Pharmacological Interventions for Electronic Cigarette Cessation: A Systematic Review and Meta-analyses

Abstract

Objectives:

The rising global prevalence of electronic cigarette (e-cigarette) use poses significant public health concerns, including nicotine dependence. This systematic review and meta-analysis evaluated the efficacy and safety of pharmacological interventions for e-cigarette cessation.

Methods:

We systematically searched 5 databases, including Cochrane Library, PubMed, Ovid Embase, Sinomed, and Web of Science Core Collection, up to May 1, 2024, that evaluated the efficacy and safety of pharmacological intervention compared with placebo for e-cigarette cessation. Two investigators independently screened studies, extracted data, and assessed the risk of bias. The outcomes of interest were (1) continuous abstinence rate (CAR); (2) 7-day point prevalence of e-cigarette abstinence at different weeks; and (3) adverse effects. This study was registered with PROSPERO (CRD42024577356).

Results:

Of 1567 initially identified records, 7 studies involving 1161 participants were included. Most studies had a low risk of bias. The results indicated that varenicline (4 studies, n=463; OR=2.52, 95% CI: 1.14–5.58) was effective, while the effect of nicotine replacement therapy (NRT; 2 studies, n=538; OR=1.38, 95% CI: 0.61–3.14) was not statistically significant. One study on cytisine also suggested potential benefit.

Conclusions:

Varenicline has demonstrated efficacy in supporting e-cigarette cessation, whereas the current evidence for NRT and cytisine remains insufficient to draw firm conclusions. Further well-designed studies are needed to clarify the effectiveness of these 2 interventions. Although pharmacotherapies used for cigarette smoking show potential for repurposing, their utility in e-cigarette cessation should be interpreted cautiously until stronger evidence becomes available.

E-cigarettes are battery-powered devices that heat liquid containing nicotine, propylene glycol, and/or glycerin and flavoring.¹ In China, the prevalence of e-cigarette use among adults is rising, increasing from 1.3% to 1.6% during 2015–2019.² Similarly, in the United States, e-cigarettes are also becoming more popular, especially among young adults. The prevalence of e-cigarette use reached 4.5% among adults and 11.0% of those aged 18–24 years in 2021.^3,4 Though developed as an alternative to cigarettes, the emissions from e-cigarettes contain toxic substances that can also be harmful to people who use e-cigarettes.⁵ Nicotine, the primary psychoactive component in most e-cigarettes, exhibits well-documented addictive properties through its action on nicotinic acetylcholine receptors. Acute nicotine exposure can induce cardiovascular effects, including elevated blood pressure, tachycardia, and increased catecholamine release, while chronic exposure may contribute to endothelial dysfunction and oxidative stress, potentially increasing cardiovascular disease risk.^6–8 Beyond nicotine-related effects, e-cigarette aerosols contain potentially harmful constituents that have been associated with respiratory injury, particularly in cases of e-cigarette or vaping product use–associated lung injury (EVALI). In addition, nicotine exposure during adolescence and young adulthood may adversely affect neurodevelopment, as brain maturation continues into the mid-twenties.⁹

Given the health hazards and not-well-established long-term risks of e-cigarette use, summarized data on e-cigarette cessation methods are necessary. While some people succeeded in quitting e-cigarettes on their own, many needed assistance.¹⁰ Despite the World Health Organization’s (WHO) 2024 clinical treatment guidelines for smoking cessation, no evidence-based recommendations were included for those wishing to quit e-cigarettes. The guideline recommended 5 pharmacological interventions for quitting smoking: nicotine replacement therapy (NRT), bupropion, varenicline, cytisine, and combination pharmacotherapy.¹¹ However, it remains unclear whether smoking cessation methods can be applied to people who use e-cigarettes. Furthermore, the effectiveness and safety of e-cigarette cessation have not been systematically studied.¹²

There is a lack of summaries of pharmacological cessation interventions for e-cigarettes. Although previous studies have investigated effective interventions for e-cigarette cessation, due to factors such as small sample sizes, diverse intervention designs, limited number and low quality of studies on identical interventions, and varying durations of experimental research, no evidence-based effective e-cigarette cessation interventions have been established.^13,14 A meta-analysis is ongoing to summarize the nonpharmacological interventions for e-cigarette cessation among adolescents.¹⁵ In addition to young people, adults may use e-cigarettes as an alternative to traditional cigarettes, primarily to aid smoking cessation or to maintain nicotine intake in smoke-free environments, and may continue their use over an extended period.^16,17 Thus, this systematic review and meta-analysis aimed to accumulate evidence on pharmacological interventions for e-cigarette cessation (ie, NRT, varenicline, cytisine, and bupropion in any form). It comprehensively examined the efficacy and adverse effects in the design of randomized control trials to support those attempting to quit e-cigarettes.

