Multiple Pharmacotherapy Adaptations for Smoking Cessation Based on Treatment Response in Black Adults Who Smoke

This randomized clinical trial evaluates the efficacy of multiple smoking cessation pharmacotherapy adaptations based on treatment response in Black adults who smoke daily at a health center in Kansas City, Missouri.


Introduction
About 1 of every 7 non-Hispanic Black adults in the United States smoke. 1 Smoking remains the leading cause of preventable disease and death for all people in the United States, including Black adults, and is the leading cause of cancer deaths. 2 Although smoking prevalence among Black adults is similar to the US national average, 1 Black adults tend to smoke fewer days per month and fewer cigarettes per day 3 but bear a disproportionate share of tobacco-related disease. 2,4 This makes the improvement of tobacco-related interventions for Black adults a national health priority. 5 Early abstinence is an important but overlooked target for intervention. More than two-thirds of individuals who smoke and do not achieve abstinence within 4 weeks of initiating pharmacotherapy do not achieve abstinence at later time points. [6][7][8] The current standard care for tobacco treatment recommends continuing an individual on pharmacotherapy for 8 to 12 weeks even if they continue to smoke, which is in direct contrast to treatment of other chronic diseases (eg, hypertension, diabetes) where altering medications to achieve desired outcomes are commonplace. [9][10][11][12][13] Within tobacco dependence, existing studies suggest changing pharmacotherapy results in higher rates of abstinence for those who do not show initial treatment response. [14][15][16][17][18][19] One key study assessed response to a nicotine patch (NP) after 1 week of precessation therapy and 1 week after the targeted quit date (TQD). 17 Participants who responded (ie, quit) remained on the NP, while those who did not respond (ie, continued smoking) were switched to either continuation of NP (control condition), rescue treatment with bupropion + NP, or rescue treatment with varenicline. Findings suggested that early treatment of individuals who continued smoking and were switched to bupropion + NP or varenicline had abstinence rates that were almost twice as high as those who continued smoking and remained on NP at the end of treatment and at the 6-month follow-up. In a large prospective cohort of individuals who smoke and use smoking cessation medications across multiple quit attempts, individuals who switched to a different medication for each quit attempt had significantly higher abstinence rates relative to early and late users of medication or those who repeated the same medication across all quit attempts. 15 Importantly, the act of switching drove better success regardless of the designated medication, which is consistent with other studies that have found that previous pharmacotherapy failure does not affect treatment outcomes. 20,21 Prior studies have adapted pharmacotherapy only once and/or focused on adaptation distal to a failed quit attempt, [14][15][16][17][18][19] despite evidence that adapting therapy early minimizes decreases in selfefficacy, enhances treatment engagement, and could improve treatment response. 17

Trial Design
The study was a 26-week unblinded and open label randomized clinical trial to evaluate the efficacy of multiple pharmacotherapy adaptations in combination with counseling for smoking cessation among Black adults who smoke daily. 33

Participants
The study was conducted at Swope Health, a federally qualified health center in Kansas City, Missouri. Participants were recruited through in-clinic and community-based efforts in the metropolitan Kansas City area, including flyers; physician letters; direct referrals; health fairs; in-clinic recruitment; radio, television, and social media advertisements; and word-of-mouth referrals from current and former participants. 33 Eligible participants self-identified as non-Hispanic African American or Black, were 18 years or older, smoked between 5 to 30 cigarettes per day for at least 25 of the past 30 days, registered an exhaled carbon monoxide (CO) of 5 ppm or higher at enrollment, and were interested in quitting smoking and willing to set a TQD within 7 days of enrollment.
Exclusion criteria included medical contraindications for NP, varenicline, or bupropion, a pharmacotherapy-assisted quit attempt in the past 30 days, past 30-day use of noncigarette tobacco products including e-cigarettes, and unwillingness to refrain from the use of electronic cigarettes or nonstudy provided smoking cessation pharmacotherapy.

Randomization
After providing written consent, participants were individually randomized 1:1 to ADT (n = 196) or UC (n = 196) groups. Randomization was stratified based on sex assigned at birth and baseline cigarettes per day smoked.

Intervention
Treatment included 18 weeks of study medication (NP, varenicline, bupropion + NP), 7 sessions of counseling, and follow-up through week 26 (Figure 1). Participants were eligible to receive $350 in total compensation for completing study activities. Remuneration was based on visit attendance and not smoking status.

