[ Original Research COPD ]

Determinants of Smoking Cessation in

Patients With COPD Treated in the
Outpatient Setting
Sandra S. Tøttenborg, MSc; Reimar W. Thomsen, DMSc; Søren P. Johnsen, DMSc; Henrik Nielsen, MSc;
and Peter Lange, DMSc

BACKGROUND: The beneficial effects of smoking cessation on the progression of COPD are
well established. Nevertheless, many patients with COPD continue to smoke.
METHODS: In this nationwide hospital-based prospective follow-up study, we examined rates
of smoking cessation and clinical and sociodemographic determinants of smoking cessation
in 3,233 patients with COPD who smoked on outpatient contact during 2008 to 2012. Using
multivariate Cox regression, we calculated hazard ratios (HRs) of quitting.
RESULTS: Within 1 and 5 years from first outpatient contact, the probability of quitting was
19% and 45%, respectively. In adjusted analyses, patients were less likely to quit if they were
younger, with an HR of 0.84 (95% CI, 0.71-0.99) for patients aged 50 to 69 years and 0.53
(95% CI, 0.37-0.76) for patients aged 30 to 49, compared with those aged 70 years or older,
who had lower income (HR, 0.79; 95% CI, 0.67-0.94), lived alone (HR, 0.75; 95% CI,
0.64-0.88), were unemployed (HR, 0.70; 95% CI, 0.54-0.90), had milder COPD with an HR
of 0.67 (95% CI, 0.53-0.84) for Global Initiative for Chronic Obstructive Lung Disease
(GOLD) A and 0.61 (95% CI, 0.47-0.80) for GOLD B compared with GOLD D, had Medical
Research Council (MRC) dyspnea scale score < 4 (HR 0.80, 95% CI 0.68-0.95), or no history
of exacerbations treated on an outpatient basis. (HR, 0.80; 95% CI, 0.68-0.93).
CONCLUSIONS: These findings reinforce that young and socioeconomically disadvantaged patients
have more difficulties achieving timely smoking cessation. A novel finding is that patients with
milder COPD are less likely to quit. The findings suggest a need for interventional studies focusing
on these subgroups to ensure abstinence to halt disease progression.
CHEST 2016; 150(3):554-562

KEY WORDS: COPD; GOLD; predictors; smoking cessation; socioeconomic status

ABBREVIATIONS: CVD = cardiovascular disease; DrCOPD = Danish
Register of COPD; GOLD = Global Initiative for Chronic Obstructive
Lung Disease; GOLD A-D = combined assessment of COPD based on
symptoms, airflow limitation, and exacerbations; GOLD 1-4 = COPD
staging based on spirometry; ICD-10 = International Classification of
Diseases, tenth revision; MRC = Medical Research Council; NPR =
National Patient Registry
AFFILIATIONS: From the Department of Public Health (Ms
Tøttenborg and Dr Lange), Section of Social Medicine, University of
Copenhagen, Copenhagen, Denmark; Department of Clinical Epide-
miology (Drs Thomsen and Johnsen and Mr Nielsen), Aarhus Uni-
versity Hospital, Aarhus, Denmark; and Respiratory Section (Dr
Lange), Hvidovre Hospital, Hvidovre, Denmark.
FUNDING/SUPPORT: This work was supported by grants from the
Danish Lung Association, the University of Copenhagen, and Boeh-
ringer Ingelheim. The work of R. W. T. and S. P. J. was funded by the
Program for Clinical Research Infrastructure (PROCRIN) established
by the Lundbeck Foundation and the Novo Nordisk Foundation and
administered by the Danish Regions.
CORRESPONDENCE TO: Sandra S. Tøttenborg, MSc, Section of Social
Medicine, Department of Public Health, University of Copenhagen,
Øster Farimagsgade 5, Postboks 2099, 1014 Copenhagen K, Denmark;
e-mail: sato@sund.ku.dk
Copyright Ó 2016 American College of Chest Physicians. Published by
Elsevier Inc. All rights reserved.
DOI: http://dx.doi.org/10.1016/j.chest.2016.05.020