METHODS

The study protocol was registered in PROSPERO (CRD42024577356). The search and report followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) (Supplementary Table S1, Supplemental Digital Content 1, http://links.lww.com/JAM/A744).

Search Strategy

We searched the following platforms from inception to May 1, 2024, without language limits: Cochrane Library, PubMed, Ovid Embase, Sinomed, and Clarivate Web of Science Core Collection. We used PICOS principles for the literature search. Our search terms included published search filters for “Pharmacological or drug or nicotine replacement therapy or varenicline or bupropion or cytisine,” “electronic nicotine delivery systems (ENDS) or electronic cigarette or e-cigarette or e-cigar,” “quit or cessation or reduce or prevention,” and “RCT or clinical trial” (details are presented in Supplementary Table S2, Supplemental Digital Content 1, http://links.lww.com/JAM/A744).

Study Selection

One investigator (AA) conducted the search and data extraction, which was reviewed by another investigator (YX). Differences were resolved by negotiation or asking a third researcher (LZ). We screened references in 2 stages. First, we screened titles and abstracts for studies that may be eligible. Second, 2 authors independently screened each full text for eligibility studies.

Inclusion Criteria

We considered studies eligible if (1) including adults who were aged 18 years or older using electronic cigarettes; (2) using any pharmacological intervention approved for cigarette cessation (ie, any forms of NRT, varenicline, cytisine, bupropion, combination use of more than one of these interventions or pharmacological interventions combined with any type of cognitive behavior therapy); (3) with outcome indicators measured before and after the pharmacological intervention; and (4) in a randomized control trial (RCT) design. Meanwhile, we excluded studies (1) including young people under the age of 18 years; (2) in nonrandomized control designs; (3) targeting animals rather than humans; (4) without control groups; (5) the data were not complete or not extractable; and (6) excluded studies that focused solely on behavioral interventions without involving pharmacological treatment.

Quality Assessment

We assessed risk of bias for each included study using the Cochrane RoB 2 tool¹⁸ and assessed the following domains: (1) random sequence generation (selection bias); (2) allocation concealment (selection bias); (3) blinding of participants and personnel (performance bias); (4) blinding of outcome assessment (detection bias); (5) incomplete outcome data (attrition bias); (6) selective reporting (reporting bias); and (7) other bias. We assessed each domain as being at low, unclear, or high risk of bias, according to the guidance of RoB 2 in the Cochrane Handbook for Systematic Reviews of Interventions.¹⁹

Data Extraction and Synthesis

We extracted the following data from each qualified study using an extraction form: (1) Study information: author’s name, title, publication year, country, study design, setting, duration, sample size, number of experimental conditions, and number of participants. (2) Participant characteristics: age, sex, race, and baseline e-cigarette use status. (3) Intervention details: pharmacotherapy type, dose, course of use, administration, and initiation of use (ie, before or on quit day). (4) If combined with other support: type (no support, self-help, or interactive behavioral support). (5) Outcomes: definition of abstinence, number of abstinent participants per arm, follow-up time, and number of participants abstinent per arm. (6) Harm and safety outcomes, including adverse events, and tolerability. (7) Risk of bias: information related to any of the risks of bias domains, information related to any other potential biases identified. Two investigators (AF and YX) extracted the relevant data independently and discussed the discrepancies or consulted with a third person (Table 1).

TABLE 1.

