A Randomized Pilot of a Tailored Smoking Cessation Quitline Intervention for Individuals Who Smoke and Vape

Abstract

Introduction

Although e-cigarettes are not a federally approved tobacco cessation aid in the United States, many smokers use them to quit or cut down on smoking. Tailored behavioral support could improve rates of complete smoking cessation for those individuals.

Aims and Methods

A novel behavioral treatment to help dual cigarette and e-cigarette users quit smoking was tested in a randomized pilot with a state tobacco quitline. Ninety-six dual users of cigarettes and e-cigarettes were recruited from incoming state quitline callers and randomized to receive enhanced e-cigarette coaching (EEC) or quitline treatment as usual (TAU) to examine EEC feasibility and acceptability. Outcomes at 3 months were treatment satisfaction, engagement, beliefs, and smoking cessation. This pilot was not powered to detect differences in quit rates.

Results

Sixty-nine percent responded to the 3-month survey. EEC treatment satisfaction was noninferior to TAU: 93.8% (30/32) of EEC and 73.5% (25/34) of TAU reported being “very satisfied” or “satisfied” with treatment, respectively. EEC participants completed more coaching calls than TAU (M = 3.4 vs. M = 2.7, p = .03), and the majority in both groups elected to receive nicotine replacement therapy (EEC: 100%, TAU: 94%, p = .24). With missing data imputed as smoking, intent-to-treat 7-day point prevalence smoking abstinence rates were 41.3% (19/46) for EEC and 28.0% (14/50) for TAU (p = .20).

Conclusions

The EEC quitline intervention for dual cigarette and e-cigarette users demonstrated high levels of treatment satisfaction and engagement. This pilot was not powered to detect significant differences in smoking cessation; however, cessation rates were promising and warrant evaluation in a fully powered trial.

Implications

If this scalable behavioral treatment to help dual cigarette and e-cigarette users quit smoking proves to be effective in a larger trial, quitlines could implement this harm reduction approach to improve outcomes for callers who already use e-cigarettes and are planning to use them while quitting smoking.

Introduction

Although e-cigarettes are not Food and Drug Administration (FDA) approved for smoking cessation, they are commonly used in the United States by smokers trying to quit.¹ Their use as a cessation aid is less controversial in other parts of the world, such as the United Kingdom where e-cigarettes are regulated and promoted as a cessation aid.² Using e-cigarettes to quit smoking combustible cigarettes is a harm reduction strategy that has increasing support from randomized trials, which have shown, in general, that e-cigarettes are at least as helpful as FDA-approved nicotine replacement therapy (NRT). A recent Cochrane review determined that nicotine-containing e-cigarettes were more effective than NRT at helping people quit smoking and rated the evidence as moderate.³ There is also preliminary evidence that vaping can be safely and effectively combined with NRT.^3–5 Although this accumulating evidence has not yet led to US regulatory bodies promoting vaping as a smoking cessation tool, adults in the United States are using e-cigarettes about as frequently as FDA-approved cessation medications.¹

Despite the growing evidence that e-cigarettes can help smokers quit using combustible products, there are few guidelines for how to best use e-cigarettes to completely quit smoking. For example, the National Health Service of the United Kingdom, which promotes e-cigarettes as a quit-smoking aid, does not offer comprehensive guidance for how to best use them.⁶ The US Centers for Disease Control and Prevention (CDC) advises educating smokers that e-cigarettes are not an approved quitting aid and recommends use of FDA-approved medications, such as NRT, as first-line treatment.^7,8 However, for smokers who have tried FDA-approved medications without success and are committed to using e-cigarettes for cessation, providers should support their quit attempt.^9–11 For the CDC, supporting the attempt includes suggesting frequent use of an e-cigarette with nicotine.⁷ As with NRT where research has shown that use in conjunction with specific education and behavioral guidelines can improve outcomes,¹² individuals planning to quit smoking utilizing e-cigarettes may be more successful with empirically supported education and behavioral support. Qualitative interviews with quitline callers who both smoke and vape found misunderstandings about the relative risks of smoking, vaping, and using NRTs, as well as the use of strategies that were not likely to lead to quitting smoking.¹³ Behavioral support guidelines for how to best utilize e-cigarettes to completely quit smoking need to be developed and tested, with a recent study supporting the efficacy of such an approach.¹⁴ Furthermore, although e-cigarettes are likely far less harmful than combustible cigarettes, they are not completely safe. Any plan to switch completely from smoking to vaping should also include support for a long-term goal to quit vaping.

