Varenicline versus placebo for waterpipe smoking cessation: a double-blind randomized controlled trial

Waterpipe tobacco smoking is a growing public health concern. There is limited research using pharmacotherapy and no research using varenicline (established treatment for smoking cessation) in waterpipe smokers. We tested the efficacy of varenicline in achieving abstinence from all tobacco use among waterpipe smokers.

Two-arm, parallel group, placebo-controlled, double-blind, multi-centre (n = 4), individually randomized trial with follow-up to 25 weeks.

District general hospitals and catchment communities within four districts of Punjab, Pakistan.

Adult daily waterpipe smokers (n = 510; 253 in varenicline and 257 in placebo arms), who were interested in quitting, were recruited and analysed between March and November 2016. Of these, 220 (87%) in the varenicline and 239 (93%) in the placebo arms completed all follow-ups. Participants were on average aged 49 [standard deviation (SD) = 15.2] years, daily smokers and smoked for the last 27 (SD = 15.9) years. More than half (261, 51.2%) also smoked cigarettes.

All trial participants received two structured sessions of behavioural support (of 30 and 10 minutes) one at the time of registration and the other 1 week later. Participants were randomized to varenicline (active arm) and placebo (control arm) stratified on district, sex and concomitant cigarette smoking. Varenicline and placebo were dispensed as identical unlabelled tablets for 12 weeks: 0.5 mg for 1 week (once on days 1-3, twice on days 4-7) and 1 mg for the subsequent 11 weeks (twice daily).

The trial participants were followed-up for a period of 25 weeks post-randomization. The primary outcome was 7-day repeated point prevalence abstinence from all forms of tobacco, self-reported at each of weeks 5, 12 and 25, verified by carbon-monoxide cut-off < 10 parts per million.

No evidence of statistically significant difference in repeated point prevalence abstinence between the varenicline (12 of 253; 4.7%) and placebo (11 of 257; 4.3%) arms (relative risk = 1.11, 95% confidence interval = 0.50-2.47, P = 0.80) was observed (Bayes factor = 0.048). Adverse events reported in 27 participants were 34 (15 in varenicline and 19 in placebo); none was serious.

Varenicline was not more effective than placebo in aiding cessation of tobacco use in long-term daily waterpipe smokers.