Baseline Characteristics and Outcomes of Studies

Author, Reference	Settings	Sample Size	Age, Mean (SD)	Sex	Race	Study Design	Interventions	Control	Medication Course Dosage	e-Cigarette Cessation Outcome
Fucito et al.26	Yale University Medical School and the Medical University of South Carolina	44	28.2 (8.5)	Male: 20; Female: 20	White (32); non-Hispanic	RCT	Varenicline and self-guided cessation booklet	Placebo	Varenicline (daily titration to 2 mg)	Varenicline vs placebo At week 8 quit rates (9/20; 45% vs 6/20; 30%; RR=1.51; CI, 0.68−3.37) At week 12 follow-up (8/20; 40% vs 6/20; 30%; RR=1.36; CI, 0.59−3.13).
Pasquale Caponnetto et al.23	Centro per la Prevenzione e Cura del Tabagismo (CPCT), the University-run smoking cessation center in Italy.	155	52.8 (9.4)	Male: 96; Female: 59	NA	RCT	Varenicline and smoking cessation counseling	Placebo	Participants in the varenicline group received varenicline (1 mg, administered twice daily for 12 wk) plus counseling. Participants in the placebo group received placebo (administered twice daily, for 12 wk) plus counseling.	CAR between weeks 4 and 12: 50.0% vs 16.9% (OR=4.9; 95% CI, 2.3–10.4; P<0.0001); CAR between weeks 4 and 24: 48.7% vs 14.3% (OR=5.7; 95% CI, 2.6–12.3; P<0.0001). 7-d point prevalence: Week 12: 51.2% vs 22.0% (OR=3.7; 95% CI, 1.9–7.5; P [0.000816] > 0.0001) Week 24: 48.7% vs 18.1% (OR=4.3; 95% CI, 2.1–8.9; P [0.00034068] >0.0001)
Pasquale Caponnetto-2 et al.10	University-run smoking cessation center in Italy	140	52.6 (9.1)	Male: 69; Female: 71	NA	RCT	Varenicline and smoking cessation counseling	Placebo	Varenicline was titrated to full dose by the time of their TQD (0.5 mg/d for 2–3 d, 0.5 mg twice daily for 4–5 d; then 1 mg twice daily for 11 wk)	CAR varenicline vs placebo: Weeks 4–12: 40.0% vs 20.0% (OR=2.67, 95% CI, 1.25–5.68, P=0.011); Weeks 4–24: 34.3% vs 17.2% (OR=2.52, 95% CI, 1.14–5.58, P=0.0224). 7-d point prevalence of e-cigarette abstinence varenicline vs placebo: At week 12: 40.0% vs 20.0% (OR=2.67, 95% CI, 1.25–5.68, P=0.011); At week 24: 34.3 vs 17.2 (OR=2.52, 95% CI, 1.14–5.58, P= 0.0224).
Rigotti et al.22	5US sites	160	33.6（11.1）	Female: 83 [51.9%]	6 (3.8%) were Asian, 14 (8.8%) were Black or African American, 135 (84.4%) were White, 5 (3.1%) were another race, and 9 (5.6%) were of Hispanic ethnicity	RCT	Cytisine and behavioral support	Placebo	3 mg of cytisine or placebo taken orally 3 times daily for 12 wk	Continuous e-cigarette abstinence rate in cytisine vs placebo groups: At weeks 9–12 (end of treatment): 31.8% vs 15.1% (OR=2.64; 95% CI, 1.06–7.10; P=0.04) At weeks 9–12: 23.4% vs 13.2% During weeks 9–16 (OR=2.00; 95% CI, 0.82–5.32; P=0.15).
Palmer et al.24	Online advertisements (Craigslist) across South Carolina	30	32	Male: 50%	33.3% identified as a racial or ethnic minority	RCT	NRT and behavior support: supportive booklet or Quitline referral	Behavior support	28-d supply of NRT (21 mg patches, 4 mg lozenges)	At the end of treatment (EOT; day 28) and follow-up (FU; day 56) in the intervention group, 6 (33.3%) reported abstinence from e-cigarette use at EOT, whereas 0 participants in the control group endorsed abstinence. At FU, 5 participants (27.7%; 4/6 that were abstinent at EOT) in the intervention group remained abstinent from e-cigarette use, whereas 2 (16.66%) in the control group endorsed quitting e-cigarettes. Within those who continued to vape at EOT and FU, 5 participants in the treatment group (5/12; 41.67%) and 4 participants in the control group (33.33%) endorsed reducing e-cigarette use.
Klein25	The Ohio State University	504	NA	Female:362; Male: 126; Other: 20	Caucasian: 350; African: 43; Hispanic: 48; Other: 66	RCT	NRT and behavioral: phone counseling or behavioral: digital coaching	Behavior support	NRT treatment will consist of up to an 8-week supply of nicotine patch, gum, and/or lozenge. NRT will be sent in two 4 wk shipments. Participants may be dosed for a single form of NRT or combination NRT (patch plus gum or lozenge) based on coach assessment of nicotine use and participant preference.	7-day point prevalence e-cigarette Abstinence no NRT, no digital (55/134 41%); digital, no NRT (54/126 42.9%); NRT, no digital (61/126 48.4%); or NRT and digital (59/122 48.4%). e-cigarette abstinence: no NRT, no digital (46/134 34.3%); digital, no NRT (45/126 35.7%); NRT, no digital (52/126 41.3%); or NRT and digital (53/122 43.4%).
Evins et al.27	Center for Addiction Medicine, Boston, MA 02114	261	Mean age: 21.4 y	Female: 53%	Asian: 48; Black: 16; Multiracial: 28; White: 158; Other: 11; Hispanic or Latino(a) ethnicity: 43	RCT	Varenicline plus behavioral counseling	Placebo plus behavioral counseling, or enhanced usual care	Participants were randomized (1:1:1) to 12 wk of varenicline titrated to 1 mg twice daily over 7 d (standard titration); Both varenicline and placebo were administered for a treatment period of 12 wk. The study included a follow-up period of up to 24 wk. Assessments were conducted at weeks 16, 20, and 24 to observe the participants’ continuous abstinence from vaping	For varenicline and placebo, continuous abstinence rates were 51% vs 14% during weeks 9 through 12 (adjusted odds ratio [aOR], 6.5 [95% CI, 3.0–14.1]; P<0.001) and 28% vs 7% during weeks 9 through 24 (aOR, 6.0 [95% CI, 2.1–16.9]; P<0.001). Varenicline had higher continuous abstinence rates vs enhanced usual care during weeks 9 through 12 (51% vs 6%; aOR, 16.9 [95% CI, 6.2–46.3]) and during weeks 9 through 24 (28% vs 4%; aOR, 11.0 [95% CI, 3.1–38.8]). Continuous abstinence rates were not significantly different between the placebo and enhanced usual care groups

