More than half of U.S. smokers attempt to quit each year. More and more, these quit attempts have involved the use of electronic nicotine delivery systems (ENDS), such as e-cigarettes. Unlike pharmaceutical nicotine replacement therapy (NRT) products, ENDS are not approved by any medicines regulator around the world as smoking cessation devices.
Many types of studies have sought to examine how effective ENDS are at improving the odds of quitting smoking for good, including randomized controlled trials (RCTs). Because RCTs are considered the “gold standard” of evidence for assessing how well an intervention works, we will review the RCTs done to date to see what they can tell us about whether ENDS can help people quit smoking.
A Cochrane review from 2016 found relatively weak evidence that early-generation ENDS helped people quit smoking based on the two RCTs that had been conducted in Italy and New Zealand up to that point. Since then, three more major RCTs have been conducted, each with different strengths and weaknesses.
A study based in the UK by Hajek et al. from 2019 is arguably the most important single study to date due to its large sample size and participants’ adherence to the product(s) they were assigned. The ENDS group was provided with a current-generation (to that time) tank-style device, nicotine liquid, and instructions for use. The NRT group was provided with any requested combination of patches, gums, lozenges, nasal or mouth sprays, or inhalers. Both groups received behavioral support. At 52 weeks, 18% of the ENDS group had reported quitting smoking, compared to 9.9% of the NRT group. Notably, 80% of quitters in the ENDS group were still using ENDS at follow-up.
The other major trial that allows for direct comparison of an experimental group provided with ENDS and one that was not provided with ENDS (what researchers considered the “usual care”, which is typically NRT and counselling) was a US-based study by Halpern et al. (from 2018). Unfortunately, this trial had very low compliance across all groups, and so, although they found a higher rate of biochemically verified quits in the ENDS group than the “usual care” group, the difference was insubstantial. The authors reran their analysis on a non-random subset of the study population that logged onto the trial website at least once after randomization, finding that the difference in cessation rate was larger, but the statistical significance of this difference was reduced.
Walker et al. (2020) analyzed ENDS use in combination with the nicotine patch in a New-Zealand-based cohort, finding a biochemically verified cessation rate of 7% at six months in a group given nicotine patches and nicotine-containing ENDS devices, compared to 4% for a group given nicotine patches and ENDS that did not contain nicotine. A smaller third group that was given only patches had a cessation rate of 2%. Unlike the difference between the ENDS groups, the difference between this group and the others was not statistically significant due to the low number of participants in this group compared to the first two.
Several smaller RCTs have been conducted as well. In 2017, Carpenter et al. conducted a pilot study on 68 US smokers, with no significant difference in cessation between groups given e-cigarettes with different strengths of nicotine and a control group. Lee, Ahn, and Cheong (2019) reported no significant difference between a group provided with ENDS and one provided with nicotine gum in a South Korean trial with150 participants. Also in 2019, Hatsukami et al. analyzed a population of US smokers uninterested in immediately quitting smoking and found a significantly higher 7-day abstinence at eight weeks for a group of 76 smokers who were instructed to completely switch to e-cigarettes compared to another group of 76 smokers instructed to switch to NRT. Masiero et al. (2019) found higher cessation rates among groups of motivated Italian smokers given ENDS with and without nicotine compared to a control group that was not given ENDS.
Based on these studies, how much evidence is there to support claims that e-cigarettes help people quit smoking? First, it is worth noting that RCT results are always specific to the devices used — e-cigarettes are not standardized, and there are many differences between brands and models — and accompanying support (and potentially other context), and so their results may not generalize to other products and settings (the same can also be said about RCTs for other cessation aids and/or support). For example, newer generation e-cigarettes typically have liquid with a higher nicotine concentration and/or are more efficient at getting nicotine to the bloodstream of the user.
Second, one large study found that ENDS worked significantly better than NRT, and several other studies did not find a significant difference, with none finding ENDS performed significantly worse. Taken together, it is likely that ENDS are at least as effective as NRT at helping people quit smoking, but more studies are needed across a variety of other settings before this can be concluded with confidence.
Third, ENDS-associated quits are more likely to include sustained nicotine use at follow-up than NRT-associated quits. As a result, the overall advisability of using ENDS for cessation is heavily dependent on the direct harms associated with ENDS use, which have not been definitively established.
All this said, successful quit rates for both ENDS and NRT remain low — emphasizing the need for multiple quit attempts and ongoing support, as well as even more vigorous prevention strategies to stop people from taking up smoking in the first place.