Free tobacco quitlines are available in all 50 US states, as well as other countries around the world.¹⁵ Each year, approximately 57 000 adults (15% of callers) who call US tobacco quitlines for help quitting traditional cigarettes report dual use of e-cigarettes at registration.^16,17 E-cigarette use among quitline callers increased from approximately 10% to 15% from 2015 to 2018.¹⁷ Most report vaping to quit or cut down on their smoking.¹⁷ Current practice in the quitlines operated by Optum, the largest service provider for quitlines in the United States, is to advise dual users to quit both cigarettes and e-cigarettes and to use FDA-approved cessation aids.^17,18 Given recent evidence that e-cigarettes may be more efficacious than NRT and that use of e-cigarettes for quitting smoking is common, an alternative approach would be to support dual users who prefer to completely switch to e-cigarettes with education and specific behavioral support.

The purpose of the present study was to pilot a novel behavioral intervention designed to be delivered by quitline coaches. The target audience was quitline callers who were using both combustible cigarettes and e-cigarettes at enrollment and who were interested in using e-cigarettes as part of their smoking cessation plan. This randomized pilot study examined the feasibility and acceptability of the new behavioral support and educational intervention and explored outcomes compared to the standard quitline intervention.

Methods

Study methods and the enhanced e-cigarette coaching (EEC) intervention were previously described in detail.¹⁹ Research protocols were approved by the Western Institutional Review Board. The trial is registered with ClinicalTrials.gov, NCT03575468.

Participants

Participants were recruited from incoming callers to the Oklahoma Tobacco Helpline (OTH) between November 2018 and March 2020 who were smoking and had used e-cigarettes in the past 30 days. The OTH is a free, public health resource offering tobacco cessation services via phone, web, text, and/or print materials, as well as NRT. OTH callers received a study offer if they met the following criteria: smoked at least one cigarette per day (CPD), interested in phone coaching program enrollment, ready to quit smoking within 30 days, not pregnant or planning to become pregnant in the next 3 months, at least 18 years old, English speaking, and did not report schizophrenia diagnosis. Individuals who met these criteria were read a brief study description and invited to be screened. Screening criteria included currently using e-cigarettes some days or every day; possible or very likely to use e-cigarettes while trying to quit smoking (responses of “not at all likely” were excluded); access to an Android smartphone, operating system 6.0 or higher, and email; willing to use an app to complete daily surveys; no contraindications for NRT (eg, no recent heart attack or stroke); and no other household members already enrolled in the study.

Design

Individuals who met screening criteria completed a verbal consent and a baseline survey by phone with a research-trained coach. In this parallel group study, participants were then randomized to receive quitline treatment as usual (TAU) or EEC, using blocked randomization stratified by gender (male vs female; other response options for gender were not included for the quitline at the time of this trial, but have since been added) with an allocation ratio of 1:1. The quitline participant record software accessed a randomization table to automatically assign a participant to a group after a coach clicked a randomize button. Participants were required to complete their first coaching call to remain in the study to ensure all were exposed to treatment. In both groups, participants were contacted for four additional coaching calls, were encouraged to call in for support as needed, and received the NRT benefits standardly available to them through the OTH (2–8 weeks of patch, gum, and/or lozenge). The five coaching calls for the two groups were delivered over an approximately 2-month period, with call 1 intended to take approximately 20–25 minutes, and later calls to take approximately 15 minutes. Participants downloaded a smartphone app to complete daily questionnaires on product use for 12 weeks. Participants received $20 for completing the baseline survey and up to $110 for daily questionnaires.

Interventions

The quitline intervention, based on the US Public Health Service clinical guidelines and grounded in Social Cognitive Theory,^19–21 focused on five key strategies for quitting: committing to a quit date, coping with urges, using medications effectively, disposing of tobacco paraphernalia, and utilizing social support. The TAU protocol included recommending participants stop both cigarettes and e-cigarettes on their quit date when they start NRT.

In contrast, the EEC protocol added four intervention components to standard evidence-based quitline treatment: e-cigarette education, a shared decision-making model (SDM) for quit plan development (offering selection of NRT, e-cigarette, both NRT and e-cigarette, or no nicotine replacement aid; in these conversations coaches discussed e-cigarettes as a quitting tool similar to NRT and as an alternative form of nicotine replacement), behavioral support tailored to the selected quit plan, and a requirement for coaches to assess and address e-cigarette use on every call. The underpinnings of these added components were the theory of reasoned action,^22–24 which highlights the need to change beliefs (eg, misconceptions about e-cigarettes or NRT) prior to changing behavioral intention or behavior, and self-determination theory,^25,26 an underpinning of SDM and patient-centered care which asserts that a participant’s natural inclinations towards motivation, engagement, and growth can be supported or thwarted based on their experiences within a social context, particularly with regard to autonomy, competence, and relatedness.

