Author ORCID Identifier

Defense Date


Document Type


Degree Name

Doctor of Philosophy



First Advisor

Dr. Caroline Cobb

Second Advisor

Dr. Andrew Barnes

Third Advisor

Dr. Rashelle Hayes

Fourth Advisor

Dr. Thomas Eissenberg

Fifth Advisor

Dr. Paul Perrin


Introduction: Smokers with mental illness (MI) are disproportionately affected by negative health outcomes. Electronic nicotine delivery systems (ENDS) may represent a harm reduction tool for those who reduce and/or replace their cigarettes with ENDS. Little previous research has examined how smokers with MI respond to ENDS. This analysis aimed to address this research gap using secondary data from a randomized controlled trial of ENDS varying in nicotine delivery among smokers with and without current MI. The aims were to test 1) the effects of MI status, condition, and time on changes in smoking behavior and negative mood and stress measures, 2) whether changes in negative mood and stress mediate condition-related effects on smoking, and 3) whether this mediation was moderated by MI status.

Methods: Smokers (n=520) interested in reduction but not cessation were randomized to receive either a non-nicotine-containing plastic cigarette substitute (CIG SUB) or ENDS differing in liquid nicotine concentration (0, 8, or 36 mg/ml) for 24 weeks. MI status was assessed at baseline. Smoking behavior (cigarettes per day; CPD) and negative mood (depression, psychological distress) and perceived stress measures were assessed at week 0, 4, 8, 16, and 24. Conditions were collapsed by nicotine-containing status (CIG SUB/0 mg/ml vs. 8/36 mg/ml), and participants were categorized by MI status (yes, no). Linear mixed models and mediation models were used. Sensitivity analyses included covariate adjustment.

Results: CPD reduction was significantly greater among smokers without MI at week 16 and 24 for the unadjusted analysis only. Nicotine conditions were associated with significantly greater CPD reduction at all time points, and both condition groupings resulted in significant CPD reduction relative to baseline. Significantly greater depressive symptoms were observed for non-nicotine conditions at week 4; significantly greater psychological distress was observed for non-nicotine conditions at week 24. With covariate adjustment, negative mood measures were significantly higher at later study time points for those with MI. Perceived stress differed by MI status but not condition grouping. Changes in negative mood and stress did not mediate CPD reduction, but direct effects of condition as well as changes in negative mood on CPD reduction were observed.

Conclusions: Our findings indicate that smokers with MI may experience greater difficulty reducing CPD, but nicotine conditions had similar effectiveness in reducing CPD relative to non-nicotine conditions among smokers with and without MI. Smokers with MI reported increased negative mood at some time points, but changes in negative mood and stress did not explain the relationship between condition and CPD reduction. Results highlight the need for mood management during smoking reduction and cessation efforts for smokers with MI and support the idea that ENDS may be an effective tool for smoking reduction for this group.


© The Author

Is Part Of

VCU University Archives

Is Part Of

VCU Theses and Dissertations

Date of Submission