Varenicline for smoking reduction in smokers not yet ready to quit: A double-blind, proof-of-concept randomized clinical trial

Varenicline has demonstrated efficacy for quitting smoking. Its agonist and antagonist effects suggest that it would be efficacious for reducing cigarettes per day in smokers not yet ready to quit. To conduct a proof-of-concept placebo-controlled randomized clinical trial of varenicline for smokers willing to reduce, but not quit smoking. Smokers (N = 53) were randomized to receive 28-days of varenicline plus brief counseling or 28-days of placebo plus brief counseling. They were instructed to reduce their cigarettes per day with the goal of reducing by 50% from baseline to end-of-treatment. Most (82.7%) participants attended all four counseling sessions and more than half took their medication as prescribed on at least 80% of days. Approximately half of our participants reported at least one adverse event, but no serious adverse events were reported or discovered. Although twice as many smokers receiving varenicline made a quit attempt as compared to those receiving placebo, the data did not support the hypothesis that taking varenicline would have a benefit over placebo in meeting a 50% smoking reduction goal. Those who reduced their cigarettes per day by at least 50% and those who reduced their CO intake by at least 50% from baseline to end-of-treatment showed higher self-efficacy for quitting at end-of-treatment, however. These data support the proof-of-concept that smokers not ready to quit are willing to use varenicline while reducing the number of cigarettes smoked per day, and that successful reduction is associated with increased self-efficacy for later quitting. •Smokers meeting their 50% reduction goal showed higher self-efficacy for quitting at end-of-treatment.•Those receiving varenicline were twice as likely to make a quit attempt as those receiving placebo.•Participants in both groups were able to significantly reduce and maintain their reduced cigarette intake.•Neither participants, nor their clinicians could guess the group to which they were assigned at rates better than chance.