Calls were recorded, and 20% were reviewed for fidelity. EEC calls were assessed for the presence of eight required call elements as described previously.¹⁹ Across EEC calls, 94.4% of the required elements were present. For TAU, 20% of calls were assessed for discussion about e-cigarettes to examine distinction between the treatments. Of these, 41% included conversation about e-cigarettes; all except one (3%; 1/32) aligned with the standard protocol described above. This TAU call, with a participant using lozenges and e-cigarettes to quit smoking, included more extensive assessment, education, and behavioral support on e-cigarettes.

Outcome Assessment at 3 Months

Outcomes were assessed 3 months after enrollment via an online survey or by phone with interviewers blinded to the study group. Nonresponders to phone calls were emailed a shortened outcome survey with questions assessing product use (tobacco, e-cigarette, NRT) and quitline satisfaction. Participants who self-reported smoking abstinence for 7 or more days were sent a carbon monoxide (CO) monitor to confirm abstinence (ie, CO readings of 6 ppm or less on 3 consecutive days). Participants received a $50 gift card for outcome survey completion and a $50 gift card for CO completion. Participants were not made aware that the invitation to complete a CO test was related to their reported smoking status on the 3-month follow-up survey.

Measures

At OTH enrollment, participant demographics and tobacco use history were collected, including age, gender, education, insurance status, amount, and type of tobacco used, and the Heaviness of Smoking Index (HSI).²⁷ Quitline computer systems recorded call completion and whether NRT was sent. The baseline and 3-month surveys included questions about smoking and vaping habits, use of NRT, knowledge and beliefs about e-cigarettes and smoking, and psychosocial functioning and substance use. The 3-month survey additionally included an assessment of program satisfaction, quit plan development experience, participant recommendations for quitting smoking, and side effects that participants attributed to using e-cigarettes or NRT.

Primary outcomes focused on treatment acceptability and included (1) treatment satisfaction measured on a 6-point scale from “very dissatisfied” to “very satisfied” and (2) number of coaching calls completed. Secondary outcomes included quit plan development experience and beliefs about relative harm of products. Utilization of shared decision-making principles during the participant’s quit plan development was measured with an adapted version of the three-item CollaboRATE shared decision-making scale.²⁸ To further assess perceptions of differences between treatments, we assessed satisfaction with time spent talking about e-cigarettes during coaching calls and the degree to which e-cigarettes were taken into account during quit plan development. Beliefs about relative harm of cigarettes, e-cigarettes, and smoking cessation medications were assessed with questions adapted from assessment recommendations from Pearson et al.²⁹ Finally, self-reported 7-day point prevalence smoking abstinence (7dppa) from smoking at 3 month was an exploratory outcome measured with the question: “When did you last use tobacco or smoke a cigarette, even a puff? Please do not include electronic cigarette or other vaping product use.”

Data Analysis

We hypothesized that, at 3 months, EEC participants would (1) have completed as many, or more calls; (2) report satisfaction rates at least as high, and rate quit plan development experiences more positively; and (3) report more accurate beliefs about e-cigarettes, smoking, and quit medications, compared to TAU. Although this pilot was not powered to detect differences in abstinence rates, we hypothesized EEC would have a higher 7ddpa from smoking compared to TAU. Differences between groups were examined using Fisher’s exact test (categorical variables) and rank-sum tests (scale scores, call completion), with an alpha level of .05. For treatment satisfaction, we also examined the 95% confidence interval (CI) for the difference in mean treatment satisfaction to perform a noninferiority analysis.

We examined 7dppa from smoking between treatment groups using logistic regression for 3-month survey respondents and for the intent-to-treat sample (assuming missing data meant continued smoking). To examine the potential influence of variables other than treatment group on respondent smoking abstinence rates, we conducted sensitivity analyses with the following covariates included age, gender (male vs. female), education (high school or less vs. more than high school), insurance status (uninsured vs. Medicaid-insured vs. Medicare or other insurance), and HSI score. All analyses for this study were performed using version 9.4 of SAS statistical software.³⁰

Results

Between November 2018 and March 2020, 918 OTH callers met eligibility criteria to be offered additional screening for the study and 110 individuals were interested in screening, met study inclusion criteria, and completed the study consent, baseline survey, and randomization (Figure 1). Call 1 completion was a requirement to remain in the study; four participants in TAU and 10 participants in EEC did not complete call 1, resulting in a final sample of 50 TAU participants and 46 EEC participants. These different rates of connection for call 1 were likely due to operational processes for connecting participants with the small group of four EEC coaches versus the larger group of TAU coaches available 24 hours per day. Of the 96 participants, 34/50 (68.0%) TAU and 32/46 (69.6%) EEC participants completed the 3-month survey (p = .99).

Figure 1.

CONSORT flow diagram.

