A combined smoking cessation intervention within a lung cancer screening trial: a pilot observational study


Aims and Background

The time of lung cancer screening may provide the ideal setting to discuss and initiate a smoking cessation plan that includes pharmacologic aids. No studies to date have fully investigated the potential effectiveness of such combined approach.


We prospectively evaluated the biochemically verified 1-year continuous abstinence rate from smoking of 187 persistent smokers enrolled within the Multicentric Italian Lung Detection Trial (MILD), who received a pharmacologic aid to quit smoking with varenicline along with behavioral counseling. The propensity of study subjects to succeed in smoking cessation was also monitored.


At 12 months, the continuous abstinence rate from smoking was achieved in 37 out of 187 patients (19.8%), with a propensity to succeed in smoking cessation for the assisted attempt equal to 1.43, as compared to an unassisted MILD patient. At the end of the third month of therapy, 48.7% of subjects showed a continuous abstinence rate, while only 33.7% of patients were abstinent from smoking at 6 months. At baseline, the subgroup of MILD participants who were originally allocated to lung tomography showed higher smoking intensity than those allocated to no screening.


A combined smoking cessation intervention can be implemented with satisfactory results within a lung cancer screening program; this preliminary observation needs to be replicated in a prospective investigation. Clinicians should consider that lung cancer screening may be falsely reassuring for persistent smokers; therefore it should always be coupled with a smoking cessation program.

Tumori 2015; 101(3): 306 - 311




Paolo Pozzi, Elena Munarini, Francesca Bravi, Marta Rossi, Carlo La Vecchia, Roberto Boffi, Ugo Pastorino

Article History


Financial support: This study has been funded by the Italian Ministry of Health, Programma “Centro Nazionale per la Prevenzione e il Controllo delle Malattie” anno 2008. Paolo Pozzi is supported by the “5 × 1000 anno 2008” contribution. Francesca Bravi, Marta Rossi and Carlo La Vecchia were supported by an AIRC grant 10068.
Conflict of interest: None.

This article is available as full text PDF.

Download any of the following attachments:


Lung cancer, which annualy causes 1.4 million deaths worldwide, is the most common cause of cancer death (1). In the Early Lung Cancer Action Program study, low-dose ­computed tomography (LDCT) showed potential to detect early-stage lung cancer (2), and the recent National Lung Screening Trial showed that a 20% reduction in mortality from lung cancer could be achieved through LDCT screening programs (3). Low-dose computed tomography also has been shown to serve as a “teachable moment” for ­smoking ­cessation (SC) (4). In the experience of Cox et al (5) and Townsend et al (6), 1-year self-reported smoking abstinence was 14%. In the Danish Lung Cancer Screening Trial, Ashraf et al (7) described a net quit rate from smoking of 6%. In the Dutch-Belgian NELSON LDCT trial, van der Aalst et al (8) reported a 2-year SC rate of 14.5%. These trials offered only cessation advice in spite of an integrated SC intervention, although their results were encouraging when compared to the SC prevalence of an unassisted population (9). Nevertheless, it would be useful to determine whether offering lung cancer screening without SC programs could be cost-effective (10) and whether LDCT contributes to altering smoking behaviors (11). In fact, being in an active arm of LDCT screening or receiving a negative screening result may provide relief to persistent smokers (8), rather than discouraging quit attempts (12). Furthermore, due to study design, most investigations routinely regarded the point prevalence rather than continuous abstinence rate from smoking (4-5-6-7).

To our knowledge, only one small study investigated the feasibility of a SC intervention in the LDCT setting (13). The subjects received nicotine replacement therapy or varenicline before or after the LDCT screening; the investigators registered a point prevalence of SC equal to 16.7% at the end of the study and they concluded that it might be better to offer SC support before the screening test (13).

We report here the results of a combined pharmacologic and behavioral SC intervention within a prospective LDCT lung cancer screening trial; for pharmacologic support, we chose the first-line antitobacco drug varenicline (14), a partial agonist of the α-4 β-2 nicotinic acetylcholine receptor. Varenicline reduces craving and satisfaction from smoking (15) and helps smokers to continuously abstain from smoking (16), the primary objective of this study.


