Effects of E-cigarettes on Combustible Cigarette Smoking Among Adults With Opioid Use Disorder on Buprenorphine: Single Arm ERASER Pilot Trial

Abstract

Introduction

People with opioid use disorder (OUD) on buprenorphine smoke at high rates and have low cessation rates, even with evidence-based medications. Electronic cigarettes (EC) are a promising harm reduction strategy for combusted cigarette (CC) smokers unable to quit. Unfortunately, people with OUD are underrepresented in EC research.

Aims and Methods

A pilot study assessed the feasibility, acceptability, and preliminary effectiveness of EC as a harm reduction tool among CC smokers with OUD on buprenorphine (N = 30). Participants were provided with an EC and freebase nicotine liquid (6 mg/mL) with a choice of flavor and a brief training session. Research visits were scheduled in person at baseline, week 4, and week 8 (follow-up). Daily diary assessments were completed during the 4-week EC period.

Results

Most visits (>74%) and 61.4% of daily diary assessments were completed. During the 4-week study period, 90% of participants used the EC at least one day, 66.7% used the EC for at least 16 days, and 43.3% used the EC every day. Significant reductions were observed between baseline and both weeks 4 and 8 in cigarettes smoked per day (CPD_baseline = 16.2[8.3], CPD_{week4 =} 9.6[9.3], CPD_{week8 =} 8.4[8.3]) carbon monoxide (CO) levels (CO_baseline = 21.5[15.0], CO_{week4 =} 16.9[9.6], CO_{week8 =} 15.7[10.0]), and nicotine dependence measured using the Fagerström Test for Nicotine Dependence (FTND_baseline = 5.4[2.5], FTND_week4 = 4.2[2.6], FTND_week8 = 4.4[2.6]), with all p-values < .05.

Conclusions

Implementing an EC protocol in outpatient maintenance treatment programs is feasible and acceptable. Preliminary results suggest that ECs may facilitate reductions in cigarettes per day, CO levels, and nicotine dependence. Future research should explore the effect of prolonged EC use on harm reduction and cessation milestones.

Implications

EC are a potentially promising harm reduction strategy for adult CC smokers with OUD on buprenorphine who are unable to quit using evidence-based medications. However, previous studies have largely overlooked people with OUD on buprenorphine with recent drug use. This study addresses this gap through a pilot trial investigating the feasibility, acceptability, and preliminary effects of EC on CC behavior. The brief and standardized nature of the protocol and its implementation in outpatient settings highlights its potential for widespread implementation in facilities providing care to people with OUD on buprenorphine.

Introduction

Opioid use disorder (OUD) and combustible cigarette (CC) smoking are public health burdens in the United States, which co-occur at high rates. Buprenorphine maintenance represents an effective treatment for OUD, proven to reduce opioid and other illicit drug use as well as the risk of fatal and non-fatal overdose.¹ Despite the potential for successful long-term recovery from OUD with maintenance treatment, over 75% of people on maintenance treatment continue to smoke CC,² putting people with OUD at a disproportionately increased risk of tobacco-related morbidity and mortality.³ Although ~80% of CC smokers with OUD have attempted to quit,⁴ cessation rates are extremely low, even with Food and Drug Administration (FDA) approved medications.² Therefore, alternative strategies to address CC smoking in people with OUD are urgently needed.

Electronic cigarettes (EC) represent a promising harm reduction strategy for CC smokers unable to quit smoking with FDA-approved medications.⁵ Evidence from randomized clinical trials and meta-analyses among CC smokers without OUD supports the use of EC as a tool to reduce cigarettes per day (CPD) and biomarkers of exposure and potential harm,⁶ as well as in facilitating switching away from CC.⁵ Cumulative evidence continues to show that EC may be effective in promoting biochemically validated CC abstinence,^7,8 even more so than FDA-approved nicotine replacement therapy.⁹ An important gap in the literature is that people with OUD on buprenorphine are underrepresented in clinical studies testing EC as a harm reduction tool. To our knowledge, only one prior pilot study has shown that EC, with or without nicotine, can promote self-reported CPD reduction among CC smokers maintained on buprenorphine.¹⁰ Although this study reported high self-reported adherence to the EC (80.6%–91.2%), the study had several limitations, including the exclusion of CC smokers with illicit substance use in the prior 90 days, the very low amount of e-liquid used (~3 mL/week), the absence of CO reductions, and the lack of information on switching rates. These limitations underscore the need for additional studies focusing on people with OUD on buprenorphine treatment. The present pilot study aims to address these gaps in the literature by exploring the feasibility, acceptability, and preliminary effectiveness of an EC protocol delivered for CC smokers with OUD on buprenorphine treatment.