554 Original Research [ 150#3 CHEST SEPTEMBER 2016 ]

COPD is a common disease with chronic symptoms
often complicated by exacerbations and hospitalizations,1
resulting in increased mortality and reduced life
expectancy.2 Smoking cessation improves respiratory
symptoms and reduces the decline in FEV1 in all COPD
stages,3,4 lowers the risk of exacerbations5 and hospital
admissions,6 and increases survival.4,7 Achieving abstinence
is thus considered the first step in COPD management.8
Nevertheless, many patients continue to smoke.9-12 A
Danish nationwide audit indicated that > 30% of patients
with COPD followed in outpatient clinics smoke.13
Data on predictors of successful cessation are sparse and
rely largely on general population studies. Older age,
higher income, and cardiac comorbidities were
associated with quitting,14 whereas high tobacco
consumption15 and nicotine dependence,16 depression,17
and low socioeconomic status14 were associated with
low quitting rates. Some of these findings were
Methods
In Denmark, stable mild and moderate COPD is generally managed in
primary care, whereas patients with more severe COPD with frequent
exacerbations or rapidly progressing disease are referred to national
specialized pulmonary outpatient clinics. In 2008, the Danish
Register of COPD (DrCOPD) was initiated nationwide to
continuously monitor these outpatients. Details of the DrCOPD have
been described elsewhere.13,19 Briefly, all contacts between patients
with COPD and outpatient clinics are registered in the National
Patient Register (NPR) for reimbursement purposes. With the
DrCOPD, the clinics commit to register a set of clinical variables for
each patient alongside the contact date at least once yearly. Contact
dates and variables are uploaded to the DrCOPD, which consists of
a varying number of informative contacts (including smoking status
and other clinical data) and uninformative contacts for each patient.
In this study, COPD was defined by primary diagnoses according to
the Danish International Classification of Diseases, Tenth Revision
(ICD-10): COPD (J44.X), or respiratory failure, with COPD as
secondary diagnosis (J96.X þ J44.X) and age $ 30 years. During
2008 to 2012, 23,741 individuals with a first-ever outpatient contact
for COPD were enrolled, and a subsample of smokers was identified
among these patients. To detect a change in smoking patterns,
patients with at least two registrations of smoking status were
included for analysis (Fig 1). The study was approved by the Danish
Data Protection Agency (record 2012-41-0438), the Danish National
Indicator Project, Danish Regions, and the Danish Ministry of Health.
Outcome
Information on smoking cessation was based on patient self-reporting.
Quitting was defined as patients having reported abstaining from
tobacco for 6 months before the outpatient visit.
Clinical and Sociodemographic Variables
FEV1%pred, BMI, and dyspnea (Medical Research Council [MRC]
scale) was obtained from the DrCOPD. Admissions within the
previous year were identified searching the NPR for hospital
admissions lasting $ 1 day with ICD-10 codes COPD (DJ44) as the
journal.publications.chestnet.org
reproduced in patients with COPD.7,18 A Finnish study
of 739 smokers found an inverse association between
psychiatric conditions and quitting, whereas the chance
increased with age.7 An American cross-sectional study
of 89,337 veterans showed that ex-smokers were older
and had more cardiac comorbidities and better mental
health than did current smokers.18 These studies,
however, provided no measures of COPD, severity
such as degree of dyspnea, exacerbation frequency and
Global Initiative for Chronic Obstructive Lung Disease
(GOLD) A-D grouping, or insights into the role of
socioeconomic characteristics or smoking cessation
advice for successful cessation.
Identifying determinants of smoking cessation is crucial
for designing effective interventions. We therefore aimed
to determine characteristics associated with smoking
cessation in a nationwide cohort of patients from all
Danish outpatient clinics.
primary diagnosis or respiratory failure (DJ96) or pneumonia (DJ13-
DJ18) as the primary diagnosis with COPD as the secondary
diagnosis. Exacerbations treated on an outpatient basis were
identified searching the Danish Prescription Registry for
prednisolone claims the year before baseline. To be considered
independent exacerbations, claims should be > 28 days apart.5,20
GOLD 1-4 staging was used to classify severity of airflow
limitation,21 and GOLD A-D classification described COPD
severity.21 Depression was defined as any claim for an antidepressant
(excluding bupropion) during the 2 years before baseline.
Comorbidities were identified searching the NPR for diagnoses
according to the Charlson comorbidity index22 during the 10 years
before baseline and categorized into cardiovascular disease (CVD),
cancer, and other diseases (e-Table 1).
Statistics Denmark provided data on age, sex, ethnicity, cohabitation
status, and education (primary/lower secondary [elementary school],
upper secondary [high school], and tertiary [university]). Mean
family-equalized disposable income for 2 years before baseline was
used to group subjects as having more or less than the 2009 national
mean of approximately $26,500.23 Patients younger than the official
retirement age of 65 years who were unemployed, receivers of social
benefits or any kind of pension were considered unemployed.
