Engagement and Effectiveness of a Smoking Cessation Quitline Intervention in a Thoracic Surgery Clinic

Key Points

Question  What factors are associated with a higher level of engagement and smoking cessation outcomes among patients in a thoracic surgery clinic who agreed to participate in a smoking quitline intervention?

Findings  Although the level of quitline engagement in this cohort study was similar between patients who underwent surgery (55%) and those who did not undergo surgery (45%), those undergoing surgery were nearly twice as likely (64% vs 36%) to quit smoking regardless of quitline participation. Patients undergoing surgery who engaged in the quitline had a 1.8-fold higher odds of point prevalence abstinence at 6 months than those who did not.

Meaning  Patients undergoing a thoracic operation are motivated to quit smoking, and quitline engagement amplifies long-term abstinence.

Importance  Smoking quitline programs effectively promote smoking cessation in outpatient primary care settings.

Objective  To examine the factors associated with smoking quitline engagement and smoking cessation among patients undergoing thoracic surgery who consented to a quitline electronic referral.

Design, Setting, and Participants  A retrospective cohort study was conducted from January 1, 2014, to December 31, 2018, among 111 active smoking patients referred to the quitline from a thoracic surgery outpatient clinic visit. Patients were divided into operative and nonoperative cohorts.

Main Outcomes and Measures  Primary outcomes were engagement rates in the quitline program and successful smoking cessation. Secondary outcomes were self-reported point prevalence abstinence at 1 month and 6 months after the smoking quit date.

Results  Of 111 patients (62 men; mean [SD] age, 61.8 [11.2] years) who had a quitline referral, 58 (52%) underwent surgery, and 32 of these 58 patients (55%) participated in the program. Of the 53 nonoperative patients (48%), 24 (45%) participated in the program. In the operative cohort, there was no difference in the smoking cessation rate between quitline participants and nonparticipants (21 of 32 [66%] vs 16 of 6 [62%]; P = .79) or in point prevalence abstinence at 1 month (23 of 32 [72%] vs 14 of 25 [56%]; P = .27) or 6 months (14 of 28 [50%] vs 6 of 18 [33%]; P = .36). Similarly, in the nonoperative cohort, there was no difference in the smoking cessation rate between quitline participants and nonparticipants (8 of 24 [33%] vs 11 of 29 [38%]; P = .78) or in point prevalence abstinence at 1 month (7 of 24 [29%] vs 8 of 27 [30%]; P = .99) or 6 months (6 of 23 [26%] vs 6 of 25 [24%]; P = .99). Regardless of quitline participation, operative patients had a 1.8-fold higher proportion of successful smoking cessation compared with nonoperative patients (37 of 58 [64%] vs 19 of 53 [36%]; P = .004) as well as a 2.2-fold higher proportion of 1-month point prevalence abstinence (37 of 57 [65%] vs 15 of 51 [29%]; P < .001) and a 1.8-fold higher proportion of 6-month point prevalence abstinence (20 of 45 [44%] vs 12 of 48 [25%]; P = .05). Having surgery doubled the odds of smoking cessation (odds ratio, 2.44; 95% CI, 1.06-5.64; P = .04) and quitline engagement tripled the odds of remaining smoke free at 6 months (odds ratio, 3.57; 95% CI, 1.03-12.38; P = .04).

Conclusions and Relevance  Patients undergoing thoracic surgery were nearly twice as likely to quit smoking as those who did not have an operation, and smoking quitline participation further augmented point prevalence abstinence. Improved smoking cessation rates, even among nonoperative patients, were associated with appropriate outpatient counseling and intervention.

Preoperative smoking cessation is associated with a reduction in postoperative pulmonary complications, especially after thoracic surgery.^1-3 A longer time of smoking cessation prior to an elective operation is associated with a lower risk of pulmonary complications.^4,5 Therefore, the period between the initial outpatient visit with a surgeon and the surgery date is a critical time to encourage smoking cessation to reduce the risk of postoperative complications and increase the chance of long-term smoking cessation. This sentiment is acknowledged by the International Association for the Study of Lung Cancer. Recognizing the critical importance of smoking cessation after cancer diagnosis, in September 2019, the International Association for the Study of Lung Cancer recommended for patients undergoing treatment for lung cancer that evidence-based tobacco cessation assistance should be routinely and integrally incorporated into multidisciplinary cancer care for the patients and their family members.⁶ Given this importance, clinicians must have effective smoking cessation resources to provide to their patients.