Participants were 40.6 years old on average (SD = 13.5), 61.5% female, 64.6% white, 47.9% daily vapers (vs. nondaily), 85.4% daily smokers (vs. nondaily), and smoked 19.2 CPD on average (SD = 11.3). Detailed participant baseline data were reported previously; there were no significant baseline differences between the two groups on demographics or tobacco use history.¹⁹

Treatment Satisfaction and Quit Plan Development Experience

As shown in Table 1, among respondents, 93.8% (30/32) of EEC and 73.5% (25/34) of TAU reported being “very satisfied” or “satisfied” with treatment (p = .045). The mean satisfaction score was 5.5 for EEC (95% CI = 5.3% to 5.8%) and 5.0 for TAU (95% CI = 4.5% to 5.5%). Although the p-value for the difference in means was slightly too large (p = .055) to show superiority of EEC, since the lower confidence bound of a 95% CI for the difference in treatment satisfaction (EEC – TAU) was greater than −0.02, we are 95% confident that treatment satisfaction with the EEC group was noninferior to TAU with respect to mean treatment satisfaction score with a noninferiority margin of less than 0.02 calls.

Table 1.

Treatment Satisfaction Reported at 3-mo Outcome Survey

	Total sample (3-mo survey respondents: n = 66)		Enhanced e-cig coaching (3-mo survey respondents: n = 32)		Quitline treatment as usual (3-mo survey respondents: n = 34)		p
	%	95% CI	%	95% CI	%	95% CI
Treatment satisfaction, mean of 1–6 scale	M (SE) = 5.3 (0.13) Median (IQR) = 6 (5–6)		M (SE) = 5.5 (0.13) Median (IQR) = 6 (5–6)		M (SE) = 5.0 (0.22) Median (IQR) = 5.5 (4–6)		.0545
Treatment satisfaction, % very satisfied or satisfied	83.3	74.1–92.6	93.8	84.9–100.0	73.5	57.9–89.2	.0452
Quit plan development experience via adapted CollaboRATEa, mean	M (SE) = 9.6 (0.4) Median (IQR) = 11 (9–12)		M (SE) = 10.0 (0.4) Median (IQR) = 11 (9–12)		M (SE) = 9.2 (0.6) Median (IQR) = 10 (7–12)		.2869
Time spent talking to quit coach
Just right	71.7	59.9–83.4	76.7	60.6–92.7	66.7	48.8–84.6	.3437
Too much	21.7	10.9–32.4	20.0	4.8–35.2	23.3	7.3–39.4
Not enough	6.7	0.2–13.2	3.3	0.0–10.2	10.0	0.0–21.4
Time spent talking about e-cigs
Just right	69.4	57.6–81.2	83.3	69.2–97.5	56.3	38.1–74.4	.0116
Too much	3.2	0.0–7.7	6.7	0.0–16.1	0	—
Not enough	27.4	16.0–38.8	10.0	0.0–21.4	43.8	25.6–61.9
Coach took e-cig use into account when setting up quit plan
A lot	66.7	54.0–79.3	82.1	67.0–97.3	51.7	32.4–71.1	.0108
Some	15.8	6.0–25.6	10.7	0.0–22.9	20.7	5.0–36.4
A little	10.5	2.3–18.7	7.1	0.0–17.3	13.8	0.4–27.1
Not at all	7.0	0.2–13.9	0	—	13.8	0.4–27.1

CI = confidence interval; IQR = interquartile range. All variables in this table were assessed 3 mo after enrollment. Three of the 66 participants who provided outcome data completed a brief version of the outcome survey. The brief survey included assessment of treatment satisfaction, but did not include other variables reported in this table.

The adapted CollaboRATE measure included three questions (on a 5-point scale from “no effort” to “every effort was made”): “How much effort was made to help you understand your options for quitting smoking?”; “How much effort was made to listen to the things that matter most to you about your plan for quitting smoking?”; and “How much effort was made to include what matters most to you in creating your plan for quitting smoking?” Higher total scores indicate greater utilization of a shared decision-making approach during quit plan development.

There were no significant differences between groups on the shared decision-making quit plan development approach measured by the adapted CollaboRATE (p = .29) or on time spent talking to a quit coach (p = .34). There were significant differences on perceptions of time spent talking about e-cigarettes (EEC: 10.0% “not enough” vs. TAU: 43.8% “not enough”; p = .01), and degree to which coach took e-cigarette use into account when setting up quit plan (EEC: 82.1% “a lot” vs. TAU: 51.7% “a lot”; p = .01).

Treatment Engagement and Recommendations for Quitting

The average call completion for EEC exceeded that of TAU (EEC: M [SE] = 3.4 [0.24], median [interquartile range {IQR}] = 3 [2–5], 95% CI = 2.9% to 3.8%, vs. TAU: M [SE] = 2.7 [0.19], median [IQR] = 2.5 [2–4], 95% CI = 2.4% to 3.1%) (p = .03).

