Effect of Sustained Smoking Cessation Counseling and Provision of Medication vs Shorter-term Counseling and Medication Advice on Smoking Abstinence in Patients Recently Diagnosed With Cancer

Key Points

Question

How effective is sustained cessation telephone counseling plus medication treatment compared with shorter-term telephone counseling plus medication advice for smokers recently diagnosed with cancer?

Findings

In this randomized clinical trial of 303 patients, conducted at 2 National Cancer Institute–designated Comprehensive Cancer Centers, sustained telephone counseling over 6 months and provision of free medication compared with 4-week telephone counseling and medication advice resulted in 6-month biochemically confirmed quit rates of 34.5% vs 21.5%, a difference that was statistically significant.

Meaning

Among patients recently diagnosed with cancer, sustained cessation counseling and free medication may increase the likelihood of smoking abstinence compared with shorter-term counseling, but the generalizability of the study findings is uncertain and requires further research.

Abstract

Importance

Persistent smoking may cause adverse outcomes among patients with cancer. Many cancer centers have not fully implemented evidence-based tobacco treatment into routine care.

Objective

To determine the effectiveness of sustained telephone counseling and medication (intensive treatment) compared with shorter-term telephone counseling and medication advice (standard treatment) to assist patients recently diagnosed with cancer to quit smoking.

Design, Setting, and Participants

This unblinded randomized clinical trial was conducted at Massachusetts General Hospital/Dana-Farber/Harvard Cancer Center and Memorial Sloan Kettering Cancer Center. Adults who had smoked 1 cigarette or more within 30 days, spoke English or Spanish, and had recently diagnosed breast, gastrointestinal, genitourinary, gynecological, head and neck, lung, lymphoma, or melanoma cancers were eligible. Enrollment occurred between November 2013 and July 2017; assessments were completed by the end of February 2018.

Interventions

Participants randomized to the intensive treatment (n = 153) and the standard treatment (n = 150) received 4 weekly telephone counseling sessions and medication advice. The intensive treatment group also received 4 biweekly and 3 monthly telephone counseling sessions and choice of Food and Drug Administration–approved cessation medication (nicotine replacement therapy, bupropion, or varenicline).

Main Outcome and Measures

The primary outcome was biochemically confirmed 7-day point prevalence tobacco abstinence at 6-month follow-up. Secondary outcomes were treatment utilization rates.

Results

Among 303 patients who were randomized (mean age, 58.3 years; 170 women [56.1%]), 221 (78.1%) completed the trial. Six-month biochemically confirmed quit rates were 34.5% (n = 51 in the intensive treatment group) vs 21.5% (n = 29 in the standard treatment group) (difference, 13.0% [95% CI, 3.0%-23.3%]; odds ratio, 1.92 [95% CI, 1.13-3.27]; P < .02). The median number of counseling sessions completed was 8 (interquartile range, 4-11) in the intensive treatment group. A total of 97 intensive treatment participants (77.0%) vs 68 standard treatment participants (59.1%) reported cessation medication use (difference, 17.9% [95% CI, 6.3%-29.5%]; odds ratio, 2.31 [95% CI, 1.32-4.04]; P = .003). The most common adverse events in the intensive treatment and standard treatment groups, respectively, were nausea (n = 13 and n = 6), rash (n = 4 and n = 1), hiccups (n = 4 and n = 1), mouth irritation (n = 4 and n = 0), difficulty sleeping (n = 3 and n = 2), and vivid dreams (n = 3 and n = 2).

Conclusions and Relevance

Among smokers recently diagnosed with cancer in 2 National Cancer Institute–designated Comprehensive Cancer Centers, sustained counseling and provision of free cessation medication compared with 4-week counseling and medication advice resulted in higher 6-month biochemically confirmed quit rates. However, the generalizability of the study findings is uncertain and requires further research.

Trial Registration

ClinicalTrials.gov Identifier: NCT01871506

This randomized trial compares the effect of sustained telephone counseling and medication vs shorter-term telephone counseling and medication advice on 6-month smoking abstinence among patients at academic cancer hospitals recently diagnosed with cancer.

Introduction

Continued smoking after a cancer diagnosis can increase overall and cancer-specific mortality, risk for second primary cancers, treatment complications, and adverse effects.^1,2,3 However, data from 2000 to 2011 estimated that 10% to 30% of patients with cancer continued to smoke after a diagnosis,^2,4 and data from 2005 to 2007 estimated that half who quit quickly relapsed.⁵ The likelihood of cessation and its benefits are highest closer to the time of diagnosis,⁶ yet patients with cancer often do not get the assistance they need to quit.^7,8

National organizations have underscored the need to include tobacco treatment as an integral component of comprehensive cancer care. The American Association for Cancer Research policy emphasized that patients with cancer from all clinical settings be provided with evidence-based tobacco cessation assistance.⁹ The American Society of Clinical Oncology emphasized incorporating tobacco assistance into clinical practice,¹⁰ and the National Cancer Institute Moonshot initiative identified cancer treatment as a critical opportunity to provide tobacco treatment.³ The National Comprehensive Cancer Network (NCCN) published practice guidelines emphasizing combined behavioral and medication treatment.¹¹ Despite these mandates, there remains a need to identify the most effective evidence-based approach to deliver tobacco treatment integrated into cancer care.¹² A recent meta-analysis of smoking cessation intervention trials (1993-2018) for cancer survivors demonstrated that the optimal treatment has not been identified.¹³

This randomized clinical trial compared the effectiveness of 2 tobacco treatments (intensive treatment: sustained telephone counseling and free Food and Drug Administration [FDA]–approved smoking cessation medication vs standard treatment: shorter-term telephone counseling plus smoking cessation medication advice), integrated into cancer care at diagnosis, in promoting tobacco abstinence at 6 months. It was hypothesized that, compared with standard treatment, the intensive treatment would increase the proportion of smokers with biochemically confirmed, 7-day point prevalence tobacco abstinence at 6 months.

Methods

The Smokefree Support Study was a multisite, unblinded, randomized clinical trial that compared the effectiveness of 2 evidence-based tobacco treatments in promoting abstinence among adults recently diagnosed (within 4 visits or 3 months) with cancer. The study protocol and statistical analysis plan are provided in Supplement 1.¹⁴ The study was guided by theoretical frameworks for understanding chronic illness management. The Self-Regulation Model incorporates the roles of emotions, cancer beliefs, physical experiences, and environmental influences, and the Health Belief Model focuses on smoking-related beliefs.¹⁵ Study procedures were approved by the institutional review boards of the participating sites. All participants provided written informed consent. A data and safety monitoring board met annually to monitor patient safety and review potential adverse events; no study-related serious adverse events were identified.