Methods

Design and Setting

This single-arm pilot study, called Electronic cigaRettes to address combustible cigarette Smoking for pEople with opioid use disordeR (ERASER), recruited participants from an outpatient clinic providing maintenance treatment for OUD at the Prisma Health Campus in Greenville, South Carolina. This study was approved by the Institutional Review Board of Prisma Health and registered at clinicaltrials.gov (NCT06277271). All research activities were conducted in accordance with the Declaration of Helsinki’s ethical principles for medical research involving human subjects.

Participants

The sample of this pilot study comprised 30 CC smokers with OUD on buprenorphine treatment. Participants were recruited through an outpatient addiction clinic from June 2023 to February 2024. Eligibility criteria included: (1) age 21 or older, (2) being a current CC smoker, (3) smoking ≥5 CPD for ≥1 year, (4) OUD diagnosis and currently receiving buprenorphine, (5) interested in switching to EC, (66) plan to remain on buprenorphine for the duration of the study, and (7) have a smartphone with internet access. Individuals were excluded if they (1) used other tobacco or nicotine products, including EC, on ≥9 days in the past 30 days; (2) had used smoking cessation medication in the prior 30 days; (3) were pregnant or breastfeeding; (4) had a current medical or psychiatric condition that would contraindicate participation; or (5) were unable to speak/read English. All participants were recruited via medical staff referral, word of mouth, and flyers posted in the clinic. Additionally, potentially eligible participants were approached before or after their routine clinical visits with their providers while waiting in designated waiting areas.

Procedures

Potentially eligible participants received an overview of the study overview, gave oral consent for screening, and completed a standardized screening battery to assess preliminary eligibility. Those potentially eligible after the initial screening underwent the informed consent process and were scheduled for a baseline visit. Screening and baseline visits frequently occurred on the same day. During the baseline visit, participants completed a battery of assessments, including surveys that collected sociodemographic information and a brief clinical history covering tobacco use, substance use, and treatment-related variables. Additionally, saliva-based toxicology and carbon monoxide (CO) samples were collected. After all baseline measures were collected, participants received a brief education regarding the use of the study EC, as well as a brief training on how to complete the daily diary surveys and participate in the brief weekly video calls.

EC and EC Protocol

EC Device and Liquid

The EC used in this study was the RPM 4 EC device (SMOK, China), which featured a 5 mL refillable pod (dimensions 103 × 25 × 27.9 mm, 102 g, 1650 mAH, 3.3–4.2 V) and adjustable power settings ranging from 5 to 60 W. The EC was supplied with both 0.23 and 0.4 Ω coils, enabling participants to choose their preferred option. Participants were provided with additional pods and coils at no cost in the event of a malfunction. The EC liquid used was a freebase nicotine solution (6 mg/mL) containing 70% propylene glycol and 30% vegetable glycerin. The nicotine content of the EC liquid and the pharmacokinetic profile of the combination of EC device and liquid were not independently evaluated during the study. Participants were provided with a 4-week supply of EC liquid equivalent to 150% of their daily smoking rate (Supplementary Material). This amount was calculated based on self-reported daily smoking rate at baseline. Briefly, assuming a single CC delivers 1 mg of nicotine and the EC liquid had a concentration of 6 mg/mL, we estimated that 0.17 mL of liquid would provide a similar nicotine amount as one CC. To calculate daily EC liquid needs, we multiplied 0.17 mL by 150% of the daily cigarettes smoked (CPD). For example, a smoker with 10 CPD would need 2.55 mL of e-liquid per day (15 [150% of 10] × 0.17 mL). Participants were given the option to select the flavor of EC liquid, with 56.6% opting for a sweet-cooling flavor, 23.3% for a sweet non-cooling flavor, 16.7% for a tobacco flavor, and 3.3% for a menthol-flavored EC liquid.