Nearly all participants requested and were sent NRT (100.0% EEC vs. 94.0% TAU, p = .24), and similar percentages for each group were sent a nicotine patch and gum or lozenge (combination NRT: 52.2% EEC vs. 58.0% TAU, p = .20). As shown in Table 2, at 3-month follow-up, there were no significant differences in reported use of one of the seven FDA-approved quit medications during the study (83.3% EEC vs. 90.9% TAU, p = .46), in use of combination NRT during the study (46.7% EEC vs. 48.5% TAU, p = .99), in use of NRT and e-cigarettes on the same day during the study (66.7% EEC vs. 54.5% TAU, p = .44), or in current use of an FDA-approved quit medication at the time of the 3-month survey (40.6% EEC vs. 52.9% TAU, p = .89).

Table 2.

Smoking, Vaping, and Quit Medication Use at 3-mo Follow-up

	Total sample (3-mo survey respondents: n = 66)		Enhanced e-cig coaching (3-mo survey respondents: n = 32)		Quitline treatment as usual (3-mo survey respondents: n = 34)		p
	%	95% CI	%	95% CI	%	95% CI
Use of FDA-approved quit medications (3-mo survey respondents)
Used FDA-approved quit medication during quit attempt, yes any	87.3	78.8–95.8	83.3	69.2–97.5	90.9	80.6–100.0	.4616
Combination NRT, yes	47.6	34.9–60.3	46.7	27.7–65.6	48.5	30.5–66.5	.9999
Current use of FDA-approved quit medication, yes any	47.0	34.6–59.3	40.6	22.6–58.6	52.9	35.3–70.6	.8943
Quit approach and recommendations (3-mo survey respondents)
Used e-cigs and NRT together on the same day, yes	60.3	47.9–72.7	66.7	48.8–84.6	54.5	36.6–72.5	.4402
Quit approach recommendationa
E-cigs more helpful	25.4	14.3–36.4	33.3	15.4–51.2	18.2	4.3–32.1	.3075
Quit medications more helpful	19.0	9.1–29.0	13.3	0.4–26.2	24.2	8.8–39.7
Using both e-cigs and quit medication together most helpful	34.9	22.8–47.0	36.7	18.4–55.0	33.3	16.4–50.3
Either one (similarly helpful)	9.5	2.1–17.0	3.3	0.0–10.2	15.2	2.2–28.1
Neither helpful	11.1	3.1–19.1	13.3	0.4–26.2	9.1	0.0–19.4
Smoking status at 3-mo survey (intent-to-treat analysis, missing response imputed as continued smoking)
Smoking abstinent 7+ d	34.4 (33/96)	24.7–44.0	41.3 (19/46)	26.5–56.1	28.0 (14/50)	15.1–40.9	.2005
Smoking and vaping status at 3-mo survey (3-mo survey respondents)
Smoking abstinent 7+ d, among respondents	50.0	37.6–62.4	59.4	41.4–77.4	41.2	23.7–58.6	.2179
E-cig abstinent 7+ d, among respondents	35.4	23.4–47.3	18.8	4.5–33.0	51.5	33.5–69.5	.0090
Smoking and e-cig status in last 7 d, among respondents
None (no e-cigs/smoking)	16.9	7.6–26.3	12.5	0.4–24.6	21.2	6.5–35.9	.0258
E-cigs only	33.8	22.0–45.7	46.9	28.6–65.2	21.2	6.5–35.9
E-cigs and smoking	30.8	19.2–42.3	34.4	17.0–51.8	27.3	11.2–43.3
Smoking only	18.5	8.8–28.2	6.3	0.0–15.1	30.3	13.8–46.9

CI = confidence interval; FDA = Food and Drug Administration; NRT = nicotine replacement therapy. All variables in this table were assessed 3 mo after enrollment. Three of the 66 participants who provided outcome data completed a brief version of the outcome survey. The brief survey included assessment of smoking and vaping status, but did not include other variables reported in this table.

Participants were asked which of five statements were most true for them. The five statements were: (1) Electronic cigarettes were more helpful to me for quitting than quit medications (such as the nicotine patch or Chantix). (2) Quit medications were more helpful to me for quitting than electronic cigarettes. (3) It was most helpful to me to use both electronic cigarettes and quit medications at the same time. I would recommend using both together. (4) Electronic cigarettes and quit medications were similarly helpful to me for quitting. I would recommend using either one. [ie, either work similarly well]. (5) Neither electronic cigarettes nor quit medications have been helpful when I have tried to quit smoking.

Among 3-month respondents, there were no significant group differences in recommendations for others when quitting (p = .31) with the highest proportion recommending using NRT and e-cigarettes together (34.9%), followed by e-cigarettes being most helpful (25.4%), and quit medications being most helpful (19.0%) (Table 2).