Statistical Analysis

Since the outcomes were dichotomous indicators, pooled odds ratios (ORs) with their 95% CIs were calculated using the inverse variance method with Hartung-Knapp adjustment under the random effects model. Heterogeneity was evaluated using the Q statistic and I ² test. If P≤0.10, the studies were considered with heterogeneity. Those with heterogeneity were further assessed using I ² values. Heterogeneity was classified as negligible if I ² ≤ 50%, moderate if 50%<I ²≤70%, and high if I ²>70%. Inter-trial statistical heterogeneity was evaluated using the I ² test.^20,21 Considering the expected limited number and the expected heterogeneity of eligible studies, a random effects model was conducted using the inverse variance method with Hartung-Knapp adjustment. Sensitivity analysis was used to check the stability of meta-analysis results. We used Review Manager 5.3 software and R4.3.2 for bias evaluation, heterogeneity testing, data synthesizing, and generating bias plots. A 2-tailed P<0.05 was considered significant.

RESULTS

Search Results

We initially obtained 1567 relevant studies. A total of 578 duplicate papers and 964 irrelevant studies were excluded, and 25 studies were retained. After thoroughly reviewing the full texts, a total of 18 studies were excluded, leaving 7 studies included in the meta-analysis.^10,22–27The detailed screening process is shown in the flow diagram (Fig. 1).

FIGURE 1.

Flow chart of the literature selection process. RCT indicates randomized controlled trial.

Characteristics of the Included Studies

This meta-analysis included 7 studies with a total of 1161 participants (603 in the intervention group and 558 in the control group). Both males and females over the age of 18 years were included. Among them, 4 studies used varenicline;^10,23,26,27 2 studies used NRT,^24,25 and 1 study used cytisine.²² The duration of each drug intervention session was at least 4 weeks. These studies were conducted in the United States and Italy. All studies included behavioral support; 5 studies^10,22,23,26 used placebo treatment in the control group, and 2 studies^24,25 used only behavioral treatment without placebo in the control group (Table 1).