Pharmacotherapy
Participants randomized to UC received 18 weeks of 24-hour 21 mg NP. Participants randomized to ADT received 2 weeks of 24-hour 21 mg NP at baseline and up to 2 pharmacotherapy adaptations, with a first switch to varenicline and a second switch, if needed, to bupropion + NP based on CO-verified smoking status (CO Ն6 ppm) at weeks 2 and 6. Medications were dispensed at weeks 0, 2, and 6, with TQDs occurring at weeks 1, 3, and 7. The week 2 adaptation occurred after 2 weeks of patch therapy and 1 week after the first TQD. The week 6 adaptation occurred after 4 weeks of varenicline therapy and 3 weeks after the second TQD, allowing 1 week to titrate to the full dose of varenicline and 3 weeks of a quit attempt after reaching the full dose. Medication selection and order included NP because it is available over the counter and is the most used among Black adults who smoke, 24,34 varenicline as the most effective medication, 35,36 and bupropion + NP to reflect the benefit of combination therapies. 17,37 ADT participants with a CO status of 5 ppm 38 or less at follow-ups at weeks 2 and/or 6 were considered to have responded to treatment and were continued on their existing pharmacotherapy.
Those with a CO status of 6 ppm or more were considered to have not responded to treatment and were switched to the next pharmacotherapy.

Counseling
Evidence-based, individualized, and culturally specific counseling 19,20 was conducted in-person at weeks 0 (baseline), 2, 6, 12 and by phone at weeks 1, 3, and 7. Counseling focused on managing smoking cues, triggers, and the acute positive reinforcing effects of smoking in addition to dealing with nicotine withdrawal and craving. 35,39 Counselors were accredited Tobacco Treatment Specialists supervised by a clinical psychologist (L.S.C.).

Measures
Demographic and tobacco use history, adverse effects, [40][41][42][43] counseling attendance, cessation fatigue (ie, "How tired are you of trying to quit smoking or stay quit today?"), 44 commitment to abstinence (ie, "Which of the following best describes your current goal for quitting cigarettes over the next 2 weeks?" with response options of quit completely, cut back but not quit, and no longer interested in quitting), 45 withdrawal, 46 and craving 47 were collected to quantify response to treatment.
Medication adherence was assessed via 7-day Timeline Follow Back Interview at each in-person visit  In-person In-person In-person In-person a Medication was dispensed at weeks 0, 2, and 6. All participants started 2 weeks of nicotine patch and received 18 total weeks of pharmacotherapy during the study. b Smoking status was monitored via exhaled carbon monoxide at weeks 2 and 6. Carbon monoxide of 6 ppm or higher was determined a priori to indicated continued smoking (ie, treatment nonresponse), which triggered adapation to the next pharmacotherapy.
c Carbon monoxide monitoring resulted in 3 possible treatment pathways: path 1, nicotine patch; path 2, varenicline; and path 3, bupropion + NP. and corroborated with patch and pill counts. 48,49 Adherence was defined as those who took 80% or more of the prescribed doses.

Outcomes
The primary end point was anatabine-and anabasine-verified 7-day point prevalence smoking abstinence (Յ2 ng/ml 50 ) at week 12 using intent-to-treat principles and treating missing as participants who smoke. Secondary outcomes were anatabine and anabasine-verified 7-day point prevalence abstinence at week 18 (end-of-treatment) and 26 (follow-up). Week 12 was selected as the primary end point because of the focus on early treatment response and to allow time for all adaptations to occur.

Sample Size
Postulated outcomes at week 12 were 18% abstinence among participants in UC 51-53 and 32% abstinence among participants in ADT, 17,37 treating those lost to follow-up as participants who smoke per the Russell standard. 54 With 196 participants per group, the study provided 90% power to detect the expected differences with a type I error rate of 5%.

Statistical Analysis
Participants were analyzed according to their randomized treatment condition. The χ 2 test was used to compare verified abstinence at week 12 (primary end point) and weeks 18 and 26 (secondary end points) between ADT and UC. A post hoc sensitivity analysis of smoking abstinence at week 12 was performed with multiple imputation using a monotone logistic regression with treatment and sex as covariates to impute the missing data. Verified and self-reported abstinence for individuals who completed the study was also compared. Anatabine and anabasine are the reference standard for confirming abstinence in circumstances where cotinine measurements are invalid because detectable levels of cotinine could reflect nicotine replacement therapy use, smoking, or both. 55 A cut-point of 2 ng/mL or less for both alkaloids was used to differentiate individuals who smoke from those who do not smoke. 50 The study experienced less than 15% missing data at the week 12 primary end point and missingness was not related to group or baseline characteristics. χ 2 and t tests were used to compared differences in treatment adherence, cessation fatigue, commitment to abstinence, and treatment-related side effects between groups. Post hoc logistic regression analyses modeled week 2 early treatment response as an estimator of week 12 abstinence. SAS, version 9.4M7 (SAS Institute), was used for all statistical analyses with P < .05 indicating statistical significance, and all tests were 2-tailed. Initial data analysis started in March 2022 and continued through January 2023. those missing imputed as participants who smoke. Analyses based on verified and self-reported abstinence among individuals who completed the study only provided similar results. Furthermore, treatment condition had no effect on withdrawal and cravings (eFigures 1 and 2 in Supplement 2). Table 3 shows the proportion of participants in the ADT group who were adapted and demonstrated verified abstinence by path (NP, varenicline, and bupropion + NP). Excluding those who responded to NP at both time points and did not require adaptation (n = 22), 11 verified individuals who quit smoking were rescued at week 12 by using adapted therapy (8.1%).