Study population

The Multicentric Italian Lung Detection (MILD) is a randomized prospective lung cancer screening trial launched in 2005 at the Fondazione IRCCS Istituto Nazionale dei Tumori of Milan that included 4099 current or former smokers (17). The MILD participants could have been originally assigned either to a control group, where they received a longitudinal evaluation of respiratory function and brief advice for SC in each study visit, or to an early detection group, where they underwent the same program with the addition of LDCT; subjects in the LDCT arm were further randomized to receive computed tomography scan every 12 months (annual) or every 24 months (biennial).

The present research considered a subsample of patients already participating in the MILD trial who were enrolled in 2009-2010 to receive an intervention with varenicline plus individual behavioral counseling to promote continuous SC.

The study was designed as observational to follow the directive of the “Centro Nazionale per la Prevenzione e il Controllo delle Malattie” from the Italian Ministry of Health, which financially supported the study.

We selected a subcohort of MILD participants who satisfied the following criteria: age 49-75 years, persistent smoking despite brief advice for stopping smoking during prior MILD visits, and cumulative exposure to smoke of at least 20 pack-years. Individuals with renal, neurologic, or psychiatric disorders as well as those with a history of drug or alcohol abuse were excluded. The study was approved by the local institutional ethical committee (approval 09/49) and all subjects were asked to provide informed consent. The final population included 187 MILD participants who satisfied the eligibility criteria and signed the informed consent.

Baseline evaluation

At baseline, the participants completed the Fagerström Test of Nicotine Dependence (18) and underwent a forced expiratory volume in the first second evaluation (FEV1%) according to more recent ATS/ERS guidelines (19). Moreover, basal expired carbon monoxide (eCO) through an eCO reader (Micro Smokerlyzer, Bedfont Scientific Ltd., Kent, UK) was obtained (20). Participants were subsequently instructed to take 0.5 mg/day of varenicline for days 1-3, 0.5 mg twice per day for days 4-7, and 1 mg twice per day starting on day 8 and continuing until the end of week 12. Subjects were allowed to continue smoking only until the 14th day of therapy, after which they were supposed to quit smoking. Temporal or permanent dose reduction for tolerability problems was allowed. The duration of therapy was 3 months. The drug was provided free of charge by the study investigators, as the Italian Health System does not provide any reimbursement for SC drugs.

The one-on-one cognitive behavioral support consisted of 4 telephone calls, lasting at least 10 minutes, promoted by the same counselor during the months of treatment; the psychologist provided behavioral advice about how to cope with craving, and supported the motivation and the self-efficacy of the subject to quit or remain abstinent. If required by the participant, additional contacts were scheduled.

Follow-up evaluation

Follow-up assessments were set at months 1, 3, 6, and 12 after enrollment. At each medical visit, participants were asked about their smoking status, and they received an eCO reading in order to confirm smoking abstinence. A patient was defined as a sustained quitter if he or she reported being abstinent from smoking at follow-up visits and had an eCO of ≤6 ppm (21). Participants who missed visits or provided no information on their smoking status were considered persistent smokers. Therapy compliance was recorded by pill counts; the occurrence of adverse events was also monitored.

Study endpoints

The primary endpoint of the study was the rate of continuous abstinence from smoking, defined as sustained abstinence from smoking at months 3, 6, and 12, overall and in specific subgroups of subjects (basal pulmonary function and LDCT allocation arm). The point prevalence of SC and the propensity of study subjects to succeed in SC were also monitored.