Smoking cessation quitline (QL) programs have been used for more than 3 decades,^7,8 but more recent evidence has demonstrated their effectiveness in guiding patients to quit smoking in the short and long term.^9-12 Proactive electronic referrals have been associated with increased engagement with QL services,¹³ and, to our knowledge, University of California, Davis Health established the first bidirectional electronic referral to the California Smokers’ Helpline (CSH) as an adjunct to aid in outpatient clinic smoking cessation efforts.¹⁴ Our thoracic surgical outpatient clinic has prioritized the use of the QL program, accounting for 4% to 6% of the annual total health system referrals to the program since 2014. Despite the demonstrated effectiveness of these QL interventions, to our knowledge, there has yet to be a study evaluating the role that QL programs play in referring patients for thoracic surgery. To this end, we investigated the engagement and effectiveness of QL referrals from our thoracic surgery clinic. Many patients who have experienced significant life-changing events, such as screening for lung cancer,¹⁵ recovery from a myocardial infarction,¹⁶ or planning for elective surgery,¹⁷ have shown an increased motivation to succeed in smoking cessation attempts.¹⁸Quiz Ref ID We tested the hypotheses that patients undergoing surgery would acheive higher rates of smoking cessation success and longer-term abstinence compared with patients not undergoing surgery and that participation in the QL program would amplify long-term abstinence.

We performed a single-center cohort study of all actively smoking patients who consented to be referred to the CSH QL program from a thoracic surgery clinic visit during the period from January 1, 2014, to December 31, 2018. Quiz Ref IDThe CSH is a telephone smoking cessation QL program that is free to all callers and funded by the California Department of Public Health. Counselors provide an initial counseling session (approximately 30 minutes) about preparing to quit and enhancing motivation, self-efficacy, coping strategies, and a self-image as a nonsmoker; several follow-up sessions (approximately 10 minutes) are offered to address relapse within the month after the quit date.⁹ The thoracic surgery clinicians advised every patient who smokes to quit and offered a QL referral for support. We merged demographic, clinical, and perioperative data from the electronic medical record as well as our institution’s thoracic surgical database cohort of patients from this time period with the CSH QL data regarding engagement in the program and success of smoking cessation. The study was conducted under the approval and oversight of the University of California, Davis Institutional Review Board and was performed with a waiver for a requirement of informed consent as it was determined not to be research involving human participants.

We assessed the characteristics of the patients, including sex, age, race/ethnicity, comorbidities, pack-year history of cigarette use, and Eastern Cooperative Oncology Group performance status. Race/ethnicity was defined by patient-reported responses on the initial clinic visit intake form. Race/ethnicity was then dichotomized into white or nonwhite to simplify the multivariate analysis to assess whether race/ethnicity was associated with smoking cessation success or point prevalence abstinence. We divided patients into 2 groups: patients referred to QL who did not undergo an operation (nonoperative) and patients referred to QL who underwent either lung or esophageal surgery (operative). We further subdivided each group into the following 2 subgroups: (1) QL participants who were reached by QL program staff and who actually engaged in the program after referral and (2) QL nonparticipants who were unable to be contacted or who declined to participate in the program. There were no missing data in the operative group. In the nonoperative group, there was 1 QL participant lost to follow-up and 4 QL nonparticipants lost to follow-up at 6 months. These missing patients were eliminated from the denominator of the proportions of patients who quit smoking at each time point.