Statistics
Analyses were performed using STATA 14 (StataCorp LP). Patients
were followed from first contact as a smoker (baseline) until first
contact as an ex-smoker (event) or last contact as a smoker
(censored). We applied the last observation carried forward method
for intermediate uninformative contacts (for a patient with three
contacts, ie, baseline (smoker), intermediate (unknown), last
(smoker), we assumed the intermediate value “smoker”). Using the
Kaplan-Meier estimator, we calculated the percentage of events
during follow-up according to baseline factors, taking censoring of
the data into account. Thus, percentages are 100% minus the
Kaplan-Meier estimate of being event free. The log-rank test was
used for univariate comparisons of events. Effects of characteristics
on the relative probability of quitting smoking were estimated using
555
 DrCOPD
COPD patients with a first-ever visit with a COPD
outpatient clinic in the period 2008 through 2012
N = 23,741
Patients with a least one registration of smoking
status during course of outpatient visits
N = 18,206
Patients who report to be a current smoker at
first informative contact
N = 6,932
Patients with at least one follow-up registration
of smoking status
N = 3,233
Figure 1 – Flow diagram of individuals in the DrCOPD showing derivation of subgroup of 3,233 individuals with at least two registrations of smoking
status included in this study. DrCOPD ¼ Danish Register of COPD.
Cox regression. Proportionality was examined using scaled Schoenfeld
residuals. Confounders were identified drawing directed acyclic
Results
In this cohort of 18,206 outpatients with COPD,
6,932 (38%) smoked at baseline. Smoking prevalence
according to GOLD 1-4 was 38%, 40%, 37%, and 37%,
respectively. Three thousand two hundred thirty-three
smokers had $ 1 follow-up contact with smoking
information included in the analyses. During 3,569
person-years, we observed 951 cases of smoking
cessation. The last observed exit was at 4.9 years, with a
median follow-up of 320 days (p25 ¼ 154; p75 ¼ 576).
During 1 and 5 years from first contact, the probability
of quitting smoking was 19% and 45%, respectively.
Table 1 presents baseline characteristics and
corresponding cases of smoking cessation. The patients
were characterized by advanced disease and a high
556 Original Research
No information on smoking status
registered during 2008-2012
N = 5,535
Non-smokers (never smokers and ex-smokers)
at first registration of smoking status
N = 11,274
Patients with no informative follow-up registration
of smoking status
N = 3,699
At least one follow-up
No follow-up contact with missing
N = 2,356 information on smoking
N = 1,343
graphs.24 Effects of employment were computed for patients < 65
years.
degree of dyspnea, with 33% having experienced $ 1
exacerbation during the previous year. The prevalence of
depression was 31%. More than half of patients had a
lower level of education and lived alone, and 62% were
unemployed. Patients with no informative follow-up
contacts were younger, had milder COPD, and were
more likely to live alone.
Effects of clinical factors are presented in Table 2. Patients
appeared less likely to quit smoking if they had a history
of cancer (hazard ratio [HR], 0.79; 95% CI, 0.58-1.07)
or depression (HR, 0.91; 95% CI, 0.77-1.07) or if their
physician refrained from encouraging smoking cessation
(HR, 0.82; 95% CI, 0.66-1.03) compared with patients
without a history of cancer or depression as well as those
encouraged to quit smoking. However, these associations
[ 150#3 CHEST SEPTEMBER 2016 ]
TABLE 1 ] Baseline Characteristics for 3,233 Smokers With COPD With at Least One Informative Contact During
Follow-Up Compared With Characteristics in 3,699 Smokers With No Informative Follow-Up Contact:
5-Year Quitting Rates According to Clinical and Socioeconomic Factors Among 3,233 Smokers Included
for Analyses
More Than One No Informative
Informative Contact Contacts During
During Follow-Up, Follow-Up, 5-Year Quitting
n ¼ 3,233 n ¼ 3,699 Rates
Variables No. (%)/Median (IQR) No. (%)/Median (IQR) P Value Quitters (KM %)a P Valuea
Overall probability of quitting
First y 491 (19)
At 5 y 951 (45)
c
Sex, No. (%)
Male 1,544 (48) 1,841 (50) .089 422 (43) .194
Female 1,686 (52) 1,852 (50) 528 (48)
Age, median (IQR) or No. (%) 66 (16) 65 (17) .447
$70 y 1,123 (35) 1,313 (36) .002 338 (46) .003
50-69 y 1,839 (57) 1,990 (54) 553 (46)
30-49 y 271 (8) 396 (11) 60 (36)
Dyspnea (MRC score)c
$4 824 (27) 807 (24) .001 283 (51) .000
<4 2,195 (72) 2,596 (74) 599 (43)
BMI, median (IQR) 24 (7) 24 (7) .000
FEV1 %, median (IQR) 47 (25) 55 (27) .000
Spirometric GOLD 1-4, No. (%)c
Very severe 459 (15) 347 (10) .000 202 (61) .000
Severe 1,291 (41) 1,056 (29) 404 (48)
Moderate 1,249 (40) 1,852 (52) 292 (38)
Mild 141 (4) 335 (9) 26 (31)
GOLD A-D, No. (%)c
D 1,440 (48) 1,246 (37) .000 491 (51) .000
C 650 (22) 717 (21) 194 (46)
B 383 (13) 463 (14) 73 (32)
A 517 (17) 931 (28) 113 (36)
Hospital admissions, No. (%)
None 2,292 (71) 2,679 (72) .158 661 (45) .076
$1 941 (29) 1,020 (28) 290 (47)
Exacerbations, No. (%)
None 2,170 (67) 2,577 (70) .023 579 (42) .000
$1 1,063 (33) 1,122 (30) 372 (52)
Smoking cessation advice, No. (%)
No 391 (12) 330 (9) .000 116 (46) .020
Yes 2,842 (88) 3,369 (91) 835 (45)
CVD diagnosis, No. (%)
None 2,357 (73) 2,597 (70) .013 710 (46) .759
$1 876 (27) 1,102 (30) 241 (43)
Cancer diagnosis, No. (%)
None 3,004 (93) 3,419 (92) .439 897 (46) .103
$1 229 (7) 280 (8) 54 (38)