Setting and Participants

Participants were recruited from 2 National Cancer Institute–designated Comprehensive Cancer Centers (Massachusetts General Hospital/Dana-Farber/Harvard Cancer Center and Memorial Sloan Kettering Cancer Center). Adults who were identified as current smokers (self-reported smoking ≥1 cigarette in the past 30 days); undergoing cancer treatment for a recent diagnosis of breast, gastrointestinal, genitourinary, gynecological, head and neck, lymphoma, lung, or melanoma cancer; had telephone access; and were English or Spanish speaking (at 1 site) were eligible for study participation. Exclusion criteria were medical and cognitive impairment deemed likely to interfere with study participation (determined by electronic health record [EHR] and/or treating clinician), uncontrolled psychosis or suicide attempt in the past year, insufficient comprehension/literacy (determined by EHR and/or treating clinician), and not receiving cancer care at a study site. Research staff screened cancer clinic schedules to identify and contact potentially eligible patients, confirmed eligibility, and obtained patient informed consent.

Assignment to Treatment Condition

Participants were randomly assigned to a treatment group using a computer-generated schedule of randomization IDs. Treatment assignments were in blocks of 6, stratified by study site and cancer center clinic, and were generated by the study database manager. Patients completed the baseline assessment and the initial counseling session; treatment group was disclosed at the end of this session. Study staff were not blinded to treatment group.

Description of the Tobacco Treatment Conditions

Telephone counseling was selected as the delivery modality to reduce the burden of additional in-person visits for patients undergoing cancer treatment. However, patients could opt to conduct the initial counseling session at enrollment in person.

Standard Treatment

Participants were offered 4 weekly telephone counseling sessions plus education and advice regarding cessation medications. At the first session, tobacco counselors used a decision aid, designed for patients with cancer, to make medication recommendations.¹⁴

Intensive Treatment

Participants were offered 4 weekly telephone counseling sessions, 4 biweekly telephone sessions delivered over 2 months, and 3 telephone booster sessions delivered monthly. The number and length of sessions were based on pilot work with patients from the Massachusetts General Hospital thoracic oncology clinic.¹⁶ Participants in this group were also offered their choice of 12 weeks of FDA-approved smoking cessation medication at no cost; they were not required to use any medication. The medication selected by participants was appended as a prescription in the EHR and dispensed (in person or mailed) according to Public Health Service guidelines.¹⁷ Participants received an initial 4-week supply of FDA-approved smoking cessation medication (varenicline, bupropion sustained release, single or combination nicotine replacement therapy patch and/or lozenges) with the option of receiving up to 2 additional 4-week supplies.

Counseling Content

Telephone counseling was delivered by 1 of 5 certified tobacco treatment counselors using a treatment protocol guided by a Motivational Interviewing style with a focus on cognitive behavioral and stress management and resiliency strategies.¹⁴ The same tobacco treatment counselors were used for both groups. Each session was structured according to the patient’s quit readiness, cancer treatment, and challenges. Each follow-up session included an assessment of cessation progress, use any of FDA smoking cessation medication, a cancer care treatment review, an assessment of quit motivation and quit importance, and cancer-related content (eg, smoking-associated stigma, stress management, social support, symptom management).¹⁴ All cessation medication adverse effects were monitored at each session. Counselors would provide quitline information for family members who wanted quit support.

Treatment Adherence

The study principal investigators (licensed psychologists with expertise in tobacco treatment and psycho-oncology) met weekly with the tobacco counselors to review counseling sessions. All counseling sessions were audiotaped; 1 session per site was randomly selected each week to monitor adherence to treatment protocol and Motivational Interviewing style using a trial-specific checklist.¹⁴

Outcomes

Primary Outcome

The primary study outcome was biochemically confirmed 7-day point prevalence cigarette smoking abstinence at 6-month follow-up. Participants who reported smoking abstinence at either 3 or 6 months were mailed a saliva collection kit to return. Abstinence was defined as cotinine level less than 15 ng/mL.¹⁸ Participants who self-reported smoking abstinence and were still using nicotine replacement therapy or nicotine-containing e-cigarettes provided an in-person expired air carbon monoxide sample (abstinence defined as <10 particles per million expired air carbon monoxide).¹⁸ This collection process occurred separately from the counseling visit (in-person by the study research assistant).

Secondary Outcomes

Secondary smoking outcomes included biochemically confirmed past 7-day abstinence at 3 months, self-reported 7-day point prevalence cigarette smoking abstinence at 3 and 6 months, 24-hour intentional quit attempt (yes vs no) at 3 and 6 months, continuous abstinence (self-reported 3-month abstinence between quit and follow-up at 3 and 6 months) at 3 and 6 months, and sustained abstinence (biochemically confirmed repeated point prevalence abstinence at 3 and 6 months).

Exploratory Treatment Utilization Outcomes

Exploratory outcomes included the number of counseling sessions completed, the percentage of participants who received study cessation medication, and self-reported use of cessation medication.

Measures and Assessments

Sociodemographic information was collected from the baseline survey and EHR. Adhering to national mandates¹⁹ underscoring the importance of including racial/ethnic minority and underrepresented patients into clinical trials, race/ethnicity information was collected based on participant self-reported survey responses to categories; an “other” category was reviewed with the study team and recoded based on National Institutes of Health–defined categories.²⁰ Cancer history information was collected from the EHR. Cancer diagnoses were categorized as smoking related and nonsmoking related.¹ All follow-up survey data were collected by mail, electronically (Research Electronic Data Capture [REDCap]), or via telephone at 3 and 6 months. Participants were remunerated $20 for the baseline survey, $40 for each follow-up survey, and $50 for each saliva or expired air sample provided.

Smoking History

Questions included the number of cigarettes smoked per day, years smoked, changes in cigarettes per day over the past 6 months, e-cigarette use, past 24-hour quit attempts, prior smoking cessation medication use, and prior use of behavioral and alternative methods to help quit smoking. Nicotine dependence was evaluated using the 2-item Heaviness of Smoking Index.²¹ Motivation to quit smoking was assessed with a 1-item, 10-point contemplation ladder.²²

Health Belief Model Constructs

Smoking Beliefs

Quit importance was rated on a 1-item analogue scale.²³ Five analogue scale items assessed participants’ perceived benefits of quitting smoking.²³ Self-efficacy to not smoke/resist smoking urges was assessed with an 11-item measure.²⁴

Self-regulation Model Constructs

Emotions

Emotional distress within the past 2 weeks was measured using a 1-item analogue scale.²⁵ Anxiety and depression symptoms within the past 2 weeks were assessed using the General Anxiety Disorder Scale 7²⁶ and the Patient Health Questionnaire 9.²⁷ The 4-item Perceived Stress Scale 4 was used to measure appraisal of stress.²⁸ A 1-item analogue scale assessed ability to manage stress.

