Vaping can help people quit smoking whilst significantly decreasing their exposure to deadly toxins and improving their health outcomes (Lindson et al., 2025; Royal College of Physicians 2024). However, with a rise in teen vaping among those who have never regularly smoked, people are questioning whether vaping could lead to a rise in subsequent teen smoking. This idea is commonly referred to as “the gateway hypothesis”. The hypothesis originated almost 100 years ago and has been famously applied to cannabis use being a “gateway” to other, “harder”, drugs. In a recent umbrella review, Golder et al. (2025) explored the gateway hypothesis in relation to vaping and subsequent smoking and other risky behaviours—here, we unpack this.
The gateway hypothesis has been widely contested, for many reasons. The central argument against the gateway hypothesis is that we cannot truly, ethically, test whether youth use of one product (e.g., vapes) is associated with use of another (e.g., smoking) among young people. To do this, we would need a trial where young people are randomised to vape or not and are then followed up to see if they have started smoking. This is highly unlikely to pass through any research ethics committees.
The best available, ethical, methods we have to test the gateway hypothesis are population surveys. However, while population surveys might find that people who vape at age 12 are more likely to smoke at age 18, can this ever really provide sufficient evidence of a “gateway” effect? Young people being surveyed might have gone onto smoke anyway, regardless of whether they used vapes. There are also other factors like changes in the market, popular products, or the possibility that youth are simply getting older or “growing out” of vapes.
The contender to the gateway hypothesis is the common liability theory. The common liability theory proposes that people who vape and smoke have the same underlying traits, like genetics, positive attitudes/culture towards nicotine use, and being more likely to take risks. Indeed, evidence in the field of vaping and smoking suggests that there are strong genetic vulnerabilities to both behaviours (Khouja et al., 2021), and also that the association goes both ways (i.e., vaping being associated with subsequent smoking, and smoking being associated with subsequent vaping; East et al., 2018). With longitudinal surveys, it is virtually impossible to control for all the possible underlying traits (i.e., confounders) of smoking and vaping, so it is difficult to tell whether the associations are genuine or caused by these confounding factors. A meta-analysis of “gateway” studies in 2021 (Chan et al., 2021) addressed this through calculating something called an “E value”, which allows us to evaluate the sufficiency of adjustment for confounding. Unsurprisingly, this study found a relatively high E-value, indicating that estimates in vaping-smoking “gateway” studies are not robust against unmeasured confounding. Overall, this evidence points more towards a common liability for both behaviours rather than vaping being a “gateway” to smoking.
Can vaping really lead to smoking among teens — or can the ‘gateway’ effect be explained by common risk factors?
Methods
The umbrella review by Golder et al. (2025) synthesises evidence of 56 reviews that assessed the associations between youth vaping (exposure) and any physical or mental health harms (outcome; including uptake of smoking or other addictive behaviours). Studies assessing quitting smoking as an outcome or other positive benefits (e.g., reduced stress) were excluded—this is important for reasons we will mention later. The study critically appraised the included reviews using a validated risk-of-bias tool, scoring the reviews from critically low to high. A youth advisory panel provided input on the emerging results, implications, and dissemination, which is great to see and rare for systematic reviews.
Results
Most reviews assessed the associations between vaping and smoking uptake (n = 25), although many of the reviews reported on the same studies. Five reviews assessed the associations between vaping and subsequent cannabis, alcohol, and/or non-prescribed Ritalin/Adderal. Most of these reviews included longitudinal cohort studies, thus allowing for assessment of whether vaping is associated with subsequent smoking or use of other substances.
The meta-analyses included found that odds ratios for the association between vaping and smoking initiation ranged from 1.50 to 26.01, with most studies reporting that young people who vape are around three times more likely than those not using them to initiate smoking. For other substances, the odds for associations with vaping ranged from 2.13 to 6.67.
The other reviews focussed on associations between vaping and respiratory outcomes (e.g., asthma; 16 reviews), burns/injuries (7 reviews), mental health (9 reviews), and cardiovascular disorders (4 reviews). These found that vaping among youth was associated with poorer respiratory outcomes (e.g., asthma, wheezing, bronchitis) and poorer mental health. Most of the studies included in the reviews, however, were cross-sectional which is particularly important for the association between mental health and vaping because some studies have found that youth use nicotine to cope with mental health symptoms, rather than vaping causing poor mental health. Some reviews also found that vaping was associated with burns/injuries, although the majority of these were based on case reports and so might not be generalisable, and many electrical items can cause burns if used or manufactured inappropriately.
Importantly, the quality of the vast majority of the included reviews were all rated as critically low due to significant methodological limitations, including a lack of a-priori methods, not listing excluding studies, and inadequate reporting on risk of bias impacts.
Youth who vape appear more likely to smoke later, but poor study quality clouds the interpretation of the findings.
Conclusions
The conclusions of the Golder et al. (2025) umbrella review were that there were consistent associations between vaping and subsequent smoking, marijuana use, alcohol use, asthma, cough, injuries and mental health outcomes. They also claim that the “strong associations in prospective cohort studies are consistent with a causal relationship between vaping and subsequent smoking”.
The umbrella review indicated consistent associations between vaping and subsequent smoking, marijuana use, alcohol use, asthma, cough, injuries and mental health outcomes.
Strengths and limitations
As Golder et al. (2025) identify, key strengths of their review include being the most comprehensive review of vaping-associated risks to date and the inclusion of a public involvement group of young people (which is rare with reviews). They considered a vast number of reviews (56 reviews), with most being systematic reviews. Screening was undertaken by two reviewers, enhancing the reliability of the review.
