A Pilot Randomized Clinical Trial of Brief Interventions to Encourage Quit Attempts in Smokers From Socioeconomic Disadvantage

Abstract

Introduction

Cigarette smoking disproportionately affects communities of low socioeconomic status where greater smoking prevalence and poorer cessation rates have been observed. Utilizing brief evidence-based interventions to increase cessation attempts may be an effective and easily disseminable means by which to mitigate undue burden in this population.

Aims and Methods

The current intervention randomized daily smokers (N = 57) recruited from a local community soup kitchen to receive either Brief (eg, 30 m) Motivational Interviewing, Nicotine Replacement Therapy (NRT) sampling, or a Referral-Only intervention. Approximately half of participants (50.9%) reported not completing high school and many reported either just (41.4%) or not (40.4%) meeting basic expenses. Follow-up was completed approximately 1-month postintervention.

Results

Nonsignificant group differences indicated that participants randomized to the NRT sampling condition were more likely to make a quit attempt (moderate effect size). Approximately 40% of the sample reported making a serious quit attempt at follow-up. Significant differences in cigarettes per day at follow-up, controlling for baseline, were observed, with participants in the Motivational Interviewing condition, only, reporting significant reductions. Participants randomized to the NRT condition were significantly more likely to report using NRT patch and lozenge at follow-up (large effect). There were no differences between groups with respect to seeking behavioral support. Finally, we found that subjective financial strain moderated the effect of condition on change in cigarette consumption where NRT sampling was more effective for participants reporting less financial strain.

Conclusions

Findings provide initial evidence for personalizing brief interventions to promote quit attempts in low-income smokers.

Implications

While most clinical research on tobacco use and dependence focuses on successful sustained abstinence, the current study is novel because it examined three brief interventions designed to increase the number of quit attempts made by a nontreatment-seeking group suffering from health disparities (ie, smokers from socioeconomic disadvantage). These data suggest that nontreatment-seeking smokers from socioeconomic disadvantage can be influenced by Brief MIs and these interventions should be used to motivate smokers from socioeconomic disadvantage to make a quit attempt. Future studies should examine combined MIs including pharmacological and behavioral interventions.

Introduction

Despite persistent efforts to reduce the public health impact of tobacco use, cigarette smoking remains the leading preventable cause of death United States, accounting for 480 000 deaths annually.¹ Indeed, cigarette smokers’ lifespans are 10 years less than that of nonsmokers,² due to myriad smoking-related chronic illnesses including lung and heart disease, chronic obstructive pulmonary diseases, and diabetes.¹ The cost of caring for the approximately 8.6 million individuals suffering from smoking-related illnesses in the United States is estimated to be greater than $300 billion dollars annually.³

Notably, the deleterious effects of tobacco use disproportionately impact communities of low socioeconomic status. Data from the National Health Interview Survey (NHIS⁴) indicate that smoking prevalence is higher in smokers with a greater number of socioeconomic disadvantages, with each additional disadvantage contributing to this disparity.⁵ Whereas the proportion of current versus never- or ex-smokers declined over time in those with two or fewer socioeconomic disadvantages, this trend has not been observed in those with three or more disadvantages, suggesting that the remaining smokers in the United States are particularly likely to experience multiple socioeconomic disadvantages.⁵ Relatedly, smoking rates among currently homeless adults are estimated to be 73%⁶ with smoking rates among those who had ever been homeless at 57%⁷ as compared with just 14.3% of the US population living at or above the poverty level.⁸ Yet, no differences in desire to quit smoking based on homelessness status have been observed,⁷ suggesting that smokers with socioeconomic disadvantage may want to quit, but lack confidence in their ability to quit and/or the resources (eg, U.S. Food & Drug Administration approved medications) to help them.