Statistical analysis

Significant differences in the distributions of categorical variables were tested using the χ2 test, and continuous variables were tested by the paired t test. Differences were considered significant at p≤0.05. In order to compare the probability of quitting smoking in subjects who had been treated with varenicline and in an appropriate group of MILD participants who did not receive the study drug, we used the propensity score technique to find comparable couples belonging to these 2 groups (22, 23). We first identified a possible comparison group from 891 subjects who had been part of the MILD study in 2009-2010 for 1 year (median), were current smokers at that time, and had an available follow-up (including smoking status) after 10-14 months. On these subjects, we applied the propensity score technique by using a nonparsimonious multivariable logistic regression model in which the treatment with varenicline was the response variable and age, sex, education, occupation, history of smoking (duration and number of cigarettes), nicotine dependence score, FEV1%, MILD intervention group, time since enrolment in the MILD study, history of pulmonary diseases, history of cardiovascular diseases, overall health status, family history of cancer, previous attempts to quit smoking, living with smokers, and motivations for participating in the MILD study were included as covariates. This procedure allowed assembling 150 matched pairs of MILD subjects who either received varenicline or not and then comparing the 2 groups in terms of point prevalence of SC at month 12, since the biochemical verification of continuous abstinence was not envisaged in the MILD group. The odds ratio and the corresponding confidence interval for point prevalence of SC at month 12 was obtained from a conditional logistic regression model.


At month 12, 37 out of 187 subjects (19.8%) receiving the combined pharmacologic and psychological SC support achieved biochemically verified continuous abstinence from smoking. Table I shows selected characteristics of participants according to 1-year continuous abstinence: although there were no significant differences, sustained quitters tended to have a heavier history of smoking exposure and lower eCO, nicotine dependence score, and higher FEV1% as compared to persistent smokers. Even though not significant, the 1-year proportion of persistent smokers was higher in those who were actively screened for lung cancer and those who had a FEV1% <90%.

baseline characteristics of 187 participants according to continuous abstinence rate from smoking at month 12

Total (n = 187) Quitters (n = 37, 19.8%) Nonquitters (n = 150, 80.2%)
eCO = expired carbon monoxide; FTND = Fagerström Test of Nicotine Dependence; LDCT = low-dose computed tomography; MILD = Multicentric Italian Lung Detection Trial.
aThe sum does not add up to the total because of missing values.
bSubjects who did not receive active screening.
Age, y, n (%)
 <55 82 (43.9) 14 (37.8) 68 (45.3)
 55-59 62 (33.2) 14 (37.8) 48 (32.0)
 60-64 35 (18.7) 8 (21.6) 27 (18.0)
 ≥65 8 (4.3) 1 (2.7) 7 (4.7)
 Median (range) 55 (47-72) 57 (49-70) 55 (47-72)
 Mean ± SD 55.9 ± 4.6 56.3 ± 4.6 55.8 ± 4.6
Sex, n (%)
 Male 117 (62.6) 22 (59.5) 95 (63.3)
 Female 70 (37.4) 15 (40.5) 55 (36.7)
Duration of smoking, y, n (%)
 <40 115 (61.5) 21 (56.8) 94 (62.7)
 ≥40 72 (38.5) 16 (43.2) 56 (37.3)
Cigarettes per day, n, n (%)
 <20 69 (36.9) 14 (37.8) 55 (36.7)
 ≥20 118 (63.1) 23 (62.2) 95 (63.3)
Pack-years of cigarettes
 Median (range) 20.0 (0-120) 22.5 (0-90) 20.0 (0-120)
 Mean ± SD 22.4 ± 22.0 25.5 ± 25.0 21.6 ± 21.2
Baseline eCO, ppm
 Median (range) 16 (0-41) 14 (2-34) 16 (0-41)
 Mean ± SD 16.6 ± 7.5 14.9 ± 7.8 17.0 ± 7.3
FTND, score
 Median (range) 5 (0-9) 5 (0-8) 5 (0-9)
 Mean ± SD 5.1 ± 2.0 4.7 ± 2.1 5.2 ± 2.0
FEV1%, n (%)
 <90 101 (54.0) 16 (43.2) 85 (56.7)
 ≥90 86 (46.0) 21 (56.8) 65 (43.3)
Body mass index, n (%)a
 <20 13 (7.0) 1 (2.7) 12 (8.0)
 20-24.9 70 (37.4) 15 (40.5) 55 (36.7)
 ≥25 98 (52.4) 21 (56.8) 77 (51.3)
 Median (range) 25.4 (15.6-45.0) 26.0 (19.6-37.4) 25.3 (15.6-45.0)
 Mean ± SD 26.1 ± 5.0 25.9 ± 4.3 26.2 ± 5.2
MILD intervention group, n (%)
 Controlb 81 (43.3) 19 (51.4) 62 (41.3)
 LDCT 106 (56.7) 18 (48.7) 88 (58.7)
Year of enrollment (MILD), n (%)
 2005-2006 124 (66.3) 28 (75.7) 96 (64.0)
 2007-2010 63 (33.7) 9 (24.3) 54 (36.0)