Cancer Beliefs

The 5-item Internalized Shame Scale measured the extent to which shame is internalized in relation to illness.²⁹

Physical Symptoms

Cancer symptom severity over the past 2 weeks was assessed with 3 1-item analogue scale items for pain, nausea, and fatigue. Sleep quality during the past month was assessed with a 1-item measure.³⁰ The single-item Mood and Physical Symptom Scale was used to measure participants’ urges to smoke in the past 24 hours.³¹

Environmental Influences

Level of second-hand smoke exposure was assessed by the presence of additional smokers in the household and home smoking rules.³² Patient perception of social support was measured using 8 items of the Partner Interaction Questionnaire.³³

Treatment Satisfaction

Tobacco treatment satisfaction was assessed at follow-up surveys; participants rated the assistance that they received, alignment of support with their needs, helpfulness of the treatment, and the quality of assistance received.

Treatment Utilization

The number of counseling sessions completed and cessation medications (intensive treatment group) received were tracked. Participants reported any use of FDA-approved smoking cessation medications at 3- and 6-month surveys.

Analysis

A sample of 296 was planned, assuming a 10% death rate, to provide 80% power at .05 2-sided significance level to detect a 15% difference in the primary outcome. This group difference was based on biochemically confirmed group difference rates from the pilot study conducted with thoracic oncology patients that compared an intensive treatment vs a no treatment control.¹⁶ All analyses were conducted with a 2-sided significance level, with a threshold of P < .05. The Wald confidence interval is reported for all differences and 95% CIs around the difference. We examined the frequency distributions of all baseline variables and compared baseline characteristics of participants in the 2 treatment groups using χ² tests for categorical variables and Wilcoxon rank-sum tests for continuous variables. Participants who did not complete a follow-up survey or who self-reported abstinence but did not provide sufficient biochemical verification were considered smokers. Participants who were deceased or medically ineligible (eg, in inpatient hospice or psychiatrically impaired) at follow-up were not included in the final outcome analyses. SAS version 9.4 (SAS Institute) was used for all analyses.

We compared cessation outcomes by treatment group using χ² tests. Because of the potential for type I error due to multiple comparisons, findings for analyses of secondary end points should be interpreted as exploratory. Adjusting for treatment group, we assessed baseline moderators of treatment effect on the primary outcome. We also analyzed the effect changes in the Health Belief and Self-Regulation Model intervention constructs on the primary outcome (change from baseline to 6 months); changes in categorical variables were converted to an ordinal scale to denote decrements or improvements. Variables with a P value less than .10 in the bivariable tests were included in a multivariable generalized linear mixed-effects model with site included as a random effect.

We conducted logistic regression models to compare medication use (yes or no for any use reported at 3 months and 3 or 6 months) and initial session delivery modality (telephone vs in person) between groups. Within the intensive treatment group, we conducted logistic regression models to assess counseling use (number of sessions, booster sessions, and medication use) with the primary outcomes. We conducted logistic regression models to compare program satisfaction responses at 3- and 6-month follow-up between groups.

Missing Data

To verify that participants with missing outcomes did not bias the conclusion about the relationship between group and the primary outcome, we conducted sensitivity analyses with multiple imputations for the primary outcome variable. Missing items on scale measures, that were approximately 80% complete, were completed with the mean score from available items; remaining missing data were then imputed for each scale using a 2-step process. In the first step, we applied the Markov chain Monte Carlo method to generate a monotone missing data pattern for values across the 3 time points (baseline, 3 months, and 6 months).³⁴ In the second step, a standard monotone regression pattern of observations over time was used to impute all other missing values for each scale. Patient satisfaction was conducted with the entire sample; to adjust for missing data, we conducted the same 2-step imputation process as for the missing scales (over the 3-month and 6-month surveys). Results for multiply imputed data sets were analyzed according to standard methodology.³⁴ All other analyses were complete case analyses.

Results

Recruitment and Retention

Trial enrollment began in November 2013, and follow-up data collection ended in February 2018. A total of 4709 patients were screened for eligibility, and 2050 were not eligible (Figure). There were 2659 patients who met the initial EHR screening criteria (adult, current smoker, cancer diagnosis); 1808 refused the eligibility confirmation screen. Of the 851 who completed the study screen, 405 were ineligible (47.6%), 143 (16.8%) declined, and 303 were randomized (35.6%). Follow-up survey completion rates at 6 months were 81.8% (n = 121/148) in the intensive treatment group and 74.1% (n = 100/135) in the standard treatment group (difference, 7.7% [95% CI, −2.0% to 17.3%]; odds ratio [OR], 1.56 [95% CI, 0.89-2.77]; P = .12).

Figure. Flowchart of Enrollment and Intervention to Test the Effectiveness of 2 Models of Tobacco Treatment Integrated Into Cancer Care.

MSK indicates Memorial Sloan Kettering Cancer Center.

^aPatients could give multiple reasons for refusal. The research assistant categorized the reasons patients offered according to options on the study screening tool; reasons that did not fit into 1 of these predefined categories were discussed with the team to determine fit with existing categories or establishment of new categories.

^bMultiple reasons for ineligibility could have been indicated on the screener.

^cThose who were never randomized signed a consent form but did not complete a counseling session. Reasons included participants who were not able to be reached by the study counselor, participants who withdrew citing other cancer care demands, and participants who became ineligible over time.

^dThose who did not complete the follow-up survey include those who could not be reached at all to complete the follow-up assessment.

^eFollow-up survey completion rate = completed/completed + refused + withdrew + did not complete follow-up survey. Participants who were deceased or medically ineligible (eg, in inpatient hospice or psychiatrically impaired) at follow-up were not included in the final outcome analyses (n = 5 in the intensive treatment group and n = 15 in the standard treatment group). Thus, for the intensive treatment group, the denominator is 148, and for the standard treatment group, the denominator is 135.