It is especially important to evaluate how to best instigate quit attempts in this vulnerable population given the undue burden tobacco use creates among this group. Motivational Interviewing (MI) and Nicotine Replacement Therapy (NRT) sampling may be particularly suitable for this population given their brevity and potential for large scalability. To our knowledge, no previous trials have examined the effects of these brief interventions on instigating quit attempts, and whether one may be superior to the other in this population though other brief interventions have been examined in lower socioeconomic status smokers.^9,10

MI is a goal-directed, but nondirective counseling style that seeks to increase motivation for change by collaborating with a patient, eliciting information consistent with making a change from the patient, and demonstrating acceptance of the patient.¹¹ Multiple meta-analyses^12,13 have reported the efficacy of MI for tobacco dependence treatment. Multiple studies have found that a Brief MI intervention can effectively increase motivation to quit, even in smokers not interested in quitting,¹⁴ increase quit attempts,^15,16 and result in reductions in cigarettes smoked per day within 1 month.¹⁷ Evidence of the long-term effects of brief MI interventions are somewhat equivocal, with some work indicating no reduction 3-month postintervention.¹⁷ Conversely, others have found higher odds of quitting and continuous abstinence 1 year post-MI intervention in smokers receiving MI in the context of primary care,¹⁸ and higher rates of self-reported continuous abstinence in Swedish smokers 6 months post-MI intervention delivered as part of Quitline,¹⁹ as compared with controls.

Nicotine sampling involves the introduction of NRT to precessation smokers for a brief (1–2 week) period with the goal of increasing smokers’ knowledge of how NRT functions without the pressure of proximal cessation. Indeed, individuals who engage in nicotine sampling are more likely to engage in a 24-hour cessation attempt, compared with groups randomized to receive no-treatment²⁰ or a Quitline referral.²¹ Moreover, Carpenter et al.²² found that among participants who completed a practice quit attempt with telephone aided counseling, those who additionally received nicotine therapy to sample were more likely to complete a 24-hour cessation attempt. Sampling methods are also associated with an increase in positive perceptions of NRT²³ and self-reported motivation to quit.^21,24 Yet, evidence for sampling interventions increasing persistent abstinence is mixed, with participants who engaged in sampling showing superior 6-month abstinence rates than those receiving no-treatment,²⁰ but not relative to active conditions.^21,22

The current pilot randomized clinical intervention examined the effects of MI, nicotine replacement sampling, and a Referral-Only control condition in promoting quit attempts among socioeconomically disadvantaged smokers. Given evidence for both MI and NRT sampling in promoting quit attempts, we hypothesized that these conditions would be superior to the Referral-Only condition. However, whether or not MI or NRT sampling would be superior in our socioeconomically disadvantaged sample was exploratory. Our outcomes of interest were (1) making a serious quit attempt within the 30 days following the intervention and (2) engagement in treatment-seeking behavior (ie, using NRT, and/or contacting a quitline or local tobacco dependence treatment clinic). As secondary outcomes of interest, we assessed (1) changes in daily cigarette consumption; (2) biochemically verified abstinence (CO [carbon monoxide] < 5 ppm)²⁵ postintervention; (3) beliefs about NRT; and (4) importance, confidence, and readiness to quit. We examined all outcomes of interest across the sample and between intervention groups to determine whether smokers assigned to receive brief MI evidence significantly greater smoking cessation relevant behaviors and changes in beliefs supportive of smoking cessation.

Materials and Methods

Participants

We screened 138 self-identified cigarette smokers receiving services at a local food bank in a northeastern US state between June and August 2018. Participants meeting initial screening eligibility (N = 64) provided written informed consent and were randomized using block randomization to one of three intervention conditions: (1) Referral-Only (n = 21), (2) Brief MI (n = 22), and (3) NRT sampling (n = 21). Of these, seven participants were withdrawn (see Figure 1; CONSORT diagram), resulting in 20 receiving the Referral-Only intervention, 19 receiving the Brief MI intervention, and 18 receiving the NRT sampling intervention. Participants were required to be daily smokers, aged 19–65, to provide an expired breath CO reading greater than 5 ppm, receiving services at a Central NJ based social services agency, and able to read and speak English. Participants were excluded if they were already taking U.S. Food & Drug Administration approved smoking cessation aids, taking antipsychotic medications, self-reported current medical issues of potential concern to nicotine replacement users, or had pending legal issues with potential to result in incarceration. Women had to be using effective birth control and could not be pregnant or nursing or planning on becoming pregnant in the next 2 months. This study was Institutional Review Board approved before patient enrollment.

Figure 1.

Consort diagram. MI = Motivational Interviewing; .NRT = Nicotine Replacement Therapy

Measures

Fagerström Test for Cigarette Dependence²⁶

This 6-item scale was used to evaluate the quantity of cigarette consumption, the compulsion to use, and nicotine dependence. The measure has both yes/no items (scored as 1 or 0) and multiple-choice items (scored from 0 to 3). These scores are then summed to give a total score from 1 to 10 where the higher the score the more intense subject’s the nicotine dependence is.