The point prevalence of SC was 52.4%, 48.7%, 33.7%, and 32.6% at months 1, 3, 6, and 12, respectively; 28 out 91 subjects (30.7%) who were abstinent from smoking at the end of the third month of the combined SC intervention displayed an early relapse from smoking, shortly after the end of the therapy (data not shown).

In subjects who had a similar propensity to receive the combined SC intervention within the MILD trial on the basis of several covariates (Tab. II), we registered a difference in the 12-month point prevalence of SC between those who had been treated or not (34% versus 5.3%, respectively). The odds ratio of SC for the assisted attempt was 1.43 (95% confidence interval 1.11-1.84), according to the conditional model.

Covariates included in the propensity score (combined smoking cessation intervention or matched mild participant)

Mild (n = 150) Combined intervention (n = 150)
FTND = Fagerström Test of Nicotine Dependence; LDCT = low-dose computed tomography; MILD = Multicentric Italian Lung Detection Trial.
aSubjects who did not receive active screening.
Age, y, n (%)
 <55 67 (44.67) 62 (41.33)
 55-59 52 (34.67) 50 (33.33)
 60-64 24 (16.00) 30 (20.00)
 ≥65 7 (4.67) 8 (5.33)
Sex, n (%)
 Male 90 (60.00) 93 (62.00)
 Female 60 (40.00) 57 (38.00)
Duration of smoking, y, n (%)
 <40 97 (64.67) 89 (59.33)
 ≥40 53 (35.33) 61 (40.67)
Cigarettes per day, n (%)
 <20 40 (26.67) 55 (36.67)
 ≥20 110 (73.33) 95 (63.33)
FTND (score), mean ± SD 5 ± 2.2 5.0 ± 2.0
FEV1%, n (%)
 <90 38 (25.33) 85 (56.67)
 ≥90 112 (74.67) 65 (43.33)
Years since enrollment, median (range) 1 (1-4) 1 (1-4)
MILD intervention group, n (%)
 Controla 62 (41.33) 56 (37.33)
 LDCT 88 (58.67) 94 (62.67)

Table III gives the subjects’ smoking habits at baseline, according to the MILD intervention groups. There were no significant differences between the two allocation arms, except for the baseline number of cigarettes smoked daily, which tended to be higher in LDCT participants than in control subjects. A significant difference in baseline eCO was seen in LDCT participants as compared to control participants (LDCT 17.8 + 7.2 versus control 14.9 + 7.5 ppm, p = 0.0086), possibly reflecting the reassuring effect of receiving a computed tomography examination after a median of 2 years from the beginning of lung cancer screening.

Smoking habits according to mild intervention group at baseline

Baseline pack-years FTND Cigarettes eCO
eCO = expired carbon monoxide; FTND = Fagerström Test of Nicotine Dependence; LDCT = low-dose computed tomography.
Values are mean ± SD.
aSubjects who did not receive active screening.
Controla (n = 81) 22.5 ± 22.6 4.9 ± 1.9 18.5 ± 10.8 14.9 ± 7.5
LDCT (n = 106) 22.3 ± 21.7 5.2 ± 2.1 20.8 ± 10.6 17.8 ± 7.2
p Value 0.9437 0.3346 0.1579 0.0086

Table IV shows the therapy compliance and the side effects recorded during the study. Therapy retention was 61.1%, with 35.4% of participants reporting that they experienced a side effect, mostly gastrointestinal disorders, insomnia, irritability, or abnormal dreams; no suicidal intention or major cardiovascular side effects were observed. The proportion of participants who discontinued the therapy was similar among subjects who experienced at least one side effect and those who did not, whereas the proportion of subjects who reduced the therapy was higher in subjects who reported side effects, as compared to those who did not; therapy compliance did not seem to be dependent on a specific side effect.