Among participants confirmed eligible, there were no significant differences in sociodemographic characteristics among those randomized (n = 303) vs those who declined (eg, eligible but refused consent [n = 84] or were never randomized [n = 59]). Among participants who enrolled, those who identified as Hispanic or race other than White were significantly less likely to be randomized (complete initial counseling session) than those who identified as non-Hispanic White (n = 47/61 [77.1%] vs 253/288 [87.8%]; difference, −10.8% [95% CI, −22.0% to 0.4%]; OR, 0.46 [95% CI,0.23-0.93]; P = .03), and patients from thoracic and head and neck cancer center clinics were significantly less likely to be randomized compared with patients from the other clinics (n = 121/153 [79.1%] vs n = 182/209 [87.1%]; difference, −8.0% [95% CI, −15.9% to −0.1%]; OR, 1.78 [95% CI, 1.02-3.13]; P = .04).

Participants who did not complete the 3-month follow-up survey were significantly more likely to be younger (median age, 57 years [interquartile range {IQR}, 50-63] vs 59 years [IQR, 52-65]; difference, −3.25 [95% CI, −5.86 to −0.65]; OR, 1.03 [95% CI, 1.01-1.06]; P = .02), have private insurance vs Medicaid/Medicare (n = 48/171 [28.1%] vs n = 16/103 [15.5%]; difference, 12.5% [95% CI, 2.8%-22.3%]; OR, 2.03 [95% CI, 1.08-3.83]; P = .03), and have later stages of cancer vs earlier stages (n = 31/98 [31.6%] vs n = 35/176 [19.9%]; difference, 11.8% [95% CI, 0.8%-22.7%]; OR, 1.89 [95% CI, 1.08-3.34]; P = .04). Participants who did not complete the 6-month follow-up survey were significantly more likely to be younger (median age, 57 years [IQR 51-63] vs 59 years [IQR 51-65]; difference, −3.05 [95% CI, −5.70 to −0.41]; OR, 1.03 [95% CI, 1.00-1.06]; P = .03) and have later stages of cancer vs earlier stages of cancer (n = 30/98 [30.6%] vs n = 33/176 [18.8%]; difference, 11.9% [1.1%-22.7%]; OR, 2.19 [95% CI, 1.29-3.71]; P = .04).

Baseline Characteristics

The median age of participants was 59 years (IQR, 52-65); 170 participants (56.1%) were female, 255 (84.2%) were non-Hispanic White, and 93 (31.5%) had a high school education or less (Table 1). In terms of cancer characteristics, 181 (59.7%) had a smoking-related tumor, and 182 (60.1%) had early-stage disease. Thirty-one participants (10.2%) reported having a serious mental illness (major depression, bipolar disorder, or schizophrenia) (Table 2). The median number of cigarettes per day was 10 (IQR, 4-20), and 210 participants (69.3%) had decreased the number cigarettes smoked per day within the past 6 months. The median number of years smoked was 42 (IQR, 36-49), and 214 participants (72.1%) smoked within 30 minutes of waking. In terms of smoking environment, 151 patients (51.0%) allowed smoking in their homes.

Table 1. Sociodemographic and Environmental Characteristics at Baseline.

Characteristic	No. (%)
	Intensive treatment (n = 153)	Standard treatment (n = 150)
Age, median (IQR), y	59.0 (52.0-65.0)	57.0 (52.0-65.0)
Sex
Male	66 (43.1)	67 (44.7)
Female	87 (56.9)	83 (55.3)
Race and ethnicitya
American Indian	2 (1.3)	1 (0.7)
Asian	1 (0.7)	1 (0.7)
Black non-Hispanic	14 (9.2)	16 (10.67)
Black Hispanic	0	1 (0.7)
White non-Hispanic	127 (83.0)	128 (85.3)
White Hispanic	7 (4.6)	3 (2.7)
Other race/ethnicity	2 (1.3)	0
Education	n = 148	n = 147
High school graduate or less	50 (33.8)	43 (29.3)
Some college/vocational school	61 (41.2)	58 (39.5)
College graduate or more	37 (25.0)	46 (31.3)
Marital status	n = 148	n = 147
Married, living as married	89 (58.2)	75 (51.0)
Single, never married	19 (12.4)	19 (12.9)
Divorced or separated	29 (19.0)	46 (31.3)
Widowed	11 (7.2)	7 (4.8)
Employment status	n = 149	n = 147
Employed (full-time or part-time)	57 (38.3)	73 (49.7)
Unemployed/retired	56 (37.6)	44 (29.9)
Unable to work	36 (24.2)	30 (20.4)
Health insurance	n = 147	n = 146
Employer-sponsored/military	75 (51.0)	89 (61.0)
Medicaid	33 (22.4)	24 (16.4)
Medicare	29 (19.7)	26 (17.8)
Individual (self-purchased)	10 (6.8)	7 (4.8)
Additional smokers in household (among participants living with others in the household)	52/117 (44.4)	41/104 (39.4)
Home smoking rule	n = 150	n = 146
No one is allowed to smoke anywhere	80 (53.3)	65 (44.5)
Smoking permitted in some places/times	41 (27.3)	36 (24.7)
Smoking is permitted anywhere	29 (19.3)	45 (30.8)
Partner quitting support (PIQ) (range, 0-16)b
No.	133	131
Median (IQR)	1.3 (1.0-1.8)	1.33 (1.1-2.0)

Abbreviations: IQR, interquartile range; PIQ, Partner Interaction Questionnaire.

^aData for race were collected using prespecified categories of American Indian or Alaska Native, Asian, Black or African American, Native Hawaiian or other Pacific Islander, White, and other. If other was selected, participants were asked to specify. Participants were asked to check all categories that applied. Data for race were collected using prespecified categories. Ethnicity data were collected separately from race data, thus creating the overlap between Hispanic ethnicity and race.

^bPIQ measure adapted from Partner Interaction Questionnaire (range, 0-16).³³ Composite positive behavior items with a range of 0 to 4 divided by negative behavior items with a range of 0 to 4. Higher scores indicate more positive and less negative partner behaviors regarding smoking.

Table 2. Cancer, Emotional, Physical, and Smoking Characteristics at Baseline.