EC Education Session

At baseline, once all assessments were completed, trained research personnel conducted a brief educational session lasting between 15 and 25 minutes. This session aimed to educate participants on (1) CC and EC use, (2) instructions for proper EC use, and (3) practice puffing. The information provided included the health effects of both CC and EC, the benefits of switching, and tips for substituting CC with EC, including encouraging daily use, keeping the EC always handy, and substituting CC with EC use sessions.

Instructions on the proper use of the EC device were provided, covering several key points. These included device operation (eg, starting and charging the device, filling the EC with EC liquid), safety precautions (eg, proper storage of the device and liquid, avoiding exposure to high temperatures, keeping EC and liquid away from children and pets), and troubleshooting (eg, addressing common issues such as leaks and burnt coils, and how to resolve these issues). Participants received instructions on how to effectively inhale using the EC. Participants were taught to take longer, slower puffs and to reduce the number of puffs throughout the day. They were encouraged to engage in multiple short sessions of EC use rather than taking more frequent puffs during a single session, as they might have done with conventional cigarettes. They were also given the opportunity to practice their puffing technique under the observation of the research coordinator, who provided feedback on their puffing behavior. At the 8-week research visits, participants were asked to return all unused EC liquid and the study EC.

Assessments

Assessment Schedule

Research visits were conducted in person at baseline, at week 4 (end of treatment) and 8 (follow-up). These visits took place in a designated research space within the same building as the outpatient clinic. Whenever possible, research visits were scheduled to coincide with clinical appointments to reduce participant burden. Brief (≤30 minutes) video call visits were conducted weekly (weeks 1 to 4), during which participants answered questions related to their use of CC, EC, and other drugs, and self-collected biological samples. Research staff also checked whether participants had any issues with the EC and liquid and whether they had an adequate supply of EC liquid.

During weeks 1 to 4, participants were asked to complete a daily diary survey each night for 28 consecutive nights, assessing both use and craving for CC, EC, opioids, and other drugs. Diary surveys were sent every evening (between 5:00 and 10:00 PM) via an auto-send text or email, depending on participants’ preferences, with a link to a secure REDCap survey. Self-reported research data were collected through interview assessments or self-reported, depending on the instrument, and entered in REDCap, a secure and HIPAA-compliant web application for managing databases. All research assessments were conducted by members of the research team who were independent of the clinical staff at the outpatient clinic.

Measures

At baseline, participants completed a battery of assessments that collected basic demographic data (eg, sex, age, educational level, income, and living situation), as well as information on tobacco use, opioid and other drug use, and treatment-related variables.

Tobacco-related variables encompassed the use of CC, EC, and other tobacco products. Participants were asked about the number of CPD, age of smoking onset, years of regular smoking, and the number of prior quit attempts. Assessments related to EC included lifetime and recent use of EC, and reasons for prior EC use. The Fagerstrom Test for Nicotine Dependence (FTND)¹¹ and the Minnesota Nicotine Withdrawal Scale¹² were administered at baseline, weeks 4 and 8 to assess nicotine dependence and withdrawal symptoms, respectively.

Participants completed a brief clinical history that assessed recent drug use, as well as information related to their OUD treatment, including dosage and time on that medication.

Biological samples included saliva-based toxicology screens and CO samples (iCO Smokelyzer, Bedfont Scientific Ltd), which were collected at baseline, during each remote weekly visit, and at the in-person visits for weeks 4 and 8. Saliva-based toxicology screens were completed for buprenorphine, oxycodone, opiate, barbiturate, benzodiazepine, methadone, methylenedioxymethamphetamine, amphetamine, methamphetamine, and cocaine (Multi-Drug Screen Dip Card 10 Panel; American Bio Medica Corp., Kinderhook, NY).

In the daily diary surveys (3–5 minutes; 12 questions), participants were prompted to complete a survey that covered EC use (frequency of use and number of puffs in the EC), CC use (number of cigarettes smoked that day and the previous day), and drug use (yes/no for that day and the previous day), as well as craving for EC, CC, and drugs. Craving was assessed using a single 7-point Likert scale question, measuring the intensity of cravings over the past 24 hours. During each weekly follow-up, participants’ EC use was assessed, nicotine withdrawal was measured, daily use was encouraged, barriers and facilitators were discussed, and any issues were addressed.