(Continued)

journal.publications.chestnet.org 557
TABLE 1 ] (Continued)
More Than One No Informative
Informative Contact Contacts During
During Follow-Up, Follow-Up, 5-Year Quitting
n ¼ 3,233 n ¼ 3,699 Rates
Variables No. (%)/Median (IQR) No. (%)/Median (IQR) P Value Quitters (KM %)a P Valuea
Other diagnosis, No. (%)
None 3,008 (93) 3,349 (91) .000 875 (45) .020
$1 225 (7) 350 (9) 76 (51)
Depression, No. (%)
No 2,239 (69) 2,497 (68) .118 684 (47) .189
Yes 994 (31) 1,202 (33) 267 (42)
Ethnicity, No. (%)c
Danish 3,110 (96) 3,534 (96) .213 914 (45) .713
Immigrant 120 (4) 159 (4) 36 (46)
Cohabitation status, No. (%)c
Cohabitating 1,607 (50) 1,738 (47) .028 543 (51) .001
Living alone 1,623 (50) 1,952 (53) 408 (40)
Educational attainment, No. (%)c
Elementary school 1,765 (56) 1,957 (55) .387 520 (46) .263
High school 1,161 (37) 1,345 (38) 344 (46)
University 215 (7) 269 (8) 63 (45)
Income, No. (%)c 158 (67) 160 (67) .092
High 1,002 (36) 1,120 (36) .728 327 (49) .002
Low 1,797 (64) 1,971 (64) 501 (44)
Employed (subjects < 65 y),
No. (%)b
Employed 623 (38) 665 (35) .066 197 (48) .001
Unemployed 997 (62) 1,211 (65) 259 (41)

CVD ¼ cardiovascular disease; IQR ¼ interquartile range; GOLD ¼ Global Initiative for Chronic Obstructive Lung Disease; KM ¼ Kaplan-Meier.
a
Percentages are Kaplan-Meier estimates, and P values are based on log-rank test for time to event.
b
Only patients younger than 65 years of age.
c
Missing data: sex (n ¼ 3), dyspnea (n ¼ 214), GOLD 1-4 (n ¼ 93), GOLD A-D (n ¼ 243), ethnicity (n ¼ 3), cohabitation status (n ¼ 3), education (n ¼ 92),
and income (n ¼ 434).

did not achieve statistical significance. Patients with
“other” comorbidities had a higher probability of quitting
(HR, 1.31; 95% CI, 1.00-1.72) than other patients.
Patients were less likely to quit if they had milder COPD
with an HR of 0.67 (95% CI, 0.53-0.84) for GOLD A, an
HR of 0.61 (95% CI, 0.47-0.80) for GOLD B, and an HR
of 0.86 (95% CI, 0.72-1.04) for GOLD C compared with
GOLD D. In analyses of GOLD A-D components, we
observed a lower chance of quitting among patients with
moderate (HR, 0.53; 95% CI, 0.42-0.65) and severe (HR,
0.70; 95% CI, 0.57-0.85) COPD relative to very severe
COPD, among patients with a score on the MRC scale
of < 4 (HR, 0.80; 95% CI, 0.68-0.95) compared with a
score on the MRC scale of $ 4, and in those with no
history of exacerbations (HR, 0.80; 95% CI, 0.68-0.93)
relative to those with prior exacerbations (Table 3).
The probability of quitting smoking was lower for
younger patients with an HR of 0.84 (95% CI, 0.71-0.99),
for those aged 50 to 69 years and 0.53 (95% CI, 0.37-0.76),
and for those aged 30 to 49 years, compared with those
aged $ 70 years. The probability was also lower for
patients who had lower income (HR, 0.79; 95% CI,
0.67-0.94) or lived alone (HR, 0.75; 95% CI, 0.64-0.88)
(Table 4). Relative to those who were employed,
unemployed patients were less likely to quit smoking
(HR, 0.70; 95% CI, 0.54-0.90) (Table 5). This association
was more pronounced for women (HR, 0.56; 95% CI,
0.40-0.79) than for men (HR, 0.92; 95% CI, 0.64-1.34).
Discussion
This nationwide prospective study is, to our
knowledge, the first to analyze the associations