Relative Harm Beliefs

At 3 months, there were no significant differences between EEC and TAU on relative harm perceptions between cigarettes and e-cigarettes (p = .14) or between cigarettes and NRT (p = .14; Table 3). There were significant group differences in relative harm beliefs for e-cigarettes compared to NRT (p = .03) with more EEC than TAU indicating e-cigarettes were less harmful and more TAU than EEC indicating e-cigarettes were more harmful. Although not a significant difference (p = .19), a similar pattern of beliefs about relative harm of e-cigarettes and quit medications between groups was also present at baseline.¹⁹ The study was not powered to detect improvements in incorrect beliefs from baseline to follow-up.

Table 3.

Beliefs About Relative Harm of E-Cigarettes, Cigarettes, and Quit Medications at 3-mo Survey

	Total sample 3-mo survey respondents (n = 63)		Enhanced e-cig coaching respondents (n = 30)		Quitline treatment as usual respondents (n = 33)		p
	%	95% CI	%	95% CI	%	95% CI
“Compared to cigarettes, how harmful are e-cigs to a person’s health?”
Less harmful	69.8	58.2–81.5	73.3	56.5–90.1	66.7	49.7–83.6	.1403
About the same	17.5	7.8–27.1	20.0	4.8–35.2	15.2	2.2–28.1
More harmful	7.9	1.1–14.8	0	—	15.2	2.2–28.1
I prefer not to answer	4.8	0.0–10.2	6.7	0.0–16.1	3.0	0.0–9.2
“Compared to cigarettes, how harmful are quit medications like the nicotine patch to a person’s health?”
Less harmful	90.5	83.0–97.9	83.3	69.2–97.5	97.0	90.7–100.0	.1414
About the same	3.2	0.0–7.6	3.3	0.0–10.2	3.0	0.0–9.2
More harmful	1.6	0.0–4.8	3.3	0.0–10.2	0	—
I prefer not to answer	4.8	0.0–10.2	10.0	0.0–21.4	0	—
“Compared to quit medications like the nicotine patch, how harmful are e-cigs to a person’s health?”
Less harmful	25.4	14.3–36.5	40.0	21.4–58.6	12.1	0.4–23.9	.0281
About the same	22.2	11.7–32.8	16.7	2.5–30.8	27.3	11.2–43.3
More harmful	46.0	33.4–58.7	33.3	15.4–51.2	57.6	39.8–75.4
I prefer not to answer	6.3	0.2–12.5	10.0	0.0–21.4	3.0	0.0–9.2

CI = confidence interval. Variables in this table were assessed 3-mo after enrollment. Belief questions were not included in the brief outcome survey. Three of the 66 participants who provided outcome data completed a brief version of the outcome survey.

Smoking Cessation

Self-Reported 7-Day Quit Status

Respondent smoking 7dppa rates were 59.4% (19/32) for EEC and 41.2% for TAU (14/34) (p = .22) (Table 2). Among continued smokers, 51.6% reported smoking daily with no differences between groups (EEC: 50% [6/13]; TAU: 52.6% [10/20]; p = .99); in comparison, at baseline, 85% of the sample smoked daily.¹⁹ With missing data imputed as smoking, 7dppa rates were 41.3% (19/46) for EEC and 28.0% (14/50) for TAU (p = .20).

Using logistic regression, we first examined models that individually included age, HSI, gender, insurance status, and education as covariates to determine whether the inclusion of these variables impacted the relationship between treatment group and respondent quit status. The relationship remained stable with these covariates added. Similarly, when we added all five covariates in the model, the odds ratio (OR) for EEC group (OR = 2.1, p = .19) was consistent with the unadjusted model (OR = 2.1, p = .14).

Biochemical Verification of Quit Status

At 3 months, 33 participants (19 EEC and 14 TAU) self-reported 7dppa from smoking. Thirty participants (90.9%; 30/33) received their CO monitors. Twenty (67%; 20/30) participants did not provide CO readings (10 TAU and 10 EEC). Ten participants (seven EEC and three TAU) completed valid readings on 3 consecutive days. Five of seven EEC participants (26.3% of the 19 EEC participants who self-reported being quit) and two of three TAU participants (14.3% of the 14 TAU participants who self-reported being quit) recorded readings that confirmed smoking abstinence. Based on biochemically confirmed smoking abstinence, 10.9% (5/46) of EEC and 4.0% (2/50) of TAU were abstinent (p = .25); however, caution should be used in interpreting these data given the poor CO procedure compliance.