Bias Analysis of the Included Studies

In this study, the specific quality evaluation indicators included 7 aspects: random sequence generation, allocation concealment, blinding of participants, blinding of outcome assessment, incomplete outcome data, selective reporting of experimental results, and other biases. Regarding random sequence generation, 2 studies had unclear risks of bias, and others had low risks. Regarding allocation concealment, 2 studies had unclear risks of bias, and the rest had low risks. Regarding the blinding of participants and personnel, 5 studies had low risks of bias, 2 studies had unclear risks of bias. As for the blinding of the outcome assessment, 3 studies had unclear risks of bias, and the rest had low risks. Regarding incomplete outcome data, 1 study had an unclear risk of bias, and the rest had a low risk. Regarding the selective reporting of experimental results, one study had unclear risk of bias, and the rest had low risks. According to our comprehensive evaluation, the overall risk of bias in this meta-analysis was low (Fig. 2).

FIGURE 2.

A, Risk of bias of the included studies. B, Summarized risk of bias of the included studies. Notes: “?,” “+,” “–,” signify unclear risk of bias, high risk of bias, and low risk of bias, respectively.

Results of the Meta-analysis

There were 4 studies using varenicline (n=463; random effects model: OR=2.52, 95% CI, 1.14–5.58, P<0.001, Fig. 3A), and 2 studies used NRT (n=538; random effects model: OR=1.38, 95% CI, 0.61–3.14, P=0.07; Fig. 3B). Of the 7 included studies, 5 reported continuous abstinence rates of at least 4 weeks. A meta-analysis of these 5 studies (n=1091) demonstrated a significant treatment effect (random effects model: OR=2.09, 95% CI, 1.60–2.75, P<0.001; Fig. 4A). Considering that one study accounted for ~47% of the total sample, we performed a sensitivity analysis by excluding it to assess potential bias. The results remained consistent, showing an increased yet stable effect size (OR=3.76, 95% CI, 2.43–5.83, P<0.001; Fig. 4B).

FIGURE 3.

Odds ratios for NRT and varenicline on electronic cigarette cessation. A, e-Cigarette cessation rate of the study use varenicline; B, e-cigarette cessation rate of the study use NRT.

FIGURE 4.

Odds ratios for pharmacological interventions on electronic cigarette cessation. A, CAR at 12 weeks; B, CAR at 12 weeks after excluding the study with the largest number of participants; C, the 7-day point prevalence abstinence rates at 12 weeks; D, the 7-day point prevalence abstinence rates at 12 weeks after excluding the study with the largest number of participants.

In addition, 4 studies reported the 7-day point prevalence of abstinence at 12 weeks. A meta-analysis of these data (n=836) also revealed a statistically significant benefit (OR=2.17, 95% CI, 1.65–2.87, P<0.001; Fig. 4C). Similarly, due to 1 study contributing about 61% of the total sample, a sensitivity analysis was conducted. Exclusion of this study yielded a higher but still stable estimate of effect (OR=3.98, 95% CI, 2.50–6.36, P<0.001; Fig. 4D).

Furthermore, a trend toward a diminishing treatment effect over time was observed. The 7-day point prevalence abstinence rate was nominally higher at 8 weeks (OR=2.37, 95% CI, 1.43–3.92, P<0.001; Supplementary Figure S1, Supplemental Digital Content 1, http://links.lww.com/JAM/A744) than at 12 weeks (OR=2.17, 95% CI, 1.65–2.87, P<0.001; Supplementary Figure S1, Supplemental Digital Content 1, http://links.lww.com/JAM/A744).

Adverse Events

No serious adverse events were reported in the included studies. Mild adverse events occurred more frequently in the drug intervention group than in the placebo group. The most common mild adverse events were nausea, insomnia, dreaminess, nightmares, headache, dry mouth, and anxiety. Participants in both the pharmacological intervention and control groups withdrew due to adverse effects; however, only 3 studies fully reported such withdrawals, so no definitive conclusions can be drawn. The data on tolerability outcomes, specifically withdrawal due to treatment, were relatively limited. Three studies reported the number of participants who withdrew from the study due to adverse drug reactions. A total of 13 participants stopped e-cigarette cessation due to adverse drug reactions, 4 for varenicline, and 4 for cytisine, while 5 participants discontinued treatment in the placebo group.