558 Original Research [ 150#3 CHEST SEPTEMBER 2016 ]

TABLE 2 ] Probability of Smoking Cessation According TABLE 3 ] Probability of Smoking Cessation According
to Clinical Characteristics to COPD Severity
Crude Adjusted Crude Adjusteda
Variable HR (95% CI) HR (95% CI) Variable HR (95% CI) HR (95% CI)
GOLD A-Da Spirometric
D 1 (Ref) 1 (Ref) GOLD 1-4

C 0.91 (0.77-1.08) 0.86 (0.72-1.04) Very severe 1 (Ref) 1 (Ref)

B 0.62 (0.49-0.80) 0.61 (0.47-0.80) Severe 0.67 (0.57-0.79) 0.70 (0.57-0.85)

A 0.73 (0.60-0.90) 0.67 (0.53-0.84) Moderate 0.54 (0.45-0.65) 0.53 (0.42-0.65)

CVD Mild 0.56 (0.37-0.84) 0.73 (0.46-1.14)

diagnosisb Dyspnea MRC
No 1 (Ref) 1 (Ref) $4 1 (Ref) 1 (Ref)
Yes 0.98 (0.84-1.13) 0.89 (0.75-1.06) <4 0.74 (0.64-0.85) 0.80 (0.68-0.95)
Cancer Hospital
diagnosisb admissions
No 1 (Ref) 1 (Ref) Yes 1 (Ref) 1 (Ref)
Yes 0.80 (0.60-1.05) 0.79 (0.58-1.07) No 0.88 (0.77-1.01) 0.93 (0.79-1.09)
Other Exacerbations
diagnosisb Yes 1 (Ref) 1 (Ref)
No 1 (Ref) 1 (Ref) No 0.79 (0.69-0.90) 0.80 (0.68-0.93)
Yes 1.32 (1.04-1.67) 1.31 (1.00-1.72)
See Table 1 and 2 legends for expansion of abbreviations.
Depressionb a
Adjusted for age, sex, ethnicity, cohabitation, education, income, CVD,
No 1 (Ref) 1 (Ref) cancer, other diagnoses, depression, and smoking cessation encourage-
Yes 0.91 (0.79-1.05) 0.91 (0.77-1.07) ment; mutually adjusted for GOLD 1-4, MRC, admissions, and
exacerbations.
Smoking
cessation
adviceb Previous studies have assessed the relationship with
Yes 1 (Ref) 1 (Ref) FEV1 but with equivocal results: One study in the
No 0.79 (0.65-0.96) 0.82 (0.66-1.03) general population found an association between
HR ¼ hazard ratio; Ref ¼ reference. See Table 1 legend for expansion of
reduced FEV1 and smoking cessation,15 but this was not
other abbreviations. reproduced in a study of patients with COPD.7 We
a
Adjusted for age, sex, ethnicity, cohabitation, education, income, CVD, observed that patients without a history of exacerbations
cancer, other diagnoses, depression, and smoking cessation
encouragement.
treated on an outpatient basis and a lower degree of
b
Adjusted for age, sex, ethnicity, cohabitation, education, income, and dyspnea were less likely to achieve abstinence. Activity
GOLD A-D; mutually adjusted for CVD, cancer, other diagnoses, depres- limitation due to COPD has previously been associated
sion, and smoking cessation encouragement.
with a higher chance of quitting smoking.25 Thus,
deteriorating health seems to increase the motivation to
between sociodemographic and COPD severity factors quit, which may be reinforced by physicians to achieve
and smoking cessation in a population of outpatients higher quitting rates. Consistent with previous studies in
with COPD. During the 5 years after starting specialized the general population and other patient cohorts, we
outpatient care, the probability of quitting smoking was found a significant association between younger age and
45%. Considering that smoking cessation is the most difficulties with quitting smoking,7,18,25 which persisted
important intervention affecting long-term prognosis in after adjustment for disease severity. Older subjects may
COPD, this observed cessation rate seem rather low. have made more attempts at quitting smoking, which in
We observed a significant relationship between lower previous studies has been shown to predict successful
COPD severity as defined by GOLD A-D group and a cessation.14,26 Schiller and Ni25 found that although
lower chance of quitting smoking. To our knowledge, younger smokers with COPD were more likely to
this is the first time GOLD A-D has been related to attempt quitting, paradoxically their older counterparts
smoking cessation. We also found a significant were more likely to succeed. Depression has previously
relationship between lower COPD severity as defined by been associated with lower quitting rates among patients
GOLD 1-4 and a lower chance of quitting smoking. with COPD.7,18 Kupiainen et al7 found that patients with