Vaping Status at 3 Months

At 3 months, vaping 7dppa rates were 18.8% (6/32) for EEC and 51.5% (17/34) for TAU (p = .01; Table 2). For continued vapers, 63.6% (14/22) of EEC and 93.8% (15/16) of TAU reported intentions to quit vaping in the future (p = .053). There were no significant differences by group for continued vapers with regard to primary reason for e-cigarette use at follow-up (p = .64): to quit smoking (35.9%, n = 14), to avoid returning to smoking (25.6%, n = 10), to cut down on smoking (23.1%, n = 9), to use when cannot smoke (12.8%, n = 5), and curiosity (2.6%, n = 1).

Side Effects Reported at 3 Months for E-Cigarettes and Quit Medications

Overall, among respondents who reported using e-cigarettes or quit medications over the 3-month study, 50% of e-cigarette users and 40% of NRT users reported a side effect. For e-cigarettes, a somewhat higher proportion of EEC (57.7%; 14/26) reported a side effect related to using an e-cigarette, compared to TAU (42.3%; 11/26), but this was not a significant difference (p = .40). The most commonly reported side effects were: 25.0% sore or dry mouth/throat (13/52; nine EEC and four TAU), 13.5% headache (7/52; six EEC and one TAU), 13.5% breathing difficulties (7/52; three EEC and four TAU), 9.6% cough (5/52; three EEC and two TAU), 7.8% dizziness (4/52; two EEC and two TAU), and 7.8% bad or vivid dreams (4/52; three EEC and one TAU). For NRT, similar proportions of EEC (40.0%; 10/25) and TAU (40.0%; 12/30) reported a side effect related to using a quit medication. The most commonly reported side effects were as follows: 12.7% nausea or stomach upset (7/55; one EEC and six TAU), 9.1% headache (5/55; three EEC and two TAU), and 7.3% skin irritation (rash, itchiness; 4/55; two EEC and two TAU).

Discussion

This study examined a behavioral treatment for dual users of cigarettes and e-cigarettes at the time of state quitline enrollment. The intervention was feasible to implement in this real-world quitline setting and acceptable to participants, as indicated by similar or better participant satisfaction and call completion for the EEC group relative to TAU.

The OR for the impact of the EEC intervention on smoking cessation outcomes was large (2.1), but not statistically significant. The OR was also stable, regardless of other potentially important variables included in the model. Taken together, these are good indicators that the EEC intervention merits examination in a larger trial.

Although no differences between groups were shown in the adapted CollaboRATE on participants’ quit plan development experience, EEC participants were more likely to report that coaches took e-cigarette use into account during quit plan development, and spent enough time talking about e-cigarettes. No differences were found in beliefs about e-cigarettes, NRT, and smoking as measured in this study. This was unexpected given that the theory of reasoned action, which posits the need to change beliefs prior to changing behavioral intentions,^22–24 was an underpinning of the EEC approach. We did not find differences in beliefs, but we did find the behaviors and outcomes that we expected based on education and intended beliefs change. It may be that short- and long-term beliefs that drive product utilization choices are more complex and were not adequately measured with the survey questions used in this study. Moreover, although education and beliefs were addressed, most prominently on the first coaching call, this may have been a relatively small dose of education compared to other environmental inputs during the study timeframe, particularly family and friends, and regular news coverage about vaping, the rise of JUUL and the EVALI epidemic starting fall 2019.^31–33

Most EEC participants choose to use both NRT and an e-cigarette for their quit plan. The EEC intervention did not reduce NRT utilization; both groups received NRT and reported use of NRT at similar rates. Despite TAU participants being advised to quit e-cigarettes when starting NRT, both groups most frequently recommended that others use NRT and e-cigarettes together, supporting the assertion that smokers are using e-cigarettes to quit smoking regardless of professional recommendations. These data align with a recent Cochrane Review concluding that e-cigarettes may be as effective or more effective than NRT for smoking cessation.³

One contribution of this study is insight provided into how smokers utilize e-cigarettes when combined with NRT. Using combination NRT (typically a long-acting form such as the patch together with a short-acting form such as gum or lozenge) can improve quit outcomes.¹² Nicotine gum and lozenge are often used incorrectly or disliked by users.³⁴ E-cigarettes may be a more palatable and/or enjoyable short-acting replacement option. We did not control the utilization of NRT, including combination NRT, with e-cigarettes, but allowed participants to choose their own quit plan. Participants opted for a variety of options, including e-cigarette plus patch, and e-cigarette plus patch plus gum/lozenge. More research is needed on the relative effectiveness of combining e-cigarettes with various NRT options.

Arguments against e-cigarettes as a harm reduction aid include concerns that people will become dual users of cigarettes and e-cigarettes, rather than effectively using e-cigarettes to completely stop smoking.³⁵ This study demonstrated the feasibility of developing a tailored behavioral intervention directly addressing e-cigarettes as a harm reduction strategy and quitting aid with an emphasis on the harms of smoking and the need to completely quit, and, in doing so, demonstrated promising smoking abstinence rates.

