Abstract
INTRODUCTION: In a 2 × 2 factorial clinical trial with 300 individuals with current or past major depressive disorder (MDD), varenicline significantly increased smoking cessation, compared to placebo, whereas behavioral activation for smoking cessation yielded similar quit rates compared to standard behavioral treatment. In this secondary analysis, we evaluated whether participant characteristics at baseline predicted abstinence or moderated the effects of behavioral treatment or pharmacotherapy on abstinence. METHODS: The sample was racially and socioeconomically diverse and spanned varied psychiatric presentations: 49 % of participants had current MDD, 44 % had other mental health disorders, and only 27 % were being treated for their depression. Self-reported 7-day point prevalence abstinence at end-of-treatment (EOT; Week 14) and long-term follow-up (Week 27) was confirmed using breath carbon monoxide (≤6 parts per million). RESULTS: In a logistic regression model, greater likelihood of abstinence at EOT was associated with varenicline vs. placebo (OR=5.52, 95 % CI=2.62-11.60), lower cigarette dependence (OR=0.42, 95 % CI=0.19-0.90), and fast nicotine metabolism (OR=1.64, 95 % CI=1.06-2.54). In second logistic regression model, greater likelihood of Week 27 abstinence was associated with varenicline vs placebo (OR=3.06, 95 % CI=1.32-7.07) and fast nicotine metabolism (OR=2.06, 95 % CI=1.20-3.54). None of the baseline factors interacted with behavioral or pharmacological treatment to predict abstinence at either time point. CONCLUSIONS: Varenicline should be a primary treatment consideration for individuals with MDD. Stable psychiatric status is not required for favorable treatment outcome. Further effort is needed to identify behavioral interventions that can be added to varenicline to improve quit rates for individuals with MDD.