Outcome Measures

Feasibility

Intervention feasibility was measured as the proportion of enrolled participants out of the total participants who were eligible; the percentage of participants who completed weekly, week 4 and 8 visits; the percentage of CO and saliva tox assessments completed; and, the percentage of participants lost to follow-up. Daily diary feasibility was measured by the average percentage of assessments completed, and the percentage of participants who responded to at least 50% and 75% of daily diary assessments.

Acceptability

Acceptability was examined based on participants’ self-reported use and ratings. As in a recent EC trial,⁷ the percentage of participants who tried the EC at least once, used it regularly (at least 16 out of 28 days), and used it daily (28 out of 28 days) was measured. Overall satisfaction with the EC, the study, and the EC orientation session received at the baseline visit were measured at week 4 on a 5-point scale ranging from very satisfied to very dissatisfied. In weeks 4 and 8, participants were asked about the average amount of EC liquid used on the days they used the study EC.

Smoking-Related Outcomes

Outcomes were drawn from prior and ongoing studies testing EC^7,13,14 and included changes in self-reported CPD, CO levels, and nicotine dependence (FTND scores) from baseline to weeks 4 and 8. Secondary outcomes¹⁵ included the percentage of participants who partially (ie, ≥50% CPD reduction and self-reported EC use in the prior 7 days) and completely (ie, self-reported 7-day CC abstinence biochemically confirmed [CO < 6 ppm^16,17]) switched to EC at weeks 4 and 8. Additionally, participants were asked to estimate their average daily EC liquid in mL over the past 7 days at both weeks 4 and 8.

Statistical Analyses

Sample Size Determination

This pilot study was not designed to have sufficient power to detect statistically significant differences, but rather to determine feasibility and acceptability outcomes to be tested in a larger-scale efficacy trial.¹⁸ A sample size of 30 was determined based on consideration of the feasibility of recruitment within the study timeframe and recommendations for pilot trials.¹⁹

Statistical Analyses

Smoking-related outcomes data were analyzed following the intent-to-treat (ITT) approach. Descriptive statistics were conducted to analyze participants’ baseline characteristics, as well as feasibility, acceptability, and efficacy data. Repeated measures of analysis of variance (ANOVA) were conducted to explore changes in CPD, CO, and FTND scores from baseline to weeks 4 and 8. Tukey’s HSD post hoc analyses were conducted to compare changes between baseline and weeks 4 and 8. Effect sizes were calculated based on Cohen’s d, with results classified as small (d = 0.2), medium (d = 0.5), and large (≥0.8). Missing data related to CC smoking (ie, CPD, CO, and FTND) was handled using the most conservative approach for addressing missing data²⁰ and replaced with baseline values. Finally, sensitivity analyses were conducted using mixed-effects models to examine the effect of time and frequency of EC use, categorized as (1) fewer than 16 days, (2) 16 days to 27 days, and (3) all 28 days, throughout the 28-day study period on CPD, CO levels, and FTND scores. Analyses were conducted using SPSS version 28.

Results

Participants’ baseline characteristics are presented in Table 1. Briefly, half of the participants were female and, on average, 44.5 years of age. Most participants were White (93.3%) and non-Hispanic (93.3%). The sample included 23.3% who were unemployed, 20% who did not have their own method of transportation, and 85% who had an annual household income below $35k. Participants smoked, on average, 16.2 CPD with an average of 1.7 lifetime quit attempts. The majority had a drug screen positive at the baseline visit (86.7%) with the most common substance being methamphetamine (46.7%) and amphetamine (40.0%); 40% had a screen positive for more than one substance.

Table 1.