DISCUSSION

To our knowledge, this is the first systematic review and meta-analysis to analyze the effect of pharmacological intervention on e-cigarette cessation, following methods to identify eligible studies. Our comprehensive search included proactive identification of unpublished data through trial registers and conference abstracts. Data from 7 RCTs involving 1161 participants indicate that varenicline has demonstrated efficacy in helping adults quit e-cigarettes. However, the evidence for NRT and cytisine remains insufficient, and further high-quality studies are needed to determine their effectiveness. These findings highlight the emerging but still limited evidence base for developing pharmacological guidelines for e-cigarette cessation. The identified studies included 3 main medications for e-cigarette cessation: varenicline, nicotine replacement therapy, and cytisine. Previous studies have reported the potential mechanism of these medications for smoking cessation. NRT is a therapy that uses low-dose nicotine absorbed through the oral mucosa, nasal mucosa, or skin²⁸ to alleviate withdrawal symptoms.²⁹ Varenicline is a nicotine receptor partial agonist,³⁰ which helps activate the nicotinic receptors in the brain. It is usually activated by nicotine, releases dopamine, and prevents nicotine from further activating these receptors.²⁹ Cytisine, a partial agonist at α4β2 nicotinic acetylcholine receptors that mediates nicotine dependence, has shown efficacy in smoking cessation.²² Previous studies have investigated the effect of varenicline on e-cigarette cessation due to its good performance in quitting cigarettes. In addition, the use of NRT can also be effective both for e-cigarettes or dual-use cessation. While both e-cigarettes and cigarettes contain nicotine and may lead to dependence, the smoking and quitting behaviors may differ because of various doses and modes of nicotine intake. Therefore, future interventions targeting e-cigarette cessation need to clarify the dose and behavior modes of e-cigarette use.

This review provides robust evidence to guide clinicians in supporting patients who wish to cease e-cigarette use. Our meta-analysis, which synthesized data from 1161 participants across 7 studies, shows that the overall pharmacotherapy effect was largely driven by varenicline. When examined separately, varenicline demonstrated a significant benefit for vaping cessation, whereas NRT did not show statistically significant efficacy (OR=1.38, 95% CI, 0.61–3.14; P=0.07). Given these findings, only varenicline can currently be considered an effective pharmacological aid for e-cigarette cessation. Evidence for NRT and cytisine remains limited and inconclusive, underscoring the need for further high-quality research. At the end of the treatment period, the pooled abstinence rate was ~33% (197/603) in the pharmacotherapy groups, compared with a notably lower rate in control groups. However, consistent with the chronic and relapsing nature of nicotine dependence, the treatment effect showed a declining trend over time—the abstinence rate was nominally higher at 8 weeks (OR=2.37) than at 12 weeks (OR=2.17). This pattern underscores the need for ongoing support to prevent relapse. Potential reasons for this decline, as suggested by previous studies, include persistent nicotine dependence, environmental cues, and insufficient long-term support. A Cochrane review has indicated that extending the duration of pharmacotherapy—particularly with varenicline—is a promising strategy to mitigate relapse.³¹ Our findings further highlight the importance of continuous monitoring of abstinence status and the potential value of adjunctive interventions, such as digital³²or social support,^33–35 in maintaining long-term cessation.

In comparing the efficacy of different pharmacotherapies, varenicline demonstrated a stronger effect size (OR=2.52, 95% CI, 1.14–5.58, P<0.001) than NRT (OR=1.38, 95% CI, 0.61–3.14, P=0.07), though this comparison was based on a limited number of studies. This finding is consistent with evidence from traditional smoking cessation, where varenicline and cytisine have shown superior efficacy compared with NRT, bupropion, and nortriptyline.^36,37

Although varenicline has demonstrated efficacy for e-cigarette cessation, the current evidence for NRT remains limited and inconclusive. Clinicians should pay attention to the potential differences between the 2 groups. Traditional cigarette smoking is often characterized by highly ritualized patterns (eg, after meals, during breaks), whereas e-cigarette use can be more situational and ad libitum due to greater convenience and fewer restrictions. This pervasive use pattern may lead to more frequent but lower-intensity nicotine dosing, potentially establishing different behavioral cues for dependence. Consequently, cessation interventions for e-cigarettes might need to place greater emphasis on behavioral substitution strategies and addressing nontriggered, habitual use.³⁸

Clinicians can convey these findings to patients by explaining that prescription medications like varenicline and over-the-counter NRT can increase their chances of successfully quitting e-cigarettes. The shared decision-making process should include a discussion of the benefits and potential side effects of each option. For patients unwilling to use pharmacological options, behavioral support (eg, counseling, quit line services) alone has demonstrated some efficacy. Combining pharmacotherapy with behavioral support is likely to yield the best outcomes, as is standard in tobacco cessation treatment.