The primary outcomes were engagement in the QL program after referral and smoking cessation. Engagement was defined by the QL referral order results as a patient who was contacted and did not decline to participate in services. Smoking cessation was defined by either a reported quit date in the QL database or by manual electronic medical record review of a time-stamped smoking quit date under tobacco history or documentation of a quit date within a primary care or thoracic surgery clinic note. Secondary outcomes included self-reported point prevalence abstinence at 1 month after the smoking quit date (or defined as the first postoperative visit for patients in the operative group) and at 6 months after the smoking quit date. We sought to ascertain visits in the electronic medical record at or near the 1-month and 6-month marks after smoking cessation to collect data on point prevalence abstinence at 1 month and 6 months after the quit date. This information was readily available for our operative cohort because these time points correlated with 1-month and 6-month postoperative thoracic surgery clinic visits in which smoking abstinence status is regularly determined. In addition, those participating in the QL program had smoking abstinence data at the 6-month time point that were rigorously obtained via telephone follow-up, during the period from 2017 to 2019. The QL participant smoking abstinence data prior to 2017 were obtained by manual electronic medical record review as already described.

Analyses of the Fisher exact test and the t test were used to compare demographic and clinical outcomes as well as perform univariate comparisons of outcomes between groups and subgroups. All P values were from 2-sided tests and results were deemed statistically significant at P < .05. Missing smoking data at any time point were eliminated from the analysis (ie, a complete-case analysis). Multivariable logistic regression including sex, age, race/ethnicity, smoking pack-years, undergoing an operation, and participating in the QL was used to identify factors associated with participation in the QL program, smoking cessation, and point prevalence abstinence at 1 month and 6 months. We initially performed univariate logistic regression on a larger set of independent variables and then included in the final multivariable regression only variables with P < .25 or variables thought to have an a priori association with smoking cessation. An entry method with retention of all variables was used to construct the final multivariable model. Analyses were conducted with SPSS Statistics, version 25 (IBM Corp).

Of 111 patients (62 men; mean [SD] age, 61.8 [11.2] years) referred to the QL, 58 (52%) ultimately underwent surgery. The demographic characteristics of the patients are summarized in Table 1. There were 49 patients (84%) who underwent lung surgery and 9 patients (16%) who underwent esophageal surgery, with the most common indication for surgery being lung cancer (33 [57%]) (Table 2).

Of the 53 patients in the nonoperative group, 24 (45%) engaged in the QL program, and 8 (33%) of these patients successfully quit smoking (Figure). The point prevalence abstinence rate was 29% (7 of 24) at 1 month after the quit date and 26% (6 of 23) at 6 months after the quit date. Of the 29 patients (55%) in the nonoperative group who did not engage in the QL program, 11 (38%) quit smoking, with a 1-month point prevalence abstinence rate of 29% (15 of 51) and a 6-month point prevalence abstinence rate of 24% (6 of 25). Quiz Ref IDThere was no difference in the rates of initial smoking cessation and 1-month and 6-month point prevalence abstinence between nonoperative group QL participants and nonparticipants.

Of the 58 patients (52%) in the operative group referred to the QL, 32 (55%) participated in the program, and 21 (66%) of these patients successfully quit smoking preoperatively (Figure). These patients quit a mean (SD) of 31.2 (29.4) days prior to surgery, which did not differ significantly from patients in the operative group who quit without the QL a mean (SD) 60.8 (63.8) days prior to surgery (P = .10). The point prevalence abstinence rate among operative group QL participants was 72% (23 of 32) 1 month after surgery and 50% (14 of 28) 6 months after surgery. Of the 26 patients (45%) in the operative group who did not enroll in the QL program despite referral, 16 (62%) successfully quit preoperatively. The point prevalence abstinence rate decreased to 56% (14 of 25) at the first postoperative visit and to 33% (6 of 18) 6 months after surgery. Quiz Ref IDThere was no difference in rates of initial smoking cessation and in 1-month and 6-month point prevalence abstinence rates between operative group QL participants and nonparticipants. There was no difference in perioperative complication rates between patients who did and did not successfully quit smoking preoperatively (eTable in the Supplement).

On univariate analysis, patients in the operative group did not have a higher QL participation rate compared with nonoperative patients (32 of 58 [55%] vs 24 of 53 [45%]; P = .30). However, regardless of QL participation, patients in the operative group had a 1.8-fold higher proportion of successful smoking cessation compared with patients in the nonoperative group (37 of 58 [64%] vs 19 of 53 [36%]; P = .004), as well as a 2.2-fold higher proportion of 1-month point prevalence abstinence (37 of 57 [65%] vs 15 of 51 [29%]; P < .001) and a 1.8-fold higher proportion of 6-month point prevalence abstinence (20 of 45 [44%] vs 12 of 48 [25%]; P = .05).