Side effects and therapy discontinuation

Therapy compliance, n (%)
Total Discontinued Reduced Complete
aThe sum does not add up to the total of 61 quitters and 126 nonquitters because of missing values on side effects.
bGastrointestinal disorders include nausea and other gastroenteric disorders.
cPsychiatric disorders include insomnia, irritability and abnormal dreams.
Totala 175 42 (24.0) 26 (14.9) 107 (61.1)
Side effectsa
 Yes 62 15 (24.2) 15 (24.2) 32 (51.6)
Type of side effectsa
 Gastrointestinal disordersb 41 8 (19.5) 9 (22.0) 24 (58.5)
 Psychiatric disordersc 20 7 (35.0) 5 (25.0) 8 (40.0)
 Headache 1 0 (0.0) 1 (100.0) 0 (0.0)
 No 113 27 (23.9) 11 (9.7) 75 (66.4)


The present study demonstrated that combined SC support offered to persistent smokers enrolled in a lung cancer screening trial can effectively alter their smoking behaviors. By combining a 3-month course of varenicline with individual SC behavioral counseling, we were able to register a 1-year continuous abstinence rate from smoking of 19.8%.

The long-term results of our investigation are concordant with the literature (15, 16). The novelty of our study was in introducing an integrated SC intervention where it is not routinely used, as previous experiences considered only the “teachable moment” related to LDCT screening (4-5-6-7-8).

Our study confirms the feasibility data depicted by Ferketich et al (13) and supports the indication for a SC treatment in a new setting (i.e., lung cancer screening). Our study primarily examined the continuous SC rate, an innovative aspect of our study. Van der Aaslt et al (8, 12) reported sustained smoking abstinence among smokers enrolled in the NELSON trial, 14.9% and 14.6% in the screening and control arm, or 8.9% and 10.9% in smokers who received a negative or indeterminate screening result, but these were prevalent estimates with no biochemical verification. Our investigation applied a rigorous procedure (24) addressing proper duration of smoking abstinence with a strict eCO cutoff (21).

Varenicline was chosen as the study drug because it showed a nearly tripled probability to succeed in long-term SC, with respect to unassisted attempts (25); furthermore, providing a behavioral support in subjects using pharmacotherapy to stop smoking has been shown to increase their chance of success in SC (26).

Our study has several limitations, including lack of randomization to receive placebo or varenicline; application of a propensity score can reduce but not eliminate selection and other sources of bias. On the other hand, our intervention was introduced when the MILD trial had been ongoing for 4 years, resulting in two thirds of enrollees who were older MILD participants (i.e., enrolled in 2005-2006); due to this selection bias, we cannot rule out a longitudinal behavioral change that might take place in a patient who participates in several LDCT rounds (6), the effect of being treated before the LDCT screening (13), or the effect of receiving a negative/indeterminate LDCT result (12). Such a selection may result in an unpredictable outcome of our investigation.

As such, we tried to fill some gaps by applying the propensity score technique. In this scenario, the subjects of the MILD trial who received the assisted attempt had a 43% higher rate of succeeding in SC, as compared to the unassisted ones.

The present study allowed also us to further analyze the smoking habits of this MILD population. We found that, comparably to what van der Aalst et al described (8), subjects allocated to an active screening arm may perceive LDCT as a relief for their dangerous behaviour of being persistent smokers; in fact, when entering in the SC trial, LDCT participants smoked more cigarettes and showed higher eCO values than control participants, with the latter being statistically significant.

One third of subjects relapsed as soon as the active SC intervention ended; this trend was seen in other reports and represents a limitation of existing SC treatments (15, 16). Our investigation was not focused on preventing the smoking ­relapse by analyzing the quitting pattern of the subjects, thus we did not investigate the possible benefit of a lengthened varenicline treatment, as depicted by Hajek et al (27); on the other hand, we already demonstrated that there is a specific group of LDCT participants with a greater risk to develop lung cancer (28).