Participant Baseline Characteristics

Characteristic	M (SD)/N(%)
Demographics
Age (years)	44.5 (10.4)
Sex (female)	15 (50)
Race
Black	1 (3.3)
White	28 (93.3)
American Indian	1 (3.3)
Other	1 (3.3)
Ethnicity (Hispanic/Latino)	2 (6.7)
Highest educational attainment
≤Some high school	6 (20)
High school diploma/GED	12 (40)
≥Some college	12 (40)
Employment status
Employed	12 (40)
Unemployed	7 (23.3)
Receiving social security	7 (23.3)
Other	4 (13.3)
Method of transportation
Car (own or borrowed)	20 (66.7)
Public transportation/ride from others	6 (20)
Other	4 (13.3)
Annual household income
≤$14 999	13 (43.3)
$15 000-24 999	6 (20.0)
$25 000-34 999	5 (16.7)
$35 000–44 999	3(10.0)
$50 000–74 999	1(3.3)
Missing	2(6.7)
Smoking characteristics
Cigarettes per day at baseline	16.2 (8.3)
CO levels (ppm)	21.5 (15.0)
Age first started smoking cigarettes	15.5 (3.4)
Nicotine dependence (FTND)	5.4 (2.5)
Years smoking	24.3 (11.4)
Prior quit attempts	1.7 (1.3)
OUD treatment characteristics
Buprenorphine dose (mg/day)	23.1 (4.0)
Time on buprenorphine (months)	15.4 (15.4)
Self-reported drug use at baseline
Alcohol	9 (30)
Opiates	5 (16.7)
Barbiturates/sedatives	4 (13.3)
Amphetamines	14 (46.7)
Cocaine	4 (13.3)
Cannabis	11 (36.7)
More than one substance	8 (26.7)
Drug use at baseline (saliva tox)
Buprenorphine	22 (73.3)
Oxycodone	1 (3.3)
Opiates	1 (3.3)
Barbiturates/benzodiazepine	3 (10.0)
Methylenedioxymethamphetamine	2 (6.7)
Amphetamines	12 (40.0)
Methamphetamine	14 (46.7)
Cocaine	2 (6.7)
More than one substance	12 (40)

GED = general education diploma; FTND = Fagerstrom test for nicotine dependence; MNWS = Minnesota nicotine withdrawal scale; CO = carbon monoxide; ppm = parts per million; More than one substance excludes buprenorphine.

Feasibility

Figure 1 diagrams participant flow. Out of the 84 individuals screened, 61 met the inclusion criteria, and 30 agreed to participate. Thus, the recruitment rate was 49.2% (30/61). On average 74.1% (2.9/4) of remote weekly visits were completed, and 80% (24/30) and 76.6% (23/30) of participants completed the week 4 and week 8 visits, respectively. Most of the CO (76.2% or 160/210) and drug screen (78.6% or 165/210) assessments were completed, with 23.3% (7/30) of participants lost to follow-up at week 8. On average, 17.4 (SD = 8.8) or 61.4% (SD = 31.4) of daily diary assessments were completed with 70% (21/30) responding to at least half and 40% (12/30) to ≥75%.

Figure 1.

The flow of participants through the study. Participants were smokers with opioid use disorder on buprenorphine recruited locally in Greenville, SC.

Acceptability

The percentage of trial days with EC use ranged from 0% to 100% (M[SD] = 70.9% [4.5]) during the 4-week study period. At week 4, 90% (27/30) reported having used the EC at least one day, 66.7% (20/30) used the EC for at least 16 days, and 43.3% (13/30) used the EC every day. At week 4, most (95.8%) reported being very satisfied with the study and the EC (79.2%), and the brief orientation session (82.6%). On average, participants reported using 1.9 (SD = 1.9) and 1.3 (SD = 1.1) mL of EC liquid per day at weeks 4 and 8, respectively.

Smoking-Related Outcomes

Repeated measures ANOVA showed that mean self-reported CPD varied significantly across weeks (F(2,28) = 15.581, p < .001, η² = 0.51). Post hoc comparisons indicated that there was a statistically significant reduction in mean self-reported CPD from baseline to week 4 (M(SD) =16.2(8.3) vs. 9.6(9.3), p < .001,d = 0.89), and to week 8 (16.2(8.3) vs. 8.4(8.3), p < .001,d = 0.92). Similarly, there was a significant time effect (F(2,28)=3.811, p = .034, η² = 0.21) indicating that CO levels changed across baseline and weeks 4 and 8. Post hoc analyses revealed significant decreases in CO levels from baseline to week 4 (21.5(15.0) vs. 16.9(9.6), p = .005,d = 0.50), and from baseline to week 8 (21.5(15.0) vs. 15.7(10.0), p = .014,d = 0.42). Repeated measures ANOVA showed that participants’ mean FTND scores changed significantly across baseline and weeks 4 and 8 ((F(2,28) = 7.955, p = .002, η² = 0.36). Mean FTND scores significantly decreased from baseline to week 4 (baseline to week 4, 5.4(2.5) vs. 4.2(2.6), p < .001,d = 0.62) and week 8 baseline to week 8 (5.4(2.5) vs. 4.4(2.6), p = .002,d = 0.59).