On multivariable logistic regression, there were no significant factors associated with a referred patient who successfully quit to remain smoke free at 1 month. However, undergoing an operation increased the odds of initially successfully quitting smoking by 2.44 times (95% CI, 1.06-5.64; P = .04) (Table 3). Quiz Ref IDParticipating in the QL increased the odds of being abstinent at 6 months by 3.57 times (95% CI, 1.03-12.38; P = .04) (Table 4).

Quitline programs are associated with significantly improved smoking cessation efforts across many outpatient environments. Our data demonstrate that QL programs may augment cessation efforts when they are incorporated into a subspecialty preoperative clinic. Patients in the operative group were nearly twice as likely to successfully quit smoking as patient in the nonoperative group, and the only factor after adjustment that was associated with long-term smoke-free point prevalence abstinence at 6 months was participation in the QL program.

Patients undergoing an operation, especially one involving the lungs or esophagus, may be highly motivated to reduce perioperative risk factors such as active smoking.^19,20 Thus, it is perhaps unsurprising that a cohort of patients seeing a subspecialist in a preoperative clinic would have higher smoking cessation success and durability. Historically, the CSH QL program reported smoking cessation intention-to-treat rates of prolonged abstinence of 24% at 1 month and 13% at 6 months within the community setting.⁸ The patients seen in our subspecialty surgical clinic, even those who did not go on to have surgery, had higher smoking cessation rates. However, the patients undergoing an operation had over 2.5 times higher rates of initial smoking cessation, and they continued cessation at 1 and 6 months. Involvement in the QL program seemed to have little additional association with outcomes for patients in the nonoperative group but was associated with the point prevalence abstinence of the operative cohort up to 6 months after surgery. Our 6-month point prevalence abstinence results were higher than what was recently reported for electronic QL referral at the University of California, Davis from 2012 to 2016, with 576 patients having a 12.2% point prevalence abstinence rate at 6 to 12 months (all missing data on smoking cessation in that study were interpreted as the patient continuing to smoke).¹⁴

Quitline programs have been effective in real-world situations, with similar results as nicotine gum and patches.⁹ Quitline programs have further improved through the use of interactive websites and have revised strategies by including more proactive (counselor-initiated) cases,^10,21,22 as well as text message–based reminders.²³ Active referral to a program rather than reliance on a patient contacting a QL or website with a passive referral has been repeatedly demonstrated to improve enrollment rates,^24,25 and multimodality programs encompassing both educational programs as well as pharmacotherapy have shown further benefits in cessation effectiveness. A recent prospective study of such a multimodality smoking cessation program implemented in a comprehensive cancer center found that, among the 3245 patients enrolled, smoking abstinence rates were 45% at 3 months and 46% at 6 months.²⁶ These smoking cessation rates remained high regardless of a patient being currently treated for cancer, having a prior diagnosis of cancer, or having no cancer diagnosis. Thus, multimodality smoking abstinence programs can provide a significant benefit to the smoking cessation efforts of patients encountered across the entire health care system. Conversely, a recent study of inpatient smokers found that the patient-reported 30-day abstinence rate at 6 months after discharge was approximately 20% whether patients received no intervention, nicotine patches after discharge, a CSH QL telephone referral, or both nicotine patches and a QL referral.²⁷ The authors hypothesized that low QL engagement (47% of referred patients participated in QL counseling sessions) diminished the effectiveness of this intervention.

There is heterogeneity among thoracic surgeons with regard to requiring smoking cessation prior to lung surgery.^28,29 One study surveyed 200 thoracic surgeons who participate in the Society of Thoracic Surgeons General Thoracic Surgery Database and found that 40% of respondents require patients to quit smoking prior to lung resection, yet only 9.5% of these surgeons routinely refer their active smoking patients to a smoking cessation program.²⁸ A similar study found that up to 77% of thoracic surgeons would refuse to operate on a patient requiring a pneumonectomy if they were unable to achieve smoking abstinence.²⁹ Only 50% of the surveyed surgeons had institutional support for smoking cessation programs. As such, it is important that subspecialty clinics have access to efficacious smoking cessation options such as QL programs.

