Electronic Cigarette Use Intentions Mediate the Association between Low Self-Control and Future Use by Internalizing Symptoms

Abstract

Background: Adolescent electronic (e-)cigarette use intentions are related to initiation. Low self-control is also a risk factor for early stages of substance use. Yet, the impact of low self-control on use through intentions may vary across individuals; depression and anxiety may affect this association. Methods: A sample of 200 adolescents who completed waves 1 and 2 of an ongoing longitudinal study were assessed. We hypothesized that high internalizing symptoms would moderate the indirect effect of low self-control on actual e-cigarette use through e-cigarette use intentions. Results: The mediation pathway was significant at high levels of internalizing symptoms, but not at low or moderate levels. Conclusion: Specifically, those with low self-control and high internalizing symptomatology endorsed the highest e-cigarette use intentions and were more likely to subsequently use e-cigarettes. Youth low in self-control and high in depression and anxiety might be at increased risk to initiate e-cigarette use compared to youth high in self-control and high in internalizing symptomatology.

Keywords:

• e-cigarettes
• adolescence
• depression
• anxiety
• low self-control

Acknowledgements

We thank family participation in the Antecedents and Consequences of Electronic Nicotine Delivery Systems (ACE) Project and Nasreen Hidmi, the ACE Project Program Coordinator.

Notes

1 The remaining 43 adolescents who were enrolled in W1, did not complete W2 due to difficulties with retention (i.e. scheduling conflicts, unreachable, not interested anymore, parent would not consent for child, adolescent was not interested, no time to complete procedures, unfortunate personal events).

2 A post-hoc model was estimated in Mplus using full information maximum likelihood (FIML) as an effort to determine whether missingness patterns found for self-reported low self-control impacted study findings. Findings for the full sample (N = 264) were consistent with the original models. Thus, missingness seemed to have a negligible effect on the results.

3 The larger project employed a 15-month follow-up to balance obtaining comprehensive information from adolescents and caregivers while attempting to minimize participant burden as a large portion of the original sample was also invited for an additional research visit to complete a magnetic resonance imaging scan (although this data is not the focus of the current study). While year-long retrospective reporting periods of substance use have been consistently employed by national studies (Johnston et al., 2022), prior work has suggested strong reliability among adolescents reporting substance use across a 2-year study period (Shillington & Clapp, 2000).

4 Given the high degree of overlap between ethnicity and race, only ethnicity was included in the final models. Yet, findings were similar when replacing ethnicity with race as a covariate.

Additional information

Funding

This work was supported by the National Institutes of Health under Grant U54MD012393 (Sub-Project:5378) PI: Eric Wagner; K08 AA023290 PI: Trucco. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.