	No. (%)
	Intensive treatment (n = 153)	Standard treatment (n = 150)
Cancer center clinic
Thoracic	46 (30.1)	47 (31.3)
Breast	38 (24.8)	39 (26.0)
Genitourinary	25 (16.3)	26 (17.3)
Head and neck	18 (11.8)	13 (8.7)
Gastrointestinal	15 (9.8)	14 (9.3)
Lymphoma	4 (2.6)	5 (3.3)
Gynecological	3 (2.0)	4 (2.7)
Melanoma	4 (2.6)	2 (1.3)
Smoking-related tumora	92 (60.1)	89 (59.3)
Stage
0-II	90 (58.8)	92 (61.3)
III-IV	56 (36.6)	55 (36.7)
Otherb	7 (4.6)	3 (2.0)
Comorbid smoking-related diseasec	72 (47.1)	76 (50.7)
Serious mental illness (major depression, bipolar disorder, schizophrenia)	16 (10.5)	15 (10.0)
Emotional distress (range, 0-10)d
No.	147	147
Median (IQR)	8.0 (5.0-10.0)	8.0 (5.0-9.0)
Depression (PHQ-9 score >10) (range, 0-36)e	55/148 (37.2)	59/147 (40.1)
Anxiety (GAD-7 score >10) (range, 0-21)f	70/149 (47.0)	72/145 (50.0)
Perceived stress (PSS-4) (range, 0-16)g
No.	147	147
Median (IQR)	4.0 (3.0-7.0)	5.0 (3.0-8.0)
Smoking urges (MPSS) (range, 0-5)h
No.	146	146
Median (IQR)	3.0 (2.0-3.0)	3.0 (2.0-3.0)
Pain (range, 0-10)i
No.	147	148
Median (IQR)	4.0 (1.0-7.0)	4.0 (0.5-4.0)
Nausea (range, 0-10)i
No.	147	148
Median (IQR)	0 (0.0-3.0)	0 (0.0-2.0)
Fatigue (range, 0-10)i
No.	147	147
Median (IQR)	7.0 (5.0-8.0)	7.0 (4.0-8.0)
Sleep quality (range, 1-4)j
No.	148	146
Median (IQR)	3.0 (2.0-3.0)	3.0 (2.0-3.0)
Smoking behaviors and beliefs and cancer beliefs
Cigarettes per day
No.	152	149
Median (IQR)	10 (4-18)	10 (4-20)
Made changes to cigarettes per day in past 6 mo	107 (69.9)	103 (68.7)
Years smoked, median (IQR)	43 (36.0-49.0)	41 (36.0-49.0)
Time to first cigarette (≤30 min after waking)	108/151 (71.5)	106/146 (72.6)
Heaviness of Smoking Index (range, 0-6)k
No.
Median (IQR)	3 (2-3)	3 (2-3)
Ever made intentional 24-h quit attempt	138 (90.2)	136 (90.7)
e-Cigarette use in past 30 d	27/153 (17.7)	29/149 (19.5)
Quit motivation (range, 0-9)l
No.	141	134
Median (IQR)	6.0 (4.0-7.0)	6.0 (5.0-7.0)
Perceived benefits of quitting (range, 0-10)m
No.	144	140
Median (IQR)	9.4 (7.9-10.0)	9.8 (7.9-10.0)
Importance of quitting (range, 0-10)n
No.	148	145
Median (IQR)	10.0 (10.0-10.0)	10.0 (10.0-10.0)
Self-efficacy to resist smoking/urges (range, 0-10)o
No.	149	146
Median (IQR)	5.4 (4.0-7.2)	5.4 (3.9-7.0)
Internalized shame (range, 5-20)p
No.	147	146
Median (IQR)	10.0 (8.0-13.0)	10.0 (7.0-13.0)

Abbreviations: GAD-7, General Anxiety Disorder 7; IQR, interquartile range; MPSS, Mood and Physical Symptoms Scale; PHQ-9, Patient Health Questionnaire 9; PSS-4: Perceived Stress Scale 4.

^aPercentage of participants with a cancer known to be caused by smoking (ie, anal, bladder, cervical, colorectal, esophageal, gastric, head and neck, kidney, liver, lung, pancreatic, and small intestine).¹

^bStage unknown or no diagnosis; indolent and aggressive lymphoma coded as 0-II and III-IV, respectively.

^cSmoking-related diseases include emphysema/chronic obstructive pulmonary disease, stroke, hypertension, and myocardial infarction.

^dThe Distress Thermometer (range, 0-10)²⁵ is a single-item scale. 0 indicates no emotional distress and 10 indicates extreme emotional distress.

^eThe PHQ-9 (range, 0-36)²⁷ has a sum score of 9 individual items with ranges of 0 to 3. Higher scores indicate more depressive symptoms. Dichotomized P value based on the clinical cutoff for depression of 10 or greater. A score of 10 represents moderate depressive symptoms.

^fGAD-7 (range, 0-21)²⁶ has a sum score of 7 individual items with ranges of 0 to 3. Higher scores indicate more symptoms of anxiety. Dichotomized P value based on the clinical cutoff for anxiety of 10 or greater. A score of 10 represents moderate levels of anxiety symptoms.

^gThe PSS-4 (range, 0-16)²⁸ has a sum score of 4 individual items with ranges of 0 to 4; 2 items are reverse scored. Higher scores indicate higher stress.

^hThe MPSS (range, 0-5)³¹ is a single-item scale. 0 indicates no urge to smoke in the past 24 hours and 5 indicates an urge to smoke in the past 24 hours that is felt all the time.

ⁱThree separate measures of pain, nausea, and fatigue. For pain scale, 0 indicates no pain and 10 indicates extreme pain. For nausea scale, 0 indicates no nausea and 10 indicates extreme nausea. For fatigue scale, 0 indicates no fatigue and 10 indicates extreme fatigue.

^jSleep quality measure adapted from the Pittsburgh Sleep Quality Index (range, 1-4).³⁰ It is a single-item scale; 1 indicates very bad sleep quality and 4 indicates very good sleep quality.

^kHeaviness of Smoking Index ranges from 0 to 6. Scale scores of 0-1 indicate low levels of addiction to smoking, scores of 2-4 represent moderate levels of addiction, and scores of 5-6 represent high levels of addiction.²¹

^lThe Biener Scale (range, 0-9)²² is a single-item scale. Higher scores indicate a greater motivation to quit smoking. Participants are asked to select the statement that best represents their thinking at the time of the survey, ranging from 1: “I enjoy smoking so much that I will never consider quitting no matter what happens” to 9: “I have quit and I am 100% confident that I will never smoke again.”

^mThe Benefits of Quitting Scale ranges from 0 to 10.²³ Five items with response options ranging from 0 to 10 are added and averaged to create an overall score. Higher scores indicate a greater endorsement of the perceived health benefits of quitting smoking.

ⁿThe Importance of Quitting Scale (range, 0-10)²³ is a single-item scale. Higher scores indicate a greater endorsement of the importance of quitting smoking.

^oThe Quit Self-Efficacy Questionnaire ranges from 0 to 10.²⁴ Eleven individual items with response options ranging from 0 to 10 are added and averaged to create overall score. Higher scores indicate a greater endorsement of the ability to resist the urge to smoke in different situations.