Future LDCT screening trials should include a combined SC intervention for all smokers, especially those who are actively screened for lung cancer and who may have aggressive cancers. This intervention may be designed in a prospective fashion and may consider different (e.g., nicotine replacement therapy or bupropion) or extended treatment for subjects with early relapse of smoking. Lung cancer screening should move from a moment of early detection into a dynamic process of smoking prevention.


This work was made possible by a (year) law by the Italian government which allowed Italian citizens to allocate the 5 × 1000 share of their tax payment to support a research or charitable institution of their choice. We wish to thank all those citizens who decided to donate their 5 × 1000 to IRCCS Fondazione Istituto Nazionale dei Tumori, Milan.


Financial support: This study has been funded by the Italian Ministry of Health, Programma “Centro Nazionale per la Prevenzione e il Controllo delle Malattie” anno 2008. Paolo Pozzi is supported by the “5 × 1000 anno 2008” contribution. Francesca Bravi, Marta Rossi and Carlo La Vecchia were supported by an AIRC grant 10068.
Conflict of interest: None.
  • 1. Jemal A.,Bray F.,Center MM.,Ferlay J.,Ward E.,Forman D. Global cancer statistics. CA Cancer J Clin 2011; 61: 69-90 Google Scholar
  • 2. Henschke CI.,McCauley DI.,Yankelevitz DF. Early Lung Cancer Action Project: overall design and findings from baseline screening. Lancet 1999; 354: 99-105 Google Scholar
  • 3. Aberle DR.,Adams AM.,Berg CD. National Lung Screening Trial Research Team. N Engl J Med 2011; 365: 395-409 Google Scholar
  • 4. Ostroff JS.,Buckshee N.,Mancuso CA.,Yankelevitz DF.,Henschke CI. Smoking cessation following CT screening for early detection of lung cancer. Prev Med 2001; 33: 613-621 Google Scholar
  • 5. Cox LS.,Clark MM.,Jett JR. Change in smoking status after spiral chest computed tomography scan screening. Cancer 2003; 98: 2495-2501 Google Scholar
  • 6. Townsend CO.,Clark MM.,Jett JR. Relation between smoking cessation and receiving results from three annual spiral chest computed tomography scans for lung carcinoma screening. Cancer 2005; 103: 2154-2162 Google Scholar
  • 7. Ashraf H.,Tønnesen P.,Holst Pedersen J.,Dirksen A.,Thorsen H.,Døssing M. Effect of CT screening on smoking habits at 1-year follow-up in the Danish Lung Cancer Screening Trial (DLCST). Thorax 2009; 64: 388-392 Google Scholar
  • 8. van der Aalst CM.,van den Bergh KA.,Willemsen MC.,de Koning HJ.,van Klaveren RJ. Lung cancer screening and smoking abstinence: 2 year follow-up data from the Dutch-Belgian randomised controlled lung cancer screening trial. Thorax 2010; 65: 600-605 Google Scholar
  • 9. Quitting smoking among adults: United States, 2001-2010. MMWR Morb Mortal Wkly Rep 2011; 60: 1513-1519 Google Scholar
  • 10. McMahon PM.,Kong CY.,Bouzan C. Cost-effectiveness of computed tomography screening for lung cancer in the United States. J Thorac Oncol 2011; 6: 1841-1848 Google Scholar
  • 11. Bach PB.,Mirkin JN.,Oliver TK. Benefits and harms of CT screening for lung cancer: a systematic review. JAMA 2012; 307: 2418-2429 Google Scholar
  • 12. van der Aalst CM.,van Klaveren RJ.,van den Bergh KA.,Willemsen MC.,de Koning HJ. The impact of a lung cancer computed tomography screening result on smoking abstinence. Eur Respir J 2011; 37: 1466-1473 Google Scholar
  • 13. Ferketich AK.,Otterson GA.,King M.,Hall N.,Browning KK.,Wewers ME. A pilot test of a combined tobacco dependence treatment and lung cancer screening program. Lung Cancer 2012; 76: 211-215 Google Scholar
