Clinical Question

Are tobacco cessation interventions effective for patients being treated for or recovering from other substance use disorders?

Evidence-Based Answer

Pharmacotherapy alone and pharmacotherapy plus counseling are effective for tobacco cessation among patients being treated for or recovering from alcohol and substance use disorders (number needed to treat [NNT] = 21; 95% confidence interval [CI], 8 to 58 for patients receiving pharmacotherapy alone; NNT = 15; 95% CI, 9 to 28 for patients receiving pharmacotherapy plus counseling). Counseling interventions without pharmacotherapy are not effective for tobacco cessation among patients in treatment for or recovery from substance use disorders. There is no evidence that tobacco cessation interventions affect abstinence rates from alcohol or other drugs.¹ (Strength of Recommendation: B, based on inconsistent or limited-quality patient-oriented evidence.)

Practice Pointers

Smoking remains the leading cause of preventable death in the United States, and approximately 19% of adults report tobacco use.² Among adults in early recovery from alcohol and substance use disorders, smoking rates are greater than 75%, with smoking-related mortality exceeding alcohol-related mortality in this population.³ Despite strong evidence supporting smoking cessation efforts in persons recovering from alcohol and substance use disorders, physicians remain apprehensive that smoking cessation may increase the risk of alcohol or substance use relapse.¹ The authors examined the effectiveness of various smoking cessation interventions among patients in treatment for or recovery from alcohol and other non-tobacco substance use disorders, and whether smoking cessation increased the risk of relapse.

In this review, 34 randomized controlled trials involving 5,796 patients examined tobacco cessation counseling alone (11 studies); pharmacotherapy alone (i.e., nicotine replacement therapy [NRT], non-NRT, or the two combined; 11 studies); and counseling in combination with pharmacotherapy (12 studies). The primary outcome was tobacco abstinence at the longest period of follow-up, and the secondary outcome was abstinence from alcohol and other drugs. Tobacco cessation counseling alone did not affect tobacco abstinence when results were pooled across the studies. Pharmacotherapy alone increased tobacco abstinence (NNT = 21; 95% CI, 8 to 58). Counseling plus pharmacotherapy also increased abstinence (NNT = 15; 95% CI, 9 to 28). Follow-up ranged from six weeks to 18 months. Tobacco cessation interventions were not associated with a difference in abstinence rates for alcohol or other drugs.

Tobacco cessation pharmacotherapies approved by the U.S. Food and Drug Administration (FDA) include NRTs (i.e., patch, gum, lozenge, inhaler, and nasal spray), bupropion (Zyban), and varenicline (Chantix). Although all tobacco cessation pharmacotherapies are better than placebo, this review did not compare the effectiveness of the different types among persons who were in treatment for or recovering from alcohol and other non-tobacco substance use disorders. However, a separate Cochrane analysis compared studies of the effectiveness of these medications in the general population.⁴ The analysis demonstrated that combination NRT (combining a long-acting and short-acting NRT) and varenicline had similar outcomes and were associated with the highest rates of tobacco cessation. Monotherapy NRT and bupropion had similar effectiveness, with higher rates of tobacco cessation than placebo, but not as high as combination NRT or varenicline.

Current guidelines recommend screening and treatment for tobacco use among patients being treated for alcohol and substance use disorders.^5,6 None of the FDA-approved tobacco cessation medications have interactions with medications that are used in the treatment of opioid or alcohol use disorders (i.e., methadone, buprenor-phine/naloxone [Suboxone], naltrexone [Revia, Vivitrol], acamprosate, or disulfiram [Antabuse]). Bupropion lowers the seizure threshold, so it should be avoided in patients with a history of seizures or who are at increased risk of seizures because of abrupt discontinuation of alcohol or benzodiazepines.⁷ Previously, bupropion and varenicline had boxed warnings about neuropsychiatric safety and increased risk of suicidality, but these warnings were removed in December 2016 after a randomized controlled trial found no increase in neuropsychiatric events with either drug compared with NRT or control. The FDA reports that there is still a small risk of these adverse effects and that physicians should counsel patients to stop taking the medication and to call their physician right away if they notice any changes in mood, behavior, or thinking.^8,9

The practice recommendations in this activity are available at http://www.cochrane.org/CD010274.

editor's note: The numbers needed to treat reported in this Cochrane for Clinicians were calculated by the author based on raw data provided in the original Cochrane review.