We are currently transitioning from the model of an optional “opt-in” smoking cessation QL referral system within our thoracic surgery clinic to an “opt-out” model in which all active smokers are automatically referred to the QL or cancer center program with a tobacco treatment specialist who can also coordinate cessation medications. In the future, we are planning to compare the results of the opt-in program described presently with our ongoing data collection after initiation of the opt-out program. In addition, we intend to examine point prevalence abstinence beyond 6 months. Our current analysis suggests that high initial smoking cessation rates can be realized in this cohort of patients being evaluated in a specialty surgical clinic. However, we should not be content with initial short-term success if long-term smoking recidivism rates remain high. Our hope is that the transition to an opt-out QL enrollment strategy may help bolster long-term smoking point prevalence abstinence in our patients, as QL participation was the only factor associated with improved 6-month point prevalence abstinence in our analysis. In addition, our opt-out program for patients with cancer will work toward proactive follow-up of all patients with QL referrals regarding use of or interest in tobacco cessation medications, in conjunction with the health management education team that includes a pharmacist. Future studies may also try to identify more granular patient factors associated with difficulty in participating in the QL program or successfully quitting and remaining smoke free. Such factors may include socioeconomic status, geographical distance from the clinic, insurance status, and educational level and may help better identify patients at highest risk of poor QL participation and smoking recidivism.

There are limitations to our study, including the small sample size. In addition, smoking cessation was not chemically confirmed in patients; therefore, there is a possibility of patient reporting bias. We were not able to compare our smoking cessation and point prevalence abstinence rates of patients referred to the QL with those of patients seen in our clinic who refused referral to the QL. Other factors not included in our analysis were use of other tobacco cessation pharmaceuticals and whether patients had an in-network primary care clinician who provided regular follow-up. However, it is possible that the primary care visits that were used to collect data on smoking abstinence for the nonoperative and nonparticipant patient subgroups may not have been as rigorously collected, leading to reporting error.

Patients undergoing thoracic surgery were nearly twice as likely to quit smoking, regardless of QL referral. Patients in the operative group had comparable preoperative quit rates regardless of QL referral, but those participating in the QL had a higher point prevalence abstinence rate at 6 months. As such, subspecialty clinicians can improve smoking cessation rates with appropriate counseling and intervention, even among patients who do not undergo surgery.

Accepted for Publication: March 15, 2020.

Corresponding Author: David T. Cooke, MD, Section of General Thoracic Surgery, Department of Surgery, University of California, Davis Health, 2335 Stockton Blvd, North Addition Office Building, Ste 6121, Sacramento, CA 95817 (dtcooke@ucdavis.edu).

Published Online: July 1, 2020. doi:10.1001/jamasurg.2020.1915

Author Contributions: Dr Cooke had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Clark, Huynh, Brown, Cooke.

Acquisition, analysis, or interpretation of data: Mustoe, Clark, Huynh, Tong, Wolf, Cooke.

Drafting of the manuscript: Mustoe, Clark, Cooke.

Critical revision of the manuscript for important intellectual content: Clark, Huynh, Tong, Wolf, Brown, Cooke.

Statistical analysis: Clark, Huynh, Cooke.

Administrative, technical, or material support: Tong, Wolf, Cooke.

Supervision: Clark, Tong, Brown, Cooke.

Conflict of Interest Disclosures: Ms Mustoe reported receiving personal fees from a Medical Student Research Fellowship, UC Davis School of Medicine, and an American Association for Thoracic Surgery Summer Intern Scholarship in Cardiothoracic Surgery during the conduct of the study. Dr Tong and Ms Wolf reported receiving grants from National Cancer Institute during the conduct of the study. Dr Brown reported receiving grants from National Center for Advancing Translational Sciences, National Institutes of Health during the conduct of the study. No other disclosures were reported.

Funding/Support: Dr Tong and Ms Wolf were supported by award P30CA093373-15S from the National Cancer Institute of the National Institutes of Health.

Role of the Funder/Sponsor: The funding source had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Disclaimer: The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.