  • 14. Clinical Practice Guideline Treating Tobacco Use and Dependence 2008 Update Panel, Liaisons, and Staff. Am J Prev Med 2008; 35: 158-176 Google Scholar
  • 15. Gonzales D.,Rennard SI.,Nides M. Varenicline, an α4β2 nicotinic acetylcholine receptor partial agonist, vs sustained-release bupropion and placebo for smoking cessation: a randomized controlled trial. JAMA 2006; 296: 47-55 Google Scholar
  • 16. Jorenby DE.,Hays JT.,Rigotti NA. Efficacy of varenicline, an α4β2 nicotinic acetylcholine receptor partial agonist, vs placebo or sustained-release bupropion for smoking cessation: a randomized controlled trial. JAMA 2006; 296: 56-63 Google Scholar
  • 17. Pastorino U.,Rossi M.,Rosato V. Annual or biennial CT screening versus observation in heavy smokers: 5-year results of the MILD trial. Eur J Cancer Prev 2012; 21: 308-315 Google Scholar
  • 18. Heatherton TF.,Kozlowski LT.,Frecker RC.,Fagerström KO. The Fagerström Test for Nicotine Dependence: a revision of the Fagerström Tolerance Questionnaire. Br J Addict 1991; 86: 1119-1127 Google Scholar
  • 19. Miller MR.,Hankinson J.,Brusasco V. ATS/ERS Task Force. Eur Respir J 2005; 26: 319-338 Google Scholar
  • 20. Jarvis MJ.,Russell MA.,Saloojee Y. Expired air carbon monoxide: a simple breath test of tobacco smoke intake. BMJ 1980; 281: 484-485 Google Scholar
  • 21. Middleton ET.,Morice AH. Breath carbon monoxide as an indication of smoking habit. Chest 2000; 117: 758-763 Google Scholar
  • 22. Rosenbaum PR Propensity score. In: Armitage P Colton T eds. Encyclopedia of Biostatistics, volume 5 New York: John Wiley and Sons; 1998 3551 3555 Google Scholar
  • 23. Cavuto S.,Bravi F.,Grassi MC.,Apolone G. Propensity score for the analysis of observational data. Drug Dev Res 2006; 67: 208-216 Google Scholar
  • 24. West R.,Hajek P.,Stead L.,Stapleton J. Outcome criteria in smoking cessation trials: proposal for a common standard. Addiction 2005; 100: 299-303 Google Scholar
  • 25. Cahill K.,Stevens S.,Perera R.,Lancaster T. Pharmacological interventions for smoking cessation: an overview and network meta-analysis. Cochrane Database Syst Rev 2013; 5: - Google Scholar
  • 26. Stead LF.,Lancaster T. Behavioural interventions as adjuncts to pharmacotherapy for smoking cessation. Cochrane Database Syst Rev 2012; 12: - Google Scholar
  • 27. Hajek P.,Tønnesen P.,Arteaga C.,Russ C.,Tonstad S. Varenicline in prevention of relapse to smoking: effect of quit pattern on response to extended treatment. Addiction 2009; 104: 1597-1602 Google Scholar
  • 28. Boeri M.,Verri C.,Conte D. MicroRNA signatures in tissues and plasma predict development and prognosis of computed tomography detected lung cancer. Proc Natl Acad Sci USA 2011; 108: 3713-3718 Google Scholar



  • Tobacco Control Unit, Fondazione IRCCS Istituto Nazionale dei Tumori, Milan - Italy
  • Department of Epidemiology, IRCCS Istituto di Ricerche Farmacologiche Mario Negri, Milan - Italy
  • Department of Clinical Sciences and Community Health, University of Milan, Milan - Italy
  • Unit of Thoracic Surgery, Fondazione IRCCS Istituto Nazionale dei Tumori, Milan - Italy
  • Paolo Pozzi and Elena Munarini contributed equally to this work.

Article usage statistics

The blue line displays unique views in the time frame indicated.
The yellow line displays unique downloads.
Views and downloads are counted only once per session.

No supplementary material is available for this article.