Change Questionnaire²⁷

This three-item measure assessed three components of motivation to quit smoking: Confidence in ability to quit (I could quit smoking), Importance of quitting (It is important for me to quit smoking), and Readiness to quit (I am trying to quit smoking) as rated on a 10-point Likert-type scale from 0 = Definitely Not to 10 = Definitely.

Financial Strain Scale²⁸

This 9-item measure was used to assess participants’ perceptions of financial strain. Eight items (eg, How difficult is it to afford a home suitable for [yourself/your family?]) were rated on a 3-point scale (1 = No Difficulty to 3 = Very Great Difficulty). The final item asked participants to report how much money they have left over at the end of each month.

Subjective Financial Situation Scale²⁹

This single item measure asks participants to describe their subjective view of their overall personal financial situation on a 4-point scale anchored by “Live comfortably” to “Don’t meet basic expenses.”

Attitudes Toward NRT Scale³⁰

This measure assesses previous offering of NRT, perceived advantages to NRT use, perceived side effects, and if relevant, reasons for not using NRT.

Quitting Preparation and Actions Questionnaire

This measure is used at follow-up only and assesses actions consistent with preparing for or making an attempt to quit smoking.

Interventions

All interventions were provided immediately following completing of the assessment measures described above. Follow-up interviews were also conducted in-person 30 days after the interventions. Participants were provided with gift cards for completing baseline and follow-up measures.

Referral-Only

The Referral-Only intervention consisted of simply providing written information regarding how to contact the state Quitline and the location and telephone number of a local tobacco dependence clinic.

Brief MI

The Brief (30 minutes) MI intervention consisted of an individual motivational interview focused on eliciting change talk regarding reasons for quitting smoking and confidence in one’s ability to quit smoking. At the conclusion, participants were provided (in an MI consistent manner) with the referral information described above. Therapists were clinical psychology PhD students who received 25 hours of live interactive training by the first author (a member of the Motivational Interviewing Network of Trainers). Therapists were required to be “certified” as having met prespecified, MI consistent criteria before seeing study patients. Therapists were “decertified” if subsequent audiorecorded sessions did not meet criteria and were “recertified” if they produced another practice audiorecording in which they met criteria.

NRT Sampling

The NRT sampling intervention (approximately 10 minutes) consisted of a description of the 21 mg nicotine patch and 4 mg nicotine lozenge and the provision of a free 2-week, supply of both. At the conclusion, participants were provided with the referral information described above.

Analytic Strategy

To assess for differences in outcomes between groups we used chi-square for dichotomous outcomes and analysis of variance for continuous outcomes. Significant results were probed using Dunnett’s method, whereby our comparison of interest was between the treatment conditions (MI and NRT sampling) and control (Referral-Only) condition. We further report effect size estimates (ie, f for ANOVA; Cramer’s V for chi-square) to help elucidate aspects of the pilot study that may be most beneficial. Data were also examined based on the assumption of “missing equals smoking” and carrying the last observation (ie, baseline data) forward, as a relatively conservative means of handling data for the n = 5 participants lost to follow-up. We will refer to additional analyses of imputed data as intent-to-treat (ITT) analyses.

Results

A total of 57 participants (M_age = 47.4 years [SD = 10.7]), completed the study. A majority of participants reported their biological sex as male (71.9%), and 3.5% identified as transgender women. The sample was racially and ethnically diverse with 15.8% identifying as Latino, 49.1% as white, 43.9% as black/African American, and 5.3% as American Indian or Alaska Native. The majority (49.1%) reported completing high school, with 19.3% not earning a high school degree or equivalent, 15.8% completing some college, and 15.9% completing at least 4 years of college. They reported an average financial strain score of 20.5 (SD = 4.6), with 40.4% reporting not meeting and 41.4% reporting just meeting basic expenses.

We were able to complete follow-up for 85% of those randomized to the Referral-Only intervention (n = 17), 94.7% of those randomized to the MI (n = 18) intervention, and 94.4% of those randomized to NRT sampling (n = 18) intervention.