However, there are several limitations of this work. As Golder et al. (2025) mention, besides the outcomes of smoking initiation, most of the associations assessed are cross-sectional and, for burns/injuries and cardiovascular outcomes, were mainly limited to case reports, which rank low on the “hierarchy of evidence”. They also highlight that quality of the included reviews was generally low; however, the authors do not account for this in their conclusions. Therefore, the findings from this review must be treated with caution.
More broadly, the authors do not take into account the possibility that their findings could be explained by a common liability. They mention this competing theory in their discussion but place more emphasis on the gateway hypothesis. One of the reviews (Chan et al.) that was included in the umbrella review found that estimates in vaping-smoking “gateway” studies are not robust against unmeasured confounding, as mentioned earlier in this blog; however, the nuance of this study is not described by Golder et al. This is a major limitation of the review.
Finally, we also have the issue of what we’re focussing our attention on: if researchers are looking for an association between vaping and smoking, that’s what they will find. A recent evidence and gap map by Conde et al. (2024) found that the vast majority of studies assessing vaping and smoking among youth focus on vaping and subsequent associations with smoking, with very few considering whether vaping could be a “gateway” away from smoking—that is, looking at whether youth who smoke are using vapes to get themselves off of smoking. It is well-established that vapes are much less harmful than smoking and can help adults to quit smoking. Therefore, rather than repeatedly synthesising evidence of poor quality “gateway” studies, researchers would be best to focus their efforts on under-researched areas. This could include research into whether vaping is a way out of smoking among young people, as well as specific policy options that balance the need to prevent people who have never smoked from vaping with ensuring they are available to support adults who smoke to quit.
The review brings together an impressive body of research, but poor quality studies and missing nuance prevents any firm conclusions on the association between vaping and subsequent health outcomes.
Implications for practice
It important to remember that vaping is an important gateway out of smoking for many, and the arguments about youth vaping should not cloud this focus. Tobacco smoking is the leading cause of preventable death and disability globally, killing up to two thirds of people who smoke in the long term (World Health Organization, 2025). Vaping is substantially less harmful than smoking and there is high-certainty evidence that vaping can help people quit smoking (Lindson et al., 2025; Royal College of Physicians, 2024). NICE clinical guidelines recommend e-cigarettes to help people quit smoking, encouraging a full switch to reduce the harms from tobacco use. It is estimated that around 2.7 million people in Great Britain have quit smoking with a vape in the last five years (Action on Smoking and Health, 2024).
Although the support for the gateway theory of vaping is questionable, we must still acknowledge that vaping carries risks and young people and those who have never smoked should not vape (Royal College of Physicians, 2024). It is therefore important that guidelines and policy approaches balance the need to ensure that vaping does not appeal to young people with ensuring that vapes are available to adults for the purposes of quitting smoking. This could include policies to make vapes less appealing to youth and young adults, such as standardising packaging and point-of-sale restrictions combined with initiatives to make them accessible to adults who smoke, such as Swap to Stop, a government scheme that provides free vape kits and support to adults who smoke in England.
Vaping can help people who smoke to quit, but protecting young people without denying adults this tool remains a delicate balance.
Statement of interests
The authors of this blog had no personal involvement in the study. KE led the first study in Great Britain to assess the longitudinal associations between vaping and smoking among young people and this study was included in many of the reviews in the Golder et al. umbrella review. Both KE and ET have worked in the field of tobacco and nicotine research for many years, completing their PhDs on vaping and smoking norms (KE) and toxicants from vaping (ET). KE and ET are employed by the Brighton and Sussex Medical School and University College London, respectively. They have both co-authored reports on health risks and perceptions of vaping for the Office for Health Improvement and Disparities (OHID) and Royal College of Physicians.
Links
Primary paper
Golder S, Hartwell G, Barnett LM, Nash SG, Petticrew M, Glover RE. Vaping and harm in young people: umbrella review. Tobacco Control. 2025.
Other references
Action on Smoking and Health (2024). Nearly 3 million people in Britain have quit smoking with a vape in the last 5 years.
Chan GC, Stjepanović D, Lim C, Sun T, Shanmuga Anandan A, Connor JP, Gartner C, Hall WD, Leung J (2021). Gateway or common liability? A systematic review and meta‐analysis of studies of adolescent e‐cigarette use and future smoking initiation. Addiction. 116(4):743-56.
Conde M, Tudor K, Begh R, Nolan R, Zhu S, Kale D, Jackson S, Livingstone‐Banks J, Lindson N, Notley C, Hastings J (2024). Electronic cigarettes and subsequent use of cigarettes in young people: An evidence and gap map. Addiction. 119(10):1698-708.
East K, Hitchman SC, Bakolis I, Williams S, Cheeseman H, Arnott D, McNeill A (2018). The association between smoking and electronic cigarette use in a cohort of young people. Journal of Adolescent Health. 62(5):539-47.
Khouja JN, Wootton RE, Taylor AE, Davey Smith G, Munafò MR (2021). Association of genetic liability to smoking initiation with e-cigarette use in young adults: a cohort study. PLoS medicine.18(3):e1003555.
Lindson N, Butler AR, McRobbie H, Bullen C, Hajek P, Wu AD, Begh R, Theodoulou A, Notley C, Rigotti NA, Turner T, Livingstone-Banks J, Morris T, Hartmann-Boyce J (2025). Electronic cigarettes for smoking cessation. Cochrane Database of Systematic Reviews, Issue 1. Art. No.: CD010216. DOI: 10.1002/14651858.CD010216.pub9.
National Centre for Smoking Cessation and Training (2025). Supporting clients who want to stop vaping.
Royal College of Physicians (2025). E-cigarettes and harm reduction: An evidence review.
World Health Organization (2025). Tobacco Fact Sheet.