journal.publications.chestnet.org 559
TABLE 4 ] Probability of Smoking Cessation According whereas two studies found no association.15,28 We found
to Sociodemographic Factors that patients living alone had more difficulties quitting
Crude Adjusteda smoking than did those living with someone. One
HR (95% CI) HR (95% CI) previous study did not observe a relationship between
Sex cohabitation status and smoking cessation,29 whereas a
Male 1 (Ref) 1 (Ref) Romanian study found that patients cohabitating with a
Female 1.09 (0.96-1.24) 1.09 (0.93-1.27) smoker were less likely to quit than were patients who
Age
lived alone or with a nonsmoker.30 We found that
$ 70 y 1 (Ref) 1 (Ref)
unemployment was significantly associated with
continued smoking, especially in women. Morris et al31
50-69 y 0.90 (0.79-1.03) 0.84 (0.71-0.99)
showed that men who attributed their unemployment to
30-49 y 0.62 (0.47-0.82) 0.53 (0.37-0.76)
illness had significantly higher rates of smoking
Ethnicity
cessation than did both continuously employed men and
Danish 1 (Ref) 1 (Ref)
men who did not attribute their nonemployment to
Immigrant 1.06 (0.76-1.49) 1.17 (0.79-1.74)
illness.31 In our cohort, 53% of the men younger than 65
Cohabitation
years were on a disability pension, which is granted in
status
cases of permanently reduced working capacity. The
Cohabitating 1 (Ref) 1 (Ref)
effects of comorbidities on quitting smoking were
Living alone 0.80 (0.70-0.90) 0.75 (0.64-0.88)
equivocal. Although we observed no negative association
Educational
attainment
with CVD, subjects with cancer had a slightly reduced
Elementary 1 (Ref) 1 (Ref)
probability of quitting smoking. Accordingly, in the
school general population the presence of cardiac comorbidities
High school 1.11 (0.97-1.28) 1.13 (0.97-1.33) has been shown to increase the probability of quitting
University 1.12 (0.86-1.45) 0.96 (0.70-1.32) smoking,14 whereas Kupiainen et al7 found no
Income in association with either CVD or cancer in a study of
thousands patients with COPD. A possible explanation for the lack
High 1 (Ref) 1 (Ref) of association could be that patients continuing to
Low 0.81 (0.70-0.93) 0.79 (0.67-0.94) smoke after a cancer diagnosis are more recalcitrant
smokers. Lastly, we observed that patients who were not
See Table 1 and 2 legends for expansion of abbreviations.
a
Adjusted for GOLD A-D, CVD, cancer, other diagnoses, depression,
advised to stop smoking were less likely to do so. This is
smoking cessation encouragement; mutually adjusted for age, sex, consistent with previous findings.25,32 For instance,
ethnicity, cohabitation, education, and income. Tønnesen et al32 saw a doubling in the 1-year quitting
rate after minimal motivational intervention in patients
COPD with psychiatric disorders were 83% more attending a lung clinic.
likely to fail at quitting smoking. In comparison, we
observed only a modest tendency toward reduced Strengths and Limitations
smoking cessation in patients with depression. Using The prospective design ensured temporality between
antidepressant claims may have underestimated the baseline characteristics and smoking cessation, whereas
true prevalence of depression. However, the prevalence the large sample size provided good variance in
of 31% in this cohort equals that of other cohorts COPD severity. Although we were able to adjust for
of smokers with related pulmonary diagnoses.27 important factors associated with smoking cessation,
Alternatively, antidepressants may have improved there may be residual confounding from the type of
depressive symptoms to a point that they no longer tobacco smoked, as cigar smokers have higher cessation
interfered with quitting smoking. We observed low rates than cigarette and pipe smokers.29 Also, we had
income to be associated with a lower chance of quitting no information on the use of nicotine replacement
smoking. Lower disposable income may hinder access to therapy or participation in smoking interventions and
nicotine replacement therapy, which is primarily sold no detailed information on the extent of cessation
over the counter and without subsidization. In general encouragement delivered by the doctor. From the source
population studies, the effect of income has been population, 3,699 smokers were excluded because of
inconsistent: One study showed that higher income missing follow-up data on smoking. We observed that
was positively associated with successful cessation,26 smokers with severe COPD were more likely to have

560 Original Research [ 150#3 CHEST SEPTEMBER 2016 ]

TABLE 5 ] Probability of Smoking Cessation During 5 Years of Follow-Up Among 1,671 Patients < 65 Years With
COPD According to Employment Status
Crude Model 1a Model 2b
HR (95% CI) HR (95% CI) HR (95% CI) P for Interactionc
All
Employed 1 (Ref) 1 (Ref) 1 (Ref)
Unemployed 0.74 (0.61-0.89) 0.74 (0.58-0.93) 0.70 (0.54-0.90)
Men 1 (Ref) 1 (Ref) 1 (Ref) .220
Unemployed 0.79 (0.60-1.05) 0.95 (0.66-1.35) 0.92 (0.64-1.34)
Women 1 (Ref) 1 (Ref) 1 (Ref)
Unemployed 0.68 (0.53-0.87) 0.61 (0.44-0.84) 0.56 (0.40-0.79)