Smoking Behavior at Baseline

Participants reported smoking an average of 12.3 cigarettes per day (SD = 6.7), presented with an average expired breath CO reading of 28.9 ppm (SD = 17.3), and scored a 4.8 (SD = 1.9), on the Fagerström Test for Cigarette Dependence, consistent with moderate dependence.²⁶ Participants reported a mean age of regular smoking of 17.9 years (SD = 8.0), smoking for an average of 26.5 years (SD = 13.2) and making an average of 2.9 prior serious quit attempts (SD = 2.7). Groups did not differ in terms of baseline nicotine dependence, CO, age they began regular smoking, or number of years smoking. However, despite randomization procedures, significant difference were observed in terms of cigarettes smoked daily at baseline, F(2, 54) = 4.1, p = .02), with post hoc analyses indicating participants randomized to the NRT condition reporting smoking significantly fewer cigarettes per day (M = 10.2, SD = 6.0) as compared with participants randomized to the MI condition (M = 15.6, SD = 7.7).

Motivation to Quit at Baseline

At baseline, participants, on average, reported a 5.6 (SD = 2.9), 7.3 (SD = 3.2), and 4.1 (SD = 3.0) out of 10 on confidence, importance, and readiness to quit smoking, respectively, on the Change Questionnaire.²⁷ No group differences in any of these beliefs was observed at baseline.

Quit Attempts Between Baseline and Follow-up

Of the n = 53 that completed their follow-up assessment, 40.4% reported making a serious quit attempt³¹ since their initial appointment. While observed group differences failed to reach statistical significance, χ ²(2) = 3.77, p = .15, Cramer’s V = .27 indicated a moderate effect size such that a larger proportion of participants receiving the NRT sampling intervention (58.8%) made a self-reported serious quit attempt compared with those receiving the MI (27.8%) or Referral-Only (35.3%) interventions. A nearly identical pattern was observed in the ITT sample (χ ²(2) =5.74, p = .22).

Cigarettes per Day at Follow-up

A 2 (time) by 3 (condition) repeated measures ANOVA was used to examine the effect of condition on cigarettes smoked per day from baseline to follow-up (see Figure 2). All assumptions of covariance, sphericity, and variance were met. There was a significant main effect of Time F(1, 49) = 7.33, p < .001, ${\mathrm{\eta }}_{\mathrm{p}}^2=0.13$, medium-large effect, with estimated marginal means indicating significant reductions in cigarettes from baseline (M = 12.30, SE = .87) to follow-up (M = 10.09, SE = 8.14). However, there was no overall difference in cigarette consumption between conditions F(2, 49) = 2.44, p < .10, ${\mathrm{\eta }}_{\mathrm{p}}^2=0.09$small-medium effect or Time × Condition interaction F(2, 49) = 1.57, p = .22, ${\mathrm{\eta }}_{\mathrm{p}}^2=0.06$, small-medium effect. Despite these nonsignificant effects, small to moderate effect size estimates were observed and estimated marginal were examined to further characterize observed relations. Examination of cell means indicated that, overall, the MI condition (M = 13.71, SE = 1.40) smoked more cigarettes than the NRT (M = 9.55, SE = 1.44) and Referral-Only conditions (M = 10.32, SE = 1.44); however, post hoc comparisons did not indicate any significant differences between conditions. Finally, cell means indicated that over time, participants in the MI condition evidenced the greatest reductions in cigarettes from Time 1 (M = 15.67, SE = 1.48) to Time 2 (M = 11.75, SE = 1.64), the NRT condition evidenced small reductions from Time 1 (M = 10.71, SE = 1.53) to Time 2 (M = 8.39, SE = 1.69), and the Referral-Only condition did not evidence meaningful reductions in cigarettes from Time 1 (M = 10.52, SE = 1.52) to Time 2 (M = 10.13, SE = 1.69). Post hoc comparisons indicated significant differences in reductions in cigarettes over time between the MI and NRT conditions (p < .05), and trend level differences in reductions over time between the MI and Referral-Only conditions (p = .10).

Figure 2.

Average number of cigarettes smoked per day by condition at baseline and follow-up. Post hoc comparisons indicated significant differences in reductions in cigarettes over time between the MI and NRT conditions (p < .05). MI = Motivational Interviewing; NRT = Nicotine Replacement Therapy.

We further examined differences in days abstinent or percent reduction in cigarettes per day between groups. There was no effect of treatment condition on percent reduction whether examining only those who attended follow-up or the ITT sample, and whether covarying for baseline cigarettes per day or not. Only one participant (randomized to Referral-Only) was biochemically confirmed (CO < 5 ppm) abstinent at the follow-up appointment, though six reported at least 1 day of abstinence (four randomized to NRT, two randomized to MI, and two randomized to Referral-Only) with an average mean number of days abstinent in the total sample of 1.8 days (SD = 5.5).

