA,
15. Kuo LC, Chen JH, Lee CH, Tsai CW, Lin CC. End-of-life health care utiliza- Dahlin CM, Blinderman CD, Jacobsen J, Pirl WF, et al. Early palliative care
tion between chronic obstructive pulmonary disease and lung cancer for patients with metastatic non-small-cell lung cancer. N Engl J Med.
patients. J Pain Symptom Manag. 2019;57(5):933–43. 2010;363(8):733–42.
16. Nordström M, Strang P. High degree of satisfaction with the support 29. The GSF Prognostic Indicator Guidance. The National GSF Centre’s guid-
given by multidisciplinary palliative home care teams in the county of ance for clinicians to support earlier recognition of patients nearing the
Stockholm. J Palliat Care. 2018;33(2):109–14. end of life. https://​www.​golds​tanda​rdsfr​amewo​rk.​org.​uk/​cd-​conte​nt/​
17. Henoch I, Strang P, Löfdahl CG, Ekberg-Jansson A. Equal palliative care uploa​ds/​files/​Gener​al%​20Fil​es/​Progn​ostic%​20Ind​icator%​20Gui​dance%​
for patients with COPD? A nationwide register study. Ups J Med Sci. 20Oct​ober%​202011.​pdf
2019;124(2):140–7.
18. Faes K, Cohen J, Annemans L. Resource use during the last six months
of life among COPD patients: a population-level study. J Pain Symptom Publisher’s Note
Manag. 2018;56(3):318-326.e317. Springer Nature remains neutral with regard to jurisdictional claims in pub-
19. Kendzerska T, Nickerson JW, Hsu AT, Gershon AS, Talarico R, Mulpuru S, lished maps and institutional affiliations.
Pakhale S, Tanuseputro P. End-of-life care in individuals with respiratory

Ready to submit your research ? Choose BMC and benefit from:

• fast, convenient online submission

• thorough peer review by experienced researchers in your field
• rapid publication on acceptance
• support for research data, including large and complex data types
• gold Open Access which fosters wider collaboration and increased citations
• maximum visibility for your research: over 100M website views per year

At BMC, research is always in progress.

Learn more biomedcentral.com/submissions