See Table 2 legend for expansion of abbreviations.

a
Adjusted for age sex, smoking cessation advice, cohabitation status, education, and income.
b
Further adjusted for comorbidities, depression, and GOLD A-D.
c
P value for interaction between sex and employment status.

follow-up on smoking than were smokers with less
severe COPD, likely because people with less severe
COPD were transferred back to primary care to a greater
extent. However, selection bias due to loss to follow-up
would impact the study results only if the probability of
capturing an outcome event depended on both exposure
and outcome. This might occur if patients with
advanced COPD who stopped smoking were selectively
more likely to return to the clinic than were patients
who quit smoking and had less severe COPD (eg, if
more severely ill patients who quit smoking tended to
live longer than patients who did not quit and therefore
had more follow-up contacts, whereas patients who
were less severely ill and quit smoking were stabilized
and discharged to primary care and lost to follow-up).
This would lead to an overestimation of the association
between milder COPD and smoking cessation.
Another limitation is the possible misclassification of
smoking status, as this was not biologically validated.
Stelmach et al33 compared self-reported smoking with
levels of urinary cotinine (a biomarker of tobacco
exposure) levels in outpatients with COPD and found
that self-reported smoking prevalence was significantly
lower compared with smoking prevalence validated
by cotinine measurement. Consequently, we may
have overestimated cessation during follow-up.
Further, changes in smoking could only be assessed
on contact, which may have caused us to overestimate
time to smoking cessation. Finally, the study involved
outpatients only, and extrapolation of results to
patients followed by general practitioners and
nonhospital-based pulmonologists should be done
with caution.
Implications
Younger, less severely ill, and socioeconomically
disadvantaged patients with COPD are less likely to achieve
smoking cessation, underscoring the need for studies
focusing on these subgroups to limit the detrimental effects
of continued smoking.