NRT at Follow-up

Among participants who attended follow-up (regardless of treatment condition), 20.8% and 18.9% reported starting NRT patch or lozenge, respectively. At follow-up, chi-square tests revealed significant differences in NRT patch use by intervention group (χ ²(2) = 8.0, p < .05, Cramer’s V = .39, large effect size), with 41.2% randomized to the NRT sampling condition reporting patch use, and just 11.8% and 5.6% in the Referral-Only and MI conditions, respectively. The same pattern was observed in the ITT sample (χ ²(2) = 8.44, p < .05), with 38.9%, 10.0%, and 5.3% using the nicotine patch in the NRT sampling, Referral-Only, and MI conditions, respectively. A similar pattern was observed for the NRT lozenge (χ ²(2) = 18.7, p < .001, Cramer’s V = .60, large effect size), with 52.9% using the nicotine lozenge in the NRT sampling condition, 5.9% in the Referral-Only condition, and none in the MI condition. Again, the ITT group showed a similar pattern (χ ²(2) = 19.3, p < .001) with 50.0%, 5.0%, and 0.0% reporting lozenge use in the NRT sampling, Referral-Only, and MI conditions, respectively. On average, participants who reported any NRT use at follow-up, reported using the lozenge 12.1 (10.0) days, and the patch 8.3 (SD = 6.7) days.

Participants were additionally queried regarding initiation of other nicotine replacement products, including gum, nasal spray, and inhaler, as well as vaping. Very few participants reported initiating any of these strategies to reduce smoking, and no group differences were observed (see Table 1). Across the sample, changes in beliefs that NRT helps with affect, reduces smoking and promotes cessation, or the belief that NRT is necessary to quit, were not observed from baseline to follow-up. However, participants reported significant decreases in the belief that NRT causes side effects and dependence (t(51) = 2.93, p < .01). Changes in NRT beliefs did not vary by treatment condition.

Table 1.

Cessation Support Initiated Between Baseline and Follow-up (n = 52)

	NRT (n = 17), %	MI (n = 18), %	Referral-Only (n = 17), %	Overall, %
Nicotine replacement therapy
Patch	41.2	5.6	11.8	19.2*
Lozenge	52.9	0.0	5.9	19.2*
Patch or lozenge	58.8	5.6	11.8	25.0*
Gum	5.9	5.6	5.9	5.8
Nasal spray	0.0	0.0	0.0	0.0
Inhaler	0.0	0.0	5.9	1.9
Other NRT	5.9	5.6	11.8	7.7
Other pharmacotherapy
Bupropion	0.0	0.0	0.0	0.0
Varenicline	0.0	0.0	0.0	0.0
Behavioral support
Quitline	0.0	5.6	5.9	3.8
Tobacco dependence program	0.0	0.0	17.6	5.8*
Quitline or tobacco dependence program	0.0	5.6	17.6	7.7
Self-help via Internet	11.8	22.2	11.8	15.4
Physician/nurse	5.9	5.6	17.6	9.6
Mental health professional	0.0	11.1	5.9	5.8
Addiction counselor	0.0	5.6	5.6	3.8
Any other behavioral support	17.6	33.3	23.5	25.0
Other
Vaping	11.8	11.1	11.8	11.5

MI = Motivational Interviewing; NRT = Nicotine Replacement Therapy.

*p < .05.

Behavioral Support

Regarding initiation of behavioral support to reduce smoking, just 3.8% and 5.8% of participants attending follow-up reported contacting the state Quitline or the local tobacco dependence program, respectively, and no difference was observed between conditions. We additionally queried participants regarding initiation of other behavioral support and self-help. Across conditions, 17.6% reported pursuing some other form of behavioral intervention with self-help via the Internet being the most frequently endorsed method, followed by speaking to a nurse or physician; no differences by condition were observed (see Table 1).

Motivation to Quit

Among participants who attended follow-up, no differences were detected between intervention groups with regard to confidence in one’s ability to quit (F(2, 49) = 1.51, p = .23, f = .24), self-reported rating of how important it would be to quit smoking (F(2, 49) = 0.45, p = .64, f = .13), or readiness to quit (F(2, 49) = 1.51, p = .23, f = .24).