^pThe Internalized Shame from Lung Cancer Stigma Scale (range, 5-20)²⁹ has a sum score of 5 individual items with ranges of 1 to 4 (5 is not applicable). Higher scores indicate greater illness-related shame.

In terms of smoking beliefs, perceived benefits and importance of quitting were high, with median scores of 9.5 (IQR, 7.9-10) and 10 (IQR, 10-10), respectively (Table 2). Quitting motivation and self-efficacy to not smoke/resist urges were moderate, with median scores of 6 (IQR, 4-7) and 5.4 (IQR, 4-7), respectively. In terms of emotional constructs, emotional distress was prevalent and relatively high, with participants reporting a median of 8 (IQR, 5-9); 114 patients (38.6%) and 142 patients (48.3%) reported elevated depression and anxiety, respectively. In terms of cancer beliefs, internalized shame was moderate, with a median score of 10 (IQR, 7-13).

Prior experience with behavioral support for cessation was limited (eTable 1 in Supplement 2). Only 39 participants (12.9%) reported prior use of counseling, and only 22 participants (7.3%) had used a quitline. A total of 239 participants (79.4%) had used a cessation medication: 187 (62.1%) had used nicotine replacement therapy patches, 102 (33.9%) had used varenicline, 90 (29.9%) had used bupropion, and 42 (14.0%) had used nicotine replacement therapy lozenges.

Smoking Cessation Outcomes

Sensitivity analyses with multiple imputations for the primary outcome verified that participants with missing outcomes did not bias the findings between group and the primary outcome.

Primary Outcome

Biochemically confirmed 7-day point prevalence abstinence at 6-month follow-up was 34.5% (n = 51/148) for the intensive treatment group vs 21.5% (n = 29/135) for the standard treatment group (difference, 13.0% [95% CI, 3.0%-23.3%]; OR, 1.92 [95% CI, 1.13-3.27]; P < .02) (Table 3; eFigure in Supplement 2). In the intensive treatment group, 76.4% provided a cotinine sample and in the standard treatment group, 72.3% provided a cotinine sample.

Table 3. Primary and Secondary Smoking Outcome Comparisons Between Treatment Groups.

	No. (%)		Absolute difference, % (95% CI)	Odds ratio (95% CI)	P valuea
	Intensive treatment (n = 148)	Standard treatment (n = 135)
Primary outcome
Biochemically confirmed 7-d point prevalence abstinence at 6 mo	51 (34.5)	29 (21.5)	13.0 (3.0 to 23.3)	1.92 (1.13 to 3.27)	.02
Secondary outcomes
Biochemically confirmed 7-d point prevalence abstinence at 3 mo	46 (31.1)	28 (20.7)	10.3 (0.2 to 20.5)	1.72 (1.00 to 2.96)	.048
Self-reported 7-d point prevalence abstinence
At 3 mo	57 (38.5)	38 (28.1)	10.4 (−0.5 to 21.3)	1.60 (0.97 to 2.64)	.07
At 6 mo	54 (36.5)	42 (31.1)	5.4 (−5.6 to 16.4)	1.27 (0.78-2.09)	.34
24-h Intentional quit attempt
At 3 mo	91 (61.5)	80 (59.3)	2.2 (−9.2 to 13.6)	1.10 (0.68 to 1.77)	.70
At 6 mo	98 (66.2)	82 (60.7)	5.5 (−5.8 to 16.7)	1.27 (0.78 to 2.06)	.34
Continuous abstinenceb
At 3 mo	29 (19.6)	23 (17.0)	2.6 (−6.5 to 11.6)	1.19 (0.65 to 2.17)	.58
At 6 mo	39 (26.4)	28 (20.7)	5.6 (−4.3 to 15.5)	1.37 (0.79 to 2.38)	.27
Sustained abstinence at 6 moc	35 (23.7)	17 (12.6)	11.1 (2.2 to 19.9)	2.15 (1.14 to 4.05)	.02

^aThe P value represents the significance of the variable in the χ² test.

^bContinuous abstinence was defined as self-reported 3-month abstinence between quit and follow-up at 3 and 6 months.

^cSustained abstinence was defined as biochemically confirmed repeated point prevalence abstinence at 3 and 6 months.

Secondary Outcomes

Biochemically confirmed 7-day abstinence at 3-month follow-up was 31.1% (46/148) for the intensive treatment group vs 20.7% (n = 28/135) for the standard treatment group (difference, 10.3% [95% CI, 0.2%-20.5%]; OR, 1.72 [95% CI, 1.00-2.96]; P = .048) (Table 3). In the intensive treatment group, 72.6% provided a cotinine sample, and in the standard treatment group, 50.0% provided a cotinine sample. Among those biochemically confirmed abstinent at 3 months, 29.7% (n = 22/74) relapsed between 3 and 6 months. Among those biochemically confirmed abstinent at 6 months, 35.0% (n = 28/80) quit between 3 and 6 months.

At 3 months, 38.5% of patients (57/148) in the intensive treatment group self-reported 7-day abstinence vs 28.1% of patients (38/135) in the standard treatment group (difference, 10.4% [95% CI, −0.5% to 21.3%]; OR, 1.60 [95% CI, 0.97-2.64]; P = .07) (eFigure in Supplement 2). At 6 months, 36.5% of patients (54/148) in the intensive treatment group self-reported 7-day abstinence vs 31.1% of patients (42/135) in the standard treatment group (difference, 5.4% [95% CI, −5.6% to 16.4%]; OR, 1.27 [95% CI, 0.78-2.09]; P = .34). Rates of continuous abstinence and 24-hour quit attempts were not significantly different between treatment groups. Six-month sustained abstinence rates were 23.6% (35/148) in the intensive treatment group vs 12.6% (17/135) in the standard treatment group (difference, 11.1% [95% CI, 2.2%-19.9%]; OR, 2.15 [95% CI, 1.14-4.05]; P < .02).

Treatment Utilization Exploratory Outcomes

The median number of sessions completed was 4 (IQR, 3-4) for the standard treatment group and 8 (IQR, 4-11) for the intensive treatment group (eTable 2 in Supplement 2). The initial session was a mean (SD) of 44.61 (15.47) minutes in duration; follow-up sessions were a mean (SD) of 19.9 (8.1) minutes. By 6-month follow-up, 77.0% of participants (97/126) in the intensive treatment group vs 59.1% of participants (68/115) in the standard treatment group reported any cessation medication use (difference, 17.9% [95% CI, 6.3%-29.5%]; OR, 2.31 [95% CI, 1.32-4.04]; P = .003).