journal.publications.chestnet.org 561
Acknowledgments
Author contributions: P. L. and S. S. T are
the guarantors of the paper. S. S. T., R. W. T.,
S. P. J., H. N., and P. L. are responsible for the
study concept, design, data acquisition and
critical manuscript revision. S. S. T., P. L.,
R. W. T., and S. P. J. are responsible for the
analysis and interpretation of data. S. S. T. is
responsible for the first draft of the
manuscript and statistical analysis.
Financial/nonfinancial disclosures: None
declared.
Role of sponsors: The sponsors had no role
in the design of the study, the collection and
analysis of the data, or the preparation of the
manuscript.
Additional information: The e-Table can be
found in the Supplemental Materials section
of the online article.
References
1. Decramer M, Janssens W, Miravitlles M.
Chronic obstructive pulmonary disease.
Lancet. 2012;379(9823):1341-1351.
2. Shavelle RM, Paculdo DR, Kush SJ, et al.
Life expectancy and years of life lost in
chronic obstructive pulmonary disease:
findings from the NHANES III Follow-up
Study. Int J Chron Obstruct Pulmon Dis.
2009;4:137-148.
3. Godtfredsen NS, Lam TH, Hansel TT,
et al. COPD-related morbidity and
mortality after smoking cessation: status
of the evidence. Eur Respir J. 2008;32(4):
844-853.
4. Kanner RE, Connett JE, Williams DE,
et al. Effects of randomized assignment to
a smoking cessation intervention and
changes in smoking habits on respiratory
symptoms in smokers with early chronic
obstructive pulmonary disease: the Lung
Health Study. Am J Med. 1999;106(4):
410-416.
5. Au DH, Bryson CL, Chien JW, et al. The
effects of smoking cessation on the risk of
chronic obstructive pulmonary disease
exacerbations. J Gen Intern Med.
2009;24(4):457-463.
6. Godtfredsen NS, Vestbo J, Osler M, et al.
Risk of hospital admission for COPD
following smoking cessation and
reduction: a Danish population study.
Thorax. 2002;57(11):967-972.
7. Kupiainen H, Kinnula VL, Lindqvist A,
et al. Successful smoking cessation in
COPD: association with comorbidities
and mortality. Pulm Med. 2012;2012:
725024.
8. Global Initiative for Chronic Obstructive
Lung Disease. Global strategy for the
diagnosis, management, and prevention
562 Original Research
of COPD: updated 2015. http://www.
goldcopd.it/materiale/2015/GOLD_
Pocket_2015.pdf. Accessed July 25, 2016.
9. Calverley PM, Anderson JA, Celli B, et al.
Salmeterol and fluticasone propionate and
survival in chronic obstructive pulmonary
disease. N Engl J Med. 2007;356(8):
775-789.
10. Tashkin DP, Celli B, Senn S, et al. A 4-year
trial of tiotropium in chronic obstructive
pulmonary disease. N Engl J Med.
2008;359(15):1543-1554.
11. Magnussen H, Disse B, Rodriguez-
Roisin R, et al. Withdrawal of inhaled
glucocorticoids and exacerbations of
COPD. N Engl J Med. 2014;371(14):
1285-1294.
12. Martinez FJ, Calverley PM, Goehring UM,
et al. Effect of roflumilast on exacerbations
in patients with severe chronic obstructive
pulmonary disease uncontrolled by
combination therapy (REACT): a
multicentre randomised controlled trial.
Lancet. 2015;385(9971):857-866.
13. Tottenborg SS, Thomsen RW, Nielsen H,
et al. Improving quality of care among
COPD outpatients in Denmark 2008-
2011. Clin Respir J. 2013;7(4):319-327.
14. Vangeli E, Stapleton J, Smit ES, et al.
Predictors of attempts to stop smoking
and their success in adult general
population samples: a systematic review.
Addiction. 2011;106(12):2110-2121.
15. Godtfredsen NS, Prescott E, Osler M, et al.
Predictors of smoking reduction and
cessation in a cohort of danish moderate and
heavy smokers. Prev Med. 2001;33(1):46-52.
16. Zhou X, Nonnemaker J, Sherrill B, et al.
Attempts to quit smoking and relapse:
factors associated with success or failure
from the ATTEMPT cohort study. Addict
Behav. 2009;34(4):365-373.
17. Covey LS, Glassman AH, Stetner F.
Cigarette smoking and major depression.
J Addict Dis. 1998;17(1):35-46.
18. Adams SG, Pugh JA, Kazis LE, et al.
Characteristics associated with sustained
abstinence from smoking among patients
with COPD. Am J Med. 2006;119(5):
441-447.
19. Lange P, Tøttenborg SS, Sorknes AD, et al.
Danish Register of chronic obstructive
pulmonary disease. Clin Epidemiol.
2016;8:1-6.
20. Lange P, Marott JL, Vestbo J, et al.
Socioeconomic status and prognosis of
COPD in Denmark. COPD. 2014;11(4):
431-437.
21. Lange P, Marott JL, Vestbo J, et al.
Prediction of the clinical course of chronic
obstructive pulmonary disease, using the
new GOLD classification: a study of the
[ 150#3 CHEST SEPTEMBER 2016
general population. Am J Respir Crit Care
Med. 2012;186(10):975-981.
22. Charlson ME, Pompei P, Ales KL, et al.
A new method of classifying prognostic
comorbidity in longitudinal studies:
development and validation. J Chronic
Dis. 1987;40(5):373-383.
23. Danmarks Statistik. Indkomster 2009.
Copenhagen, Denmark 2011. http://www.
dst.dk/publ/Indkomster. Accessed July 25,
2016.
24. Greenland S, Pearl J, Robins JM. Causal
diagrams for epidemiologic research.
Epidemiology. 1999;10(1):37-48.
25. Schiller JS, Ni H. Cigarette smoking
and smoking cessation among persons
with chronic obstructive pulmonary
disease. Am J Health Promot. 2006;20(5):
319-323.
26. Hiscock R, Judge K, Bauld L. Social
inequalities in quitting smoking: what
factors mediate the relationship between
socioeconomic position and smoking
cessation? J Public Health (Oxf).
2011;33(1):39-47.
27. Wagena EJ, Kant I, Huibers MJ, et al.
Psychological distress and depressed
mood in employees with asthma, chronic
bronchitis or emphysema: a population-
based observational study on prevalence
and the relationship with smoking
cigarettes. Eur J Epidemiol. 2004;19(2):
147-153.
28. Hagimoto A, Nakamura M, Morita T,
et al. Smoking cessation patterns and
predictors of quitting smoking among the
Japanese general population: a 1-year
follow-up study. Addiction. 2010;105(1):
164-173.
29. Osler M, Prescott E, Godtfredsen N, et al.
Gender and determinants of smoking
cessation: a longitudinal study. Prev Med.
1999;29(1):57-62.
30. Kaleta D, Usidame B, Dziankowska-
Zaborszczyk E, et al. Correlates of
cessation success among Romanian adults.
Biomed Res Int. 2014;2014:675496.
31. Morris JK, Cook DG, Shaper AG. Non-
employment and changes in smoking,
drinking, and body weight. BMJ.
1992;304(6826):536-541.
32. Tønnesen P, Mikkelsen K, Markholst C,
et al. Nurse-conducted smoking cessation
with minimal intervention in a lung clinic:
a randomized controlled study. Eur Respir
J. 1996;9(11):2351-2355.
33. Stelmach R, Fernandes FL, Carvalho-
Pinto RM, et al. Comparison between
objective measures of smoking and self-
reported smoking status in patients with
asthma or COPD: are our patients telling
us the truth? J Bras Pneumol. 2015;41(2):
124-132.
]