Financial Strain

Given the importance of identifying optimal approaches for promoting cessation in low socioeconomic status smokers, post hoc analyses were conducted to investigate the role of subjective financial strain on changes in past month (ie, 28-day) cigarette consumption by treatment condition. The PROCESS macro was used, here, which uses ordinary least squares regression to estimate models.³² PROCESS produced 95% bias-corrected confidence intervals, estimated via bootstrapping analyses in order to determine statistical significance of main and interaction effects (see Figure 3). A significant interaction between condition and financial strain was found, which accounted for 10.1% of the variance in smoking reduction at follow-up (F(1, 46) = 5.45, p = .02; f² = .11 (small-medium effect)). Specifically, participants assigned to the Referral-Only condition who reported lower financial strain evidenced significantly less reductions in smoking, whereas participants assigned to the NRT condition who reported lower financial strain evidenced significantly greater reductions in smoking. Financial strain did not appear to differentially predict reductions in smoking in the MI condition.

Figure 3.

Visualization of the conditional effect of intervention condition on smoking reduction as a function of subjective financial strain. MI = Motivational Interviewing; NRT = Nicotine Replacement Therapy.

Discussion

The current pilot randomized clinical trial examined the effects of three brief interventions (MI, NRT sampling, and Referral-Only) on promoting quit attempts among daily smokers from socioeconomic disadvantage. Overall, more than 40% of participants across the three intervention groups self-reported making a serious quit attempt in the month between baseline and follow-up. A moderate effect of intervention on quit attempts indicated that a larger proportion of participants receiving the NRT sampling intervention made a quit attempt as compared with those receiving the MI or Referral-Only interventions. Almost 60% of those receiving the NRT sampling intervention made a quit attempt as compared with 28% and 35% of those in the MI and Referral-Only interventions, respectively.

NRT sampling’s superiority in instigating serious quit attempts in the current study is somewhat consistent with that of a randomized clinical trial examining the effects of a reduction intervention plus NRT on smoking cessation compared with an MI only intervention²⁰ in which a greater percentage of individuals randomized to implement a reduction plan with concurrent NRT reported more quit attempts within the 6 weeks postintervention compared with individuals in the MI condition. In contrast, when compared with an educational control rather than with NRT, MI led to a higher number of quit attempts within a 30-day period in a nontreatment-seeking population of individuals with serious mental illness.¹⁶

None of the brief interventions used in the current study were enough to produce substantial amounts of biochemically confirmed cigarette abstinence. Indeed, only one participant across the three interventions had biochemically confirmed abstinence at the 30-day follow-up and just six self-reported at least 1 day of abstinence at some point during the 30-day follow-up period. This finding is not completely unexpected, however, as this was a nontreatment-seeking sample experiencing socioeconomic disadvantage who was only provided with NRT or behavioral counseling (MI), but not both. It is also consistent with previous work in a nontreatment-seeking sample of smokers with serious mental illness who were more likely to make quit attempts after receiving a brief adaptation of MI (as compared with an educational control), but were not more likely to be abstinent at follow-up.¹⁶ It should also be noted that while only six participants reported instances of abstinence on at least 1 day, many participants made serious quit attempts. This is aligned with current perspectives that limiting the definition of a quit attempt to indicate a period of at least 24 hours of abstinence may grossly underestimate the number of true quit attempts made in a given sample.³¹ This may be especially true in smokers from lower socioeconomic status such as the current sample who traditionally have more difficulty quitting.

Additionally, there were no differences between intervention groups with respect to cigarettes smoked per day in the 7 days leading up to the follow-up assessment. Those randomized to receive the MI intervention, however, significantly reduced the number of cigarettes they smoked per day between baseline and follow-up. Despite this, there were no significant differences in days abstinent or percent reductions in cigarettes between treatment conditions. Finally, there were no group differences in treatment seeking or in motivation to quit smoking.

The ability of the MI intervention in this study to produce decreased cigarette smoking is consistent with previous empirical research examining the effects of MI on smoking outcomes. MI for adolescent smokers¹⁷ and smokers with serious mental illness¹⁶ both led to reductions in cigarettes smoked per day. Our data are inconsistent, however, with previous work finding that cigarette reduction counseling plus NRT produced a greater reduction in cigarettes smoked per day compared with MI.²⁰ This discrepancy may be in part to the differences in samples (one being nontreatment seeking and one being treatment seeking) and because our NRT intervention was not paired with behavioral counseling.