Among intensive treatment participants, at least 1 monthly booster counseling session was completed by 63.5% of participants (94/148), and 54.7% (81/148) completed at least 2 booster sessions (eTable 3 in Supplement 2). In terms of study medication, 79.7% of participants (122/153) in the intensive treatment group received any study medication; 57.4% (88/148) received 8 to 12 weeks of study medication. Of those who received study medication, 79.5% (101/127) selected nicotine replacement therapy. Among intensive treatment participants, a greater number of sessions completed was significantly associated with higher likelihood of quitting (difference, 3.10 [95% CI, 2.00-4.19]; OR, 1.31 [95% CI, 1.12-1.52]; P < .001) (eTable 4 in Supplement 2). A greater number of monthly booster sessions was significantly associated with a higher likelihood of quitting (difference, 1.19 [95% CI, 0.78-1.62]; OR, 2.14 [95% CI, 1.56-2.92]; P < .001). Smoking cessation medication use was not associated with likelihood of quitting.

Factors Associated With 6-Month Cotinine-Confirmed Abstinence

No baseline participant characteristics (Tables 1 and 2; sociodemographic cancer, medical or smoking history, intervention constructs) were associated with 6-month biochemically confirmed abstinence. In the multivariable model, intensive treatment (difference, 13.0% [95% CI, 3.0%-23.3%]; OR, 2.01 [95% CI, 1.04-3.87]; P = .04) and clinically meaningful increases in quit motivation (difference, 1.53 [95% CI, 0.90-2.17]; OR, 1.22 [95% 1.02-1.46]; P = .03), quit self-efficacy (difference, 2.27 [95% CI, 1.57-2.97]; OR, 1.27 [95% CI, 1.09-1.46]; P = .003), establishment of home smoking rules (difference, 10.1% [95% CI, −1.7% to 22.6%); OR, 0.59 [95% CI, 0.37-0.99]; P = .03), and decreases in anxiety (difference, −2.03 [95% CI, −3.47 to −0.61]; OR, 0.93 [95% CI, 0.87-1.00]; P = .04) were significantly associated with 6-month biochemically confirmed tobacco abstinence (Table 4).

Table 4. Factors Associated With 6-Month Cotinine-Confirmed Abstinence^a.

Variable	Patients who quit smoking at 6 mo (n = 80)	Patients who continued smoking at 6 mo (n = 203)	Difference of mean change (95% CI)	Odds ratio (95% CI)	P valueb
Treatment group (intensive [n = 148] and standard [n = 135]), %	51 (34.5) vs 29 (21.5)		13.0 (3.0 to 23.3)	2.01 (1.04 to 3.87)	.04
Quit motivation (mean change from baseline to 6 mo)c	2.04	0.51	1.53 (0.90 to 2.17)	1.22 (1.02 to 1.46)	.03
Self-efficacy to not smoke/resist urges (mean change from baseline to 6 mo)d	3.33	1.06	2.27 (1.57 to 2.97)	1.27 (1.09 to 1.46)	.003
Anxiety (GAD-7) (mean change from baseline to 6 mo)e	−3.07	−1.03	−2.03 (−3.47 to −0.61)	0.93 (0.87 to 1.00)	.04
Perceived stress (mean change from baseline to 6 mo)f	−1.15	−0.33	−0.82 (−1.66 to 0.20)	0.97 (0.85 to 1.11)	.69
Partner support (PIQ) (mean change from baseline to 6 mo)g	0.81	−0.07	0.89 (−0.05 to 1.82)	1.03 (0.92 to 1.15)	.56
Establishment of home smoking rule (% change from baseline to 6 mo)h	38.8	28.6	10.1 (−1.7 to 22.6)	0.59 (0.37 to 0.99)	.03
Ability to manage stress (mean change from baseline to 6 mo)i	1.01	−0.07	0.76 (−0.16 to 1.69)	0.99 (0.88 to 1.11)	.83
Smoking urges (mean change from baseline to 6 mo)j	−1.56	−0.62	−0.94 (−1.39 to −0.50)	0.87 (0.67 to 1.11)	.26
Internalized shame (mean change from baseline to 3 mo)k	−1.36	0.04	−1.39 (−2.50 to −0.29)	0.94 (0.86 to 1.02)	.15

Abbreviations: GAD-7, General Anxiety Disorder 7; MPSS, Mood and Physical Symptoms Scale; PIQ, Partner Interaction Questionnaire; PSS-4, Perceived Stress Scale 4.

^aMissing data were imputed using the Markov chain Monte Carlo procedure, and results are from a generalized linear mixed-effects model, with site include as a random effect.

^bThe P value represents the significance of the variable in the multivariable model.

^cThe Biener Scale (range, 0-9)²² change score is from baseline to 6-month follow-up. This is a single-item scale. Higher scores indicate a greater motivation to quit smoking. Participants are asked to select the statement that best represents their thinking at the time of the survey, ranging from 1: “I enjoy smoking so much that I will never consider quitting no matter what happens” to 9: “I have quit and I am 100% confident that I will never smoke again.”

^dThe Quit Self-Efficacy Questionnaire (range, 0-10)²⁴ change score is from baseline to 6-month follow-up. Eleven items with ranges of 0 to 10 are added and averaged to create an overall score. Higher scores indicate a greater endorsement of the ability to resist the urge to smoke in different situations.

^eThe GAD-7 (range, 0-21)²⁶ change score is from baseline to 6-month follow-up. The sum score is of 7 individual items with ranges of 0 to 3. Higher scores indicate more symptoms of anxiety. Dichotomized P value based on the clinical cutoff for anxiety of 10 or greater. A score of 10 represents moderate levels of anxiety symptoms.

^fThe PSS-4 (range, 0-16)²⁸ change score is from baseline to 6-month follow-up. The sum score is of 4 individual items with ranges of 0 to 4. The second and third items are reverse scored. Higher scores indicate higher levels of stress.

^gPIQ measure adapted from the Partner Interaction Questionnaire (range, 0-16)³³ change score from baseline to 6-month follow-up. Composite positive behavior items with a range of 0 to 4 divided by negative behavior items with a range of 0 to 4. Higher scores indicate more positive and less negative partner behaviors regarding smoking.

^hPercentage of participants who advanced in establishing home smoking rule from baseline to 6-month follow-up.³²

ⁱOne-item scale (range, 0-10) change score is from baseline to 6-month follow-up. Higher score indicates better ability to manage stress.

