Varenicline versus nicotine patch with brief advice for smokers with substance use disorders with or without depression: effects on smoking, substance use and depressive symptoms

Abstract

Aims: Varenicline was compared with transdermal nicotine (NRT) for smokers with current substance use disorders (SUD) for effects on 3-month smoking abstinence (primary outcome) and, secondarily, on 3- and 6 month abstinence while adjusting for medication adherence, and on additional smoking and substance use outcomes. Moderation by major depressive disorder history (MDD) and adherence were investigated.

Design: Double-blind double-placebo-controlled randomized design, stratifying by MDD, gender and nicotine dependence, with 3 and 6 months follow-up.

Setting: University offices in Rhode Island, USA.

Participants: Adult smokers (n = 137), in SUD treatment, substance abstinent <12 months (n = 77 varenicline, 60 NRT).

Intervention and comparator: Twelve weeks of varenicline (2 mg/day, after 1-week dose run-up) or NRT (21 mg/day decreasing to 7 mg/day).

Measurements: Primary: point-prevalence smoking abstinence (7-day, confirmed) at 3 months. Secondary: point-prevalence abstinence at 6 months, quantity and frequency of smoking and substance use at 3 and 6 months, and within-treatment abstinence, medication adherence and depressive symptoms. Smoking outcome analyses were repeated controlling for adherence and investigating adherence as a moderator.

Findings: Effects on 3-month abstinence were P < 0.065 without a covariate (Bayes factor 3.35, supporting the effect strongly) and differed significantly when controlling for baseline smoking [varenicline: 13%, NRT: 3%; odds ratio (OR) = 4.81, 95% confidence interval (CI) 1.00, 23.13, P < 0.05]. The threefold difference at 6 months was not significant. Medication effect on abstinence across time was significant (P < 0.05) covarying adherence and baseline smoking (OR = 6.40, 95% CI = 1.00, 40.93). Medication differences in 3-month abstinence occurred among participants with ≥ 77% adherence (P < 0.02). No significant medication effects on heavy drinking, drug use or depressive symptoms were found.

Conclusions: Varenicline appears to improve the chances of achieving at least 3 months of smoking abstinence in smokers with substance use disorders trying to stop, compared with transdermal nicotine patches, the effect being independent of history of depressive disorder.

Keywords: Brief advice; depression; nicotine dependence; point-prevalence abstinence; smoking cessation; substance use disorders; varenicline.

Figure 1

Consolidated Standards of Reporting Trials (CONSORT) diagram of participant flow

Figure 2

Percentage of participants compliant with medication (> 85% of capsules) by assignment to varenicline versus placebo capsules after the week of dose run-up

Figure 3

Beck Depression Inventory score at baseline and weeks 2, 5, 9 and 13 after starting dose run-up. generalized estimating equations (GEE) showed a significant main effect for time, Wald χ² (d.f. = 1) = 13.43, P < 0.001, but not for medication or medication by time. Depression scores decreased from pre-treatment to week 2 [mean difference = −4.99, standard error (SE) = 0.97, 95% confidence interval (CI) = −7.65, −2.33), P < 0.01], with no significant change thereafter. NRT = nicotine replacement therapy; VAR = varenicline