Treatment Adherence
There was no difference in counseling session attendance between treatment, with participants in the ADT and UC groups completing a mean (SD

Experience With Quitting
Among participants who completed the study, there were no differences by treatment group or over time related to cessation fatigue or their commitment to abstinence. On the final TQD (week 7), 138 of 166 ADT participants (83.1%) and 158 of 177 UC participants (89.3%) remained committed to

Early Treatment Response
Post hoc analyses of individuals who completed the study (n = 374) indicated that a similar proportion of those in ADT (61 of 187 [32.6%]) and UC (60 of 165 [36.4%]) had a CO score of 5 ppm or less at week 2 (P = .45), indicating that lack of a treatment effect was not due to better response to  c Participants lost to follow-up were imputed as individuals who smoke. Missingness was not related to group or baseline covariates. NP in the first 2 weeks in UC vs ADT. Imputing those who were missing at week 2 as early treatment individuals who did not respond to treatment and imputing those lost to follow-up at week 12 as adults who smoke (n = 392), treatment response based on CO-verified abstinence at week 2 (OR, 5.0; 95% CI, 2.7-9.1; P < .001) but not treatment (OR, 1.7; 95% CI, 0.9-3.1; P = .09) estimated week 12 abstinence. Analyses including only participants who completed the study revealed similar findings.

Discussion
Adaptation to varenicline and/or bupropion + NP after the failure of NP monotherapy did not significantly improve abstinence rates for Black adults who smoke relative to continued treatment with NP. The lack of treatment association was not explained by group differences in interest of quitting at study onset, early response to NP, or cessation fatigue.
The lack of treatment association is largely explained by a lower-than-expected effect of adapted therapy, with ADT rescuing only 5% of adults who smoke as opposed to the 15% anticipated.
That UC worked as well as ADT is encouraging given that UC requires fewer components, is less costly, and is easier to manage within the health care system. Our findings add to a growing body of literature suggesting that multicomponent interventions increase participant burden and decrease adherence without conferring additional treatment benefit and lead to lower rates of abstinence than interventions with fewer components. 14,19,56,57 In the current study, adherence to new pharmacotherapy adaptations was low and resulted in only 11 additional participants who quit.
The lack of treatment association is also explained by the fact that nearly one-third of those in ADT never required adaptation (ie, remained on NP monotherapy). Participants who responded to NP during the first 2 weeks had 5 times greater odds of achieving abstinence at the primary end point relative to those who did not respond. This finding is consistent with the tipping point hypothesis, which suggests that individuals close to a hypothetical tipping point benefit greatly from limited and varied interventions, while those far from the tipping point benefit very little from even a strong intervention. 56

Limitations
This study had limitations. Our study was restricted to Black adults who smoke daily and are interested in stopping smoking. The UC intervention was more intense than treatment typically provided in primary care but is consistent with recommended practice. 35 Because this enhanced UC treatment served as a more rigorous comparator, quit rates achieved in the UC group are likely higher than those expected with standard advice to quit accompanied by treatment with nicotine replacement. The 3-weeks of full dose varenicline prior to the switch to bupropion + NP may have been insufficient to yield varenicline's full effect given evidence that some individuals quit late in a quit attempt with varenicline. 55 Although, even among those who stayed on varenicline from week 2 through week 18, quitting was low overall. Adherence to varenicline and bupropion + NP was lower than mono-NP; however, determining if the medication, the order in which it was prescribed, the process of adaptation, or other factors were associated with medication adherence are beyond the scope of this study. Findings reflect use of medication following adaptation and are relevant to clinical practice.

Conclusions
In this randomized clinical trial of adapted pharmacotherapy vs standard of care pharmacotherapy, adaptation to varenicline and/or bupropion + NP after failure of NP monotherapy did not significantly improve abstinence rates for Black adults who smoke relative to continued treatment with NP.
However, these findings suggest that those who responded to pharmacotherapy in the first 2 weeks of the study had a 5 times greater odds of being abstinent at week 12. Findings highlight the continued need to identify effective treatment, particularly for those at high risk for cessation failure and those disproportionately impacted by tobacco-related disease.