When we did our meta-analysis of the association between e-cigarette use and actual disease, we excluded mental illness because of concern that the association went in both directions, i.e., people with mental illness might be more lively to use e-cigs to self medicate with nicotine and people who used e-cigarettes (which delivered) nicotine would be more likely to develop metal illness. We just wanted to keep the discussion simple by focusing the discussion on diseases where the biology just pointed in one direction — from e-cigarettes to disease.
Now Shiya Ji and colleagues have provided a meta-analysis on the association between e-cigarette use and depression — an important mental illness — that shows that the association goes in both directions, but the arrow between e-cigarette use and depression is bigger. Specifically, their paper Bidirectional Association Between E-Cigarette Use and Depression: A Meta-Analysis found that current ecig use was associated with depression (OR 1.96, 95% CI: 1.73–2.23) as well as that depression was associated with more ecig use (OR 1.36, 95% CI: 1.15–1.61).
(They found similar associations for ever ecig use and depression.)
Most of the studies in the meta-analyses are cross-sectional, meaning that reverse causation is always a concern, a point the authors acknowledge.
The fact that the association between ecig use and depression is about 1.4 times bigger than the association in the other direction suggests to me that, while some of the observed association between ecig use and depression is because some depressed people are using ecigs to self medicate with nicotine, some of the observed association between ecigs and depression is depression caused by ecig use by non-depressed people.
In their sensitivity analysis, they showed stronger associations between e-cigarette use and depression for dual users (people who use both e-cigarettes and cigarettes; than people who just use ecigs, adding another disease for which dual use is worse.
The authors provide a good summary of what might be happening in the underlying biology to explain the associations in both directions:
The self-medication hypothesis posits that individuals with mental health disorders may utilize e-cigarettes for symptom alleviation. Among adolescents, EC use may function as either an emotional coping strategy or recreational activity. Regardless of exposure route, nicotine conversely contributes to the development or exacerbation of depressive symptoms. Chronic exposure to nicotine may lower serotonin levels, potentially exacerbating depressive symptomatology. Studies also found that prolonged nicotine exposure might dysregulate cerebral dopamine pathways, heighten stress sensitivity and compromise cognitive-emotional protective mechanisms against depression. Adolescent nicotine exposure significantly harms neurodevelopment, potentially leading to cognitive deficits and psychiatric sequelae including depression, anxiety, substance use disorders, and elevated suicide risk. Apart from nicotine, e-cigarettes also contain various levels of trace metals. Particularly, lead and aluminum affect both the central and peripheral nervous systems, potentially playing a role in the connection between e-cigarette use and depression. A recent investigation disclosed that daily use of flavored nicotine e-cigarettes might trigger neuroinflammation in brain areas associated with depressive behaviors. Additionally, EC use adversely impacts sleep health. Proposed mechanisms include nicotine-mediated disruption of sleep regulating neurotransmitters and respiratory-related sleep disturbances. Poor sleep quality correlates strongly with increased EC dependence, as sleep deprivation impairs both emotion regulation capacity and impulse control. As impaired sleep is a well-established depression risk factor, this interaction potentiates EC’s pro-depressive effects. In summary, these findings indicate bidirectional EC-depression reinforcement, creating a self-perpetuating cycle via potential mediating/synergistic mechanisms. Although these mechanisms offer plausible EC-depression pathways, they remain hypothetical without direct empirical validation in the included studies. Future experimental, clinical and longitudinal studies should directly test these mechanistic hypotheses. [emphasis added; citations deleted]
The authors sensibly note that more work needs to be done to nail these issues down, but they do show that the epidemiological results are consistent with what we know about the biology.
My one concern about the paper is that the analysis does not explicitly control for smoking (although some of the component studies did). The comparison of sole ecig users to dual users partially gets at this point a little, but it would have been better to specifically control for smoking.
Here is the abstract:
Introduction. Previous studies have indicated the potential bidirectional association between the use of e-cigarettes (ECs) and depression. However, this hypothesis is yet to be thoroughly investigated. In this meta-analysis, the two-way relationship is examined.
Methods. As of November 2024, major electronic databases were searched. Two sets of pooled risk estimates were computed by employing random effects models: one for EC use predicting depression and the other for depression predicting EC use. The forms of depressive presentation involved in this study include major depressive disorder, dysthymia, and general depressive symptoms. The results were presented in terms of adjusted odds ratios (ORs) along with 95% confidence intervals (CIs).
Results. We identified a total of 1601 references and included 27 studies involving 1 671 773 participants. Both current EC use (OR 1.96, 95% CI: 1.73–2.23) and ever EC use (OR 1.88, 95% CI: 1.59–2.21) could notably forecast depression. Pooled results also indicated that depression might significantly predict current EC use (OR 1.36, 95% CI: 1.15–1.61) and ever EC use (OR 1.53, 95% CI: 1.25–1.88).
Conclusion. EC use might be a significant predictor for depression and depression might be a significant predictor for EC use, highlighting the need to strengthen mental health warnings on e-cigarettes and integrate EC use screening in depression assessments.
Implications. Considering the bidirectional link between EC use and depression, clinical screening, public health interventions and stricter marketing regulation are crucial to mitigate risks. Future research should focus on the mechanisms and causal pathways of e-cigarettes and depression.
The full citation is: Shiya Ji, Lu Chen, Susu Zhuang, Liwen Wei, Lili Gou, Cheng Shi, Hanrui Jia, Xupeng Chen, Bidirectional Association Between E-Cigarette Use and Depression: A Meta-Analysis, Nicotine & Tobacco Research, 2025; ntaf163, https://doi.org/10.1093/ntr/ntaf163. It is available here.
Additionally, the small sample size (n=4) in this subgroup analysis heightens susceptibility to random error. The self-medication hypothesis posits that individuals with mental health disorders may utilize e-cigarettes for symptom alleviation. (Among adolescents, EC use may function as either an emotional coping strategy or recreational activity. Regardless of exposure route, nicotine conversely contributes to the development or exacerbation of depressive symptoms. Chronic exposure to nicotine may lower serotonin levels, potentially exacerbating depressive symptomatology. Studies also found that prolonged nicotine exposure might dysregulate cerebral dopamine pathways, heighten stress sensitivity and compromise cognitive-emotional protective mechanisms against depression. Adolescent nicotine exposure significantly harms neurodevelopment, potentially leading to cognitive deficits and psychiatric sequelae including depression, anxiety, substance use disorders, and elevated suicide risk. Apart from nicotine, e-cigarettes also contain various levels of trace metals. Particularly, lead and aluminum affect both the central and peripheral nervous systems, potentially playing a role in the connection between e-cigarette use and depression. A recent investigation disclosed that daily use of flavored nicotine e-cigarettes might trigger neuroinflammation in brain areas associated with depressive behaviors.
Stanton Glantz blog 