The findings of this review, drawing on evidence from countries such as the United States and Italy, carry important implications for global tobacco control policy and regulatory bodies. Currently, clinical guidelines in many nations are primarily based on evidence for combustible cigarette cessation and have not formally incorporated pharmacotherapies like varenicline for e-cigarette cessation. Our analysis provides a preliminary evidence base to inform the updating of these guidelines, supporting the integration of effective pharmacotherapy into the management of e-cigarette dependence. For drug regulatory agencies (eg, FDA, EMA), these findings highlight the need to consider approving label extensions for medications such as varenicline specifically for e-cigarette cessation, which would provide clinicians with clear regulatory backing for their prescriptions.

Several limitations should be noted when interpreting our results. First, most studies have been conducted in higher-income countries, leaving a gap in evidence from lower-income regions. In South Africa, 2.71% of adults, translating to 1.09 million people, used e-cigarettes daily or occasionally during 2018,³⁹ where interventions for e-cigarette cessation are also necessary. Future studies including data from developing countries would be valuable. Second, the number of studies on e-cigarette cessation was limited and primarily conducted in the United States (n=5) and Italy (n=2); furthermore, only 7 studies met the eligibility criteria, which limited our ability to conduct subgroup analyses by behavioral support, dosage, or participant characteristics. These are based on very limited evidence and may not justify making broad clinical conclusions. Current e-cigarette cessation interventions using medications are primarily aimed at adults, and WHO guidelines do not recommend that adolescents use these medications for quitting. Therefore, attention should be paid to the population for future research and health education. More high-quality research is needed to generate evidence for future e-cigarette cessation guidelines. Third, current studies have primarily focused on the efficacy and safety of a single drug in different settings. Future research could compare the cost-benefits of other drugs, which may provide evidence for choosing the most appropriate drug for individuals. Fourth, in this review, a judgment of “unclear” risk of bias indicates that the original study report provided insufficient or contradictory information, making a reliable judgment of “low” or “high” risk difficult. It is important to note that this signifies uncertainty in assessment rather than implying poor study quality. Lastly, a key limitation of the included studies is the variability in participant characteristics and cessation protocols. While all studies recruited current e-cigarette users, participants differed in nicotine dependence status and patterns of dual use with combustible cigarettes. Cessation approaches also varied, with some studies using predefined quit dates and others encouraging immediate or gradual reduction, and the specific protocols were not consistently reported. This heterogeneity in both participant profiles and intervention implementation may limit the generalizability of our findings and should be considered when interpreting the results.

Two of the included studies^24,25 used behavioral support alone as the comparator arm, 1 study used behavioral treatment plus placebo in the control group, while the remaining trials compared medication to placebo in the context of behavioral counseling. Behavioral support—including brief advice, counseling, and quit-line or digital support—is a well-established component of tobacco cessation programs and is often combined with pharmacotherapy to improve quit rates. Although our review focused on pharmacologic interventions, the presence of behavioral support across trials is clinically important: behavioral interventions may contribute substantially to cessation outcomes and can interact synergistically with medications. Because the 2 behavioral-only trials in this review were small and heterogeneous in design, and because cessation protocols (eg, intensity of counseling, quit-date instruction) were inconsistently reported across studies, we cannot isolate the independent effect of behavioral support from that of medications in our pooled analyses. We therefore highlight the need for future trials that clearly report behavioral intervention content and fidelity, and for studies that are powered to compare medication alone, behavioral support alone, and combined approaches in e-cigarette cessation.

In conclusion, this systematic review and meta-analysis found that varenicline has demonstrated efficacy in helping individuals quit e-cigarettes. However, the current evidence for NRT and cytisine remains limited and inconclusive, indicating that further high-quality research is needed to clarify their potential role in e-cigarette cessation. Future studies should also assess the cost-effectiveness of available interventions and support the development of individualized, evidence-based approaches for e-cigarette cessation.

Supplementary Material

adm-20-556-s001.docx ^{(264.6KB, docx)}