Half (15/30) of participants partially switched and one (3.3%) completely switched at weeks 4 and 8, and 53.3% (16/30) of participants achieved ≥50% reduction in CPD from baseline to weeks 4 and 8. Repeated measures revealed no significant changes (F(2,20) = 0.816, p = .456, η² = 0.07) in the mean Minnesota Nicotine Withdrawal Scale scores across visits (baseline 15.5[11.1], week 4 15.0[10.0], and week 8 13.0[10.3]).

Sensitivity Analyses

Although the effect of time (baseline, week 4, week 8) on CPD (F(2,48) = 17.325, p < .001, η² = 0.41) and FTND scores (F(2,48) = 6.156, p = .004, η² = 0.20) was significant, the effect of frequency of use or the interaction between frequency of use and time on any of these two measures was not (Figure 2, A and C). In addition, there was no significant temporal change in CO levels, nor was there a significant effect of frequency of use, or of interaction between frequency of use and time on CO levels (Figure 2B).

Figure 2.

Changes over time in cigarettes per day (A), CO levels (B), and Fagerstrom Test for Nicotine Dependence scores (C) by frequency of electronic cigarettes use throughout the 28-day study period: fewer than 16 days (n = 7), 16 days to 27 days (n = 7), and (c) all 28 days (n = 13). The figure displays the means, which are connected by lines. Error bars represent the standard error of the mean. Some y-axes are presented on a smaller scale to permit detailed inspection of the data.

Discussion

This pilot study evaluated the feasibility, acceptability, and potential of EC as a harm reduction approach for CC smokers with OUD on buprenorphine. We highlight the following findings: (1) the EC protocol was feasible within outpatient settings; (2) the EC and EC protocol were well accepted, based on the self-reported ranges and high rates of use; and (3) EC promoted reductions in CPD, biomarkers of exposure to harm (CO levels), and CC dependence in a population disproportionally affected by the harms of CC smoking.

The EC protocol was feasible, as evidenced by the high rates of completion for both in-person (76.6%–80%) and remote (74.1%) visits, along with the collection of biological samples (76.2%–78.6%). Our attrition rate (23.3%) was lower than in earlier trials involving CC smokers with OUD on buprenorphine receiving non-contingent reinforcement (38.7%)²¹ or bupropion (42%) for addressing CC smoking.²² People on maintenance treatment often face inequitable access due to social determinants of health (SDoH), such as unreliable transportation, unemployment, or unstable housing, which are known to influence health and OUD treatment outcomes.²³ Protocols such as the one proposed in this study, which are hybrid (online and in-person) and can be delivered in the same settings where people receive OUD care, have the potential to overcome some SDoH-related barriers. Furthermore, the brief, standardized, and easily trainable components of the protocol highlight its potential for broad implementation and maintenance within settings providing care to people with OUD on maintenance treatment. It should be noted that the average completion rate of daily diary assessments was suboptimal (61.4%) compared to what is assumed to be acceptable (ie, ≥80%)²⁴ but closer to the completion rates found in a systematic review exploring daily diary completion rates in populations with substance use disorders (ranging from 66.2% to 77.8%).²⁵ Future studies aiming at including these assessments should implement measures to increase completion including frequent monitoring, specialized apps to collect the data, or immediate reimbursement for assessment completion.

The high rates of EC use and high self-reported ratings suggest that the EC and the EC protocol are well accepted among CC smokers with OUD on buprenorphine. These findings are consistent with earlier studies involving CC smokers on buprenorphine that reported high rates of EC use as a cessation aid and high interest in using EC to reduce their CC smoking.^26,27 In our study, 90% tried the EC at least once, 66.7% used the EC frequently and 43.3% used the EC every day for the 4-week study period. Moreover, the average volume of EC use was higher (1.9 mL/day) than the values reported in an earlier pilot study testing EC acceptability among people on maintenance treatment (~3 mL/week).¹⁰ Cessation outcomes using FDA-approved medications for smoking cessation have been suboptimal among people with OUD, perhaps due to the low adherence rates.² In our study, rates of EC uptake and self-reported acceptability were relatively high, highlighting the potential for implementation as a therapeutic approach among people with OUD on maintenance treatment.²⁸