^jMPSS (range, 0-5)³¹ change score is from baseline to 6-month follow-up. It is a single-item scale; 0 indicates no urge to smoke in past 24 hours and 5 indicates an urge to smoke in the past 24 hours.

^kThe Internalized Shame from Lung Cancer Stigma Scale (range, 5-20)²⁹ change score is from baseline to 3-month follow-up. It has a sum score of 5 individual items with ranges of 1 to 4 (5 is not applicable). Higher scores indicate greater illness-related shame.

Treatment Satisfaction

At 3 and 6 months, participants in the intensive treatment group were significantly more satisfied with cessation assistance received, alignment of support with their needs, helpfulness of the treatment, and the quality of cessation assistance (eTable 5 in Supplement 2). At 6 months, 130 participants (85.0%) in the intensive treatment group vs 89 (59.3%) in the standard treatment group responded that the program met most or all of their needs (difference, 25.5% [95% CI, 16.0%-35.3%]; OR, 1.66 [95% CI, 1.24-2.24]; P = .001). In response to whether participants received the assistance they wanted, 145 participants (94.8%) in the intensive treatment group vs 117 (78.0%) in the standard treatment group responded affirmatively (difference, 16.8% [95% CI, 9.3%-24.3%]; OR, 1.76 [95% CI, 1.23-2.53]; P = .002).

Adverse Events

The most common adverse events were nausea (n = 13 for intensive treatment and n = 6 for standard treatment), rash (n = 4 for intensive treatment and n = 1 for standard treatment), hiccups (n = 4 for intensive treatment and n = 1 for standard treatment), mouth irritation (n = 4 for intensive treatment and n = 0 for standard treatment), difficulty sleeping (n = 3 for intensive treatment and n = 2 for standard treatment), and vivid dreams (n = 3 for intensive treatment and n = 2 for standard treatment).

Discussion

In this trial, compared with shorter-term telephone counseling,³⁵ providing sustained telephone counseling plus provision of free FDA-approved smoking cessation medication resulted in significantly greater biochemically confirmed abstinence at 6 months among patients with recently diagnosed cancer who had a long-term smoking history. The trial’s theory-based cancer-targeted counseling showed that improvements in smoking and cancer beliefs, emotions and physical experiences, and smoke-free home environment modifications were associated with quitting.

Participants in the intensive treatment group were significantly more likely to use cessation medication, but medication use was not associated with quitting. Although all patients received shared decision-making medication support, and any cessation medication use was assessed at each session, a stronger emphasis on dose and duration of adherence is likely warranted. More than half of participants in the intensive treatment group requested additional 4-week refills, which is consistent with prior research documenting that approximately half of patients with cancer are adherent to cessation medication regimens.³⁶ Patients with cancer might be particularly reluctant to add additional medications to their treatment regimen and this should be addressed in future studies.

Public Health Service guidelines endorse the value of intensive counseling support to promote quitting¹⁷; a greater number of sessions, as well as engagement in monthly booster sessions, was significantly associated with increased quitting. Among patients who were confirmed abstinent at 6 months, one-third quit between 3 and 6 months. Extending smoking cessation interventions has been shown to prevent relapse,³⁷ and booster cessation counseling sessions have been found to promote relapse prevention for many behavioral treatments.³⁸ A recent meta-analysis by Sheeran et al¹³ called for high-quality and effective smoking cessation interventions for patients with cancer; most trials had been conducted several years after diagnosis, and the mean number of counseling sessions in these trials was 4.6. The present intervention demonstrated that integrating an evidence-based, theory-driven, sustained tobacco treatment into the care of patients with cancer around the time of diagnosis can be effective.

Study findings underscore the importance of the American Society of Clinical Oncology and NCCN guidelines to offer smoking cessation assistance to all patients with cancer who smoke.^10,11 Results indicated that there was no variability in effectiveness across patient characteristics. This finding is particularly salient in the context of the trial’s broad eligibility criteria. Current smokers were eligible if they were willing to discuss their smoking behavior; patients with serious mental illness were eligible if there were no indications of current uncontrolled mental illness; and patients with advanced cancers were eligible. However, patients with advanced cancers were significantly less likely to complete follow-up assessments. Patients with nonsmoking-related tumors are less likely to receive tobacco treatment assistance and quit,³⁹ yet the study results suggest that patients with nonsmoking-related tumors were interested in tobacco treatment support. Having a smoking-related tumor was associated with, after consenting, decreased likelihood of being randomized; it is possible that these patients may have been offered other treatment or sought to quit on their own.

Limitations

The study has a number of limitations. First, because the study was conducted in 2 Comprehensive Cancer Centers, the effectiveness of this intervention may not be the same in other types of practice sites. Second, the trial completion rate was 78.1%. Third, the study was designed assuming what was considered to be a conservative estimate of a 15% group difference, but the observed effect size was only 13%. Fourth, the medication and counseling were provided without cost to patients, and this option may not be available in other less-resourced clinical settings. Fifth, there were challenges to approaching patients around the time of diagnosis to screen for eligibility; this speaks to the challenge of assessing and documenting tobacco status in the cancer setting.⁴⁰ Sixth, although greater session attendance was associated with achieving abstinence, the trial does not assume causality, nor does it demonstrate the effects of counseling compared with a no treatment condition. Seventh, the study design does not allow for disentanglement of effects of the counseling and medication intervention components, although the combined intervention components are adherent to NCCN and Public Health Service guideline recommendations^11,17 of counseling and medication treatment. Eighth, given the comparative effectiveness design, blinding was not possible; to decrease the potential for bias, (1) the content of the initial 4 counseling sessions, including medication advice and recommendations, was the same in both groups and (2) randomization was not conducted until the end of the first counseling session, and high-quality treatment fidelity procedures were followed.¹⁴

Conclusions

Among smokers recently diagnosed with cancer in 2 National Cancer Institute–designated Comprehensive Cancer Centers, sustained counseling and provision of free cessation medication compared with 4-week counseling and medication advice resulted in higher 6-month biochemically confirmed quit rates. However, the generalizability of the study findings is uncertain and requires further research.

Supplement 1.

Trial Protocol

Supplement 2.

eTable 1. Baseline Reports of Prior Use of Smoking Cessation

eTable 2. Treatment Utilization Across Study Groups

eTable 3. Intensive Treatment Group Treatment Utilization

eTable 4. Intensive Group Treatment Associated With 6-Month Cotinine-Confirmed Abstinence

eTable 5. Satisfaction With the Program

eFigure. Comparison of Smoking Cessation Outcomes Between Groups at 3 Months and 6 Months

Supplement 3.

Data Sharing Statement