Finally, vaping cessation was not a primary goal of the EEC treatment. Given the short follow-up time period (3 months), we expected participants would still be working towards stable cigarette abstinence and would not be ready to focus on vaping cessation. At follow-up, the primary reported reasons for vaping were to quit smoking and prevent relapse to smoking. At 3 months, half of TAU participants reported abstinence from e-cigarettes for 7 or more days compared to one-fifth of EEC participants. Differences in intentions-to-quit vaping between the two groups indicated that the EEC intervention has opportunities to improve education and intervention to continue supporting participants to achieve complete nicotine abstinence. Future studies should examine specific behavioral and NRT support for completely quitting e-cigarettes for those who are securely quit from smoking.

It is notable that although more participants were vaping at follow-up in the EEC group, a substantial number of TAU participants were also vaping at follow-up despite being advised to quit by coaches. Although not discussed during coaching calls, it is likely that at least some participants were attempting to use their e-cigarettes as part of their quit smoking attempt. Complementing recent findings from a self-help intervention for dual users,¹⁴ the promising smoking cessation rates seen in the EEC group indicate that the tailored behavioral support and education appeared to provide some benefit to those using e-cigarettes to quit smoking. This finding should be confirmed in a larger study.

This study had several limitations. First, participants were recruited from one state, Oklahoma, and needed to own an Android smartphone for study procedures; findings may not be generalizable to other populations. Second, the EEC intervention was intended to support individuals already using and planning to use e-cigarettes during their quit attempt. This treatment would need to be examined in other populations to determine its appropriateness and utility (eg, whether the EEC is acceptable to and effective for individuals not already using e-cigarettes at the start of treatment). Third, we experienced real-world challenges equalizing call connection logistics for the TAU and EEC groups, which may have led to fewer EEC participants completing their first coaching call to remain in the study. A future, larger study would allow for training a higher number of coaches on the EEC intervention, which could ameliorate this limitation. Fourth, as with all research trials that include compensation for completing study surveys, participant payment could influence an individual’s participation or outcomes. Fifth, our retention rates at the 3-month survey were approximately 70% in both groups. While a higher retention rate is desirable, this response rate is in line with other quitline research trials. Sixth, daily monitoring of substance use can lead to decreased use. Daily survey completion may have served as an intervention in itself and influenced quit rates in this trial. Both treatment groups completed daily surveys; therefore, we expect impacts were similar for both groups. Future real-world trials may want to exclude daily surveys unless they are a component of standard treatment. Finally, we utilized a smartphone-connected CO monitor for remote biochemical verification of quit status to explore feasibility of this procedure. Participants had challenges with the smartphone app required for the CO readings and completed CO readings at a low rate, suggesting limited feasibility of these methods for remote biochemical verification of smoking-specific abstinence in this real-world population. Our experience aligns with recognized feasibility challenges with biochemical verification of quit status in fully remote trials.³⁶ In future trials, we will explore including a biochemical verification option that does not require use of a smartphone app. Similarly, we did not analyze the daily surveys delivered via the smartphone app due to participant challenges using the app and poor completion. To expand the eligible sample in future trials, we would remove the requirement to have access to an Android smartphone.

Conclusion

This study piloted a brief behavioral intervention with high potential for dissemination aimed at assisting dual users of e-cigarettes and cigarettes to completely quit smoking. The new treatment was acceptable and feasible, and evidenced promising engagement and smoking cessation rates. Findings indicate support for a fully powered randomized controlled trial to test the intervention. A future, fully powered trial should include a longer follow-up timeframe and examine a staged intervention to help individuals who successfully quit cigarettes to also quit vaping.

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 supported by the National Institute on Drug Abuse (R21DA042960 to KAV) and used the InsightTM mHealth Platform and Android smartphone app, which was supported by the National Cancer Institute Cancer Center Support Grant P30CA225520 awarded to the University of Oklahoma Stephenson Cancer Center. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health or Optum.

Declaration of Interests

KAV, KMC, LNM, JMH, and KAW are employees of Optum, the provider of quitline services for the Oklahoma Helpline in this study. THB is on the Advisory Board of Hava Health, Inc., a start-up (with no connections to the tobacco industry) that is developing a therapeutic e-cigarette. All other authors declare no conflicts of interest.

Acknowledgments

The authors would like to gratefully acknowledge the Optum quitline staff from operations, technology, reporting, and product teams who supported the implementation of this trial within standard program services, particularly the research quit coaches who delivered the study intervention. We would also like to thank the Oklahoma Helpline for collaborating on this research.

Data Availability

The data underlying this article will be shared on reasonable request to the corresponding author.