Not surprisingly, we detected greater reported use of NRT patch and lozenge within the NRT intervention group as compared with the MI or the Referral-Only intervention groups. Across the entire sample, many more participants used NRT during the follow-up period than contacted the state Quitline or a community-based tobacco dependence program. These results suggest that exposure to NRT promotes NRT use, which is consistent with interventions conducted both in populations not actively engaged in cessation³³ as well as a clinical trial examining the effects of adding NRT to a practice quit attempt compared with a practice quit attempt alone.²² However, it is notable that few participants initiated other forms of NRT or vaping, underscoring the importance of direct provision of NRT. Individuals randomized to receive NRT were more likely to utilize NRT and less likely to use behavioral supports than those who were not provided medication.²² Yet, in contrast to NRT findings, participants were no more likely to seek the behavioral support recommended to them (eg, NJ Quitline, local tobacco dependence program), than other forms of behavioral support (eg, self-help via Internet, speaking with physician/nurse). Policy and regulatory research suggest that increasing access³³ and accurate information about NRT³⁴ may increase medication utilization.

While we found that participants in the current study reported significant decreases in the belief that NRT causes side effects and dependence from baseline to follow-up, these changes did not vary by treatment condition. It is unclear why these changes would occur across interventions rather primarily in those who received the NRT sampling intervention and used NRT.

Additionally, we were surprised that a study with two active interventions designed to promote quit attempts did not produce differences between active and control interventions with regard to confidence in one’s ability to quit, self-reported rating of how important it would be to quit smoking, or self-reported rating of activities related to currently trying to quit.

The current study was novel in that we examined brief smoking cessation interventions in socioeconomically disadvantaged smokers in the local community. Little is known about the effectiveness of brief smoking cessation interventions in disadvantaged groups despite the tremendous disparity in smoking prevalence among this group. Our post hoc analyses revealed that financial strain may be a significant moderator of the effect of intervention on smoking behavior. Specifically, greater reductions in smoking were observed in smokers provided with NRT who reported less financial strain. This is consistent with previous work suggesting that financial strain negatively influences smoking cessation. It is somewhat inconsistent, however, with findings that financial strain exerts its influence through its effect on withdrawal symptoms³⁵ given that NRT is specifically designed to address these symptoms.

Smokers low in financial strain, conversely, reported less reduction in cigarette consumption in the Referral-Only condition. This is somewhat surprising, given the NRT offered in this study could be viewed as a cost-effective means by which to manage nicotine craving. It is possible that for smokers moderate to high in financial strain, other unmeasured factors related to financial strain may have interfered with NRT utilization. In fact, our interaction revealed that for smokers moderate to high in financial strain, no significant differences in smoking reduction were observed across treatment conditions. It is possible that participants experiencing fewer daily stressors compared with those with greater financial strain are simply better able to implement smoking cessation strategies because they are less distracted by more immediate needs such as food and shelter. It may also be that smokers with greater financial strain are simply less concerned with the negative health effects of smoking due to other stressors. Future work may benefit from better characterizing how financial strain impedes engagement in smoking cessation. It may be that for these individuals, strategies to reduce financial strain may need to be incorporated into treatment.

Limitations of this study include the relatively small sample size inherent in a pilot study and the fact that there was no condition combining a pharmacological aid with behavioral counseling as would be recommended by the PHS Guidelines.³⁶ While we found that more than 40% of participants across the three intervention groups made a self-reported serious quit attempt in the month between the intervention and the 30-day follow-up there were very few days of abstinence overall and only one instance of CO verified abstinence at follow-up. Additionally, our analyses related to financial strain were post hoc. This study adds to the literature in that there are very few studies focused on instigating quit attempts in nontreatment-seeking populations, rather than focusing on cessation. Additional work is needed to understand whether combined interventions in socioeconomically disadvantaged smokers may yield stronger effects.

Supplementary Material

A Contributorship Form detailing each author’s specific involvement with this content, as well as any supplementary data, are available online at https://academic.oup.com/ntr.

Funding

This work was funded by a Rutgers University Office of Community Relations, Community-University Research Partnership Grant awarded to MLS, PhD and TML, PhD.

Declaration of Interests

None declared.