This is the first study to demonstrate the potential of EC in promoting significant CPD reductions, supported by CO reductions, and reducing CC dependence in a sample of CC smokers with OUD on buprenorphine treatment. Felicione et al.¹⁰ reported that self-reported CPD reduced significantly, but CO reductions did not match their CPD reductions, and no information on nicotine dependence was provided. Although a reduction in CPD is not a traditional outcome, it holds potential as a clinical outcome¹⁷ as it may indicate a decrease in toxicant exposure²⁹ and is sometimes the first step toward cessation.³⁰ Our study shows that EC represents a viable and promising tool to reduce daily CC smoking, as objectively verified by CO decreases, and to reduce CC dependence in a population disproportionately burdened by tobacco-related harms and frequently underrepresented in EC research. While EC are not an FDA-approved device to quit smoking,³¹ of our study findings add to the growing body of literature supporting the potential of EC as a tool to address CC smoking.^5,6,8

Our study findings should be interpreted with caution due to various limitations. The pilot nature of this study inherently includes a small sample size, and thus, potential type II errors of estimated large effect sizes may not be ruled out. Second, no parallel comparison condition, active or control, was considered. Third, participants were recruited from an urban area and were mostly white, so our findings may not be generalizable to people with OUD on maintenance treatment living in rural areas and racial and ethnic minority smokers. Fourth, the e-liquid was not tested to verify nicotine levels or other chemical contents. Additionally, the chosen combination of device and liquid was not tested to determine its pharmacokinetic profile. Fifth, we used only a single combination of EC/liquid, so our results may not generalize to other EC devices available on the market. Finally, the follow-up period (8 weeks) is shorter than the recommendations for cessation trials (26 weeks).¹⁷

In summary, the ERASER pilot study provides preliminary evidence of the acceptability and potential of EC as a harm reduction tool by facilitating CPD reductions among CC smokers with OUD on buprenorphine treatment with recent drug use. Furthermore, this pilot study evidences that integrating an EC protocol with in-person and remote components, including self-reported assessments and objective verification of smoking and drug use, involving people with OUD recruited from outpatient maintenance clinics is feasible. Future research is warranted to investigate the effect of prolonged EC use on tobacco use behavior over longer follow-ups.

Supplementary material

Supplementary material is available at Nicotine and Tobacco Research online.

Funding

This project was supported in part by the Clemson Faculty SUCCEEDS (CU SUCCEEDS) funding program, the Creative Inquiry (CI) program at Clemson University, and the Prisma Health Addiction Medicine Center.

Declaration of Interest

None declared.

Author Contribution

Irene Pericot-Valverde (Conceptualization [lead], Formal analysis [equal], Funding acquisition [lead], Investigation [lead], Methodology [lead], Writing—original draft [lead], Writing—review & editing [lead]), Moonseong Heo (Conceptualization [supporting], Data curation [equal], Formal analysis [supporting], Funding acquisition [supporting], Investigation [supporting], Methodology [supporting], Supervision [supporting], Writing—review & editing [supporting]), Shadi Nahvi (Conceptualization [supporting], Methodology [supporting], Writing—original draft [supporting], Writing—review & editing [supporting]), Justin Barron (Data curation [lead], Investigation [supporting], Methodology [supporting], Project administration [supporting], Writing—review & editing [supporting]), Sarah Voss (Data curation [supporting], Investigation [supporting], Project administration [supporting], Writing—review & editing [supporting]), Erik Ortiz (Investigation [supporting], Methodology [supporting], Project administration [supporting], Writing—review & editing [supporting]), Diann Gaalema (Conceptualization [supporting], Investigation [supporting], Writing—original draft [supporting], Writing—review & editing [supporting]), James Thrasher (Conceptualization [supporting], Investigation [supporting], Writing—original draft [supporting], Writing—review & editing [supporting]), Abigail Batchelder (Conceptualization [supporting], Writing—original draft [supporting], Writing—review & editing [supporting]), Kaileigh Byrne (Conceptualization [supporting], Funding acquisition [supporting], Writing—original draft [supporting], Writing—review & editing [supporting]), Deborah Kunkel (Funding acquisition [supporting], Writing—original draft [supporting], Writing—review & editing [supporting]), and Alain Litwin (Conceptualization [supporting], Funding acquisition [supporting], Resources [lead], Writing—original draft [supporting], Writing—review & editing [supporting])

Data Availability

The protocol, consent form, statistical plan, data, and other relevant study materials will be made available upon reasonable request.