Why we did this Cochrane Review

We wanted to find out:

- which treatments (medicines and e-cigarettes) help people to stop smoking;

- how these treatments compare to each other;

- whether there are ways of providing these treatments that mean they are more likely to help people stop smoking (e.g. different doses or treatment lengths);

- whether these treatments are likely to cause serious harms; and

- whether certain treatments are better tolerated, as indicated by fewer people leaving a study due to treatment.

What did we do?

We searched for studies that looked at these treatments to help adults quit smoking. We looked for randomised controlled trials, where the treatments people received were decided at random. This type of study usually gives the most reliable evidence about the effects of treatments. We compared all treatments with each other using a method called network meta-analysis.

Search date: 29 April 2022

What we found

We found 332 studies that met our criteria and 319 of these provided information that we could use in our analyses. These included 157,179 adults who smoked tobacco cigarettes. Most of the studies took place in the USA or Europe. The studies compared the effects of the stop smoking treatments listed above with:

- no medicine/e-cigarettes for stopping smoking;

- e-cigarettes that did not contain nicotine;

- placebo (a dummy medication); and

- other types of stop smoking medicine or e-cigarettes.

What are the main results of our review?

E-cigarettes, varenicline and cytisine were most likely to help people quit smoking. For every 100 people, 10 to 19 are likely to quit using an e-cigarette; 12 to 16 using varenicline; and 10 to 18 using cytisine. This is compared to the 6 in 100 people likely to quit when using no medicine/e-cigarette or placebo. People using two forms of nicotine replacement therapy at the same time, for example, a combination of nicotine patch and nicotine gum, seemed to have similar rates of quitting to people using e-cigarettes, varenicline and cytisine. Nicotine patches alone, another form of nicotine replacement therapy alone (such as gum, lozenge) and bupropion appeared to help fewer people quit but still work better than no medicine/e-cigarette or placebo (8, 9 and 9 people per 100, respectively). Nortriptyline appeared to result in the lowest number of people quitting smoking; for every 100 people using nortriptyline 6 to 11 are likely to quit.

We are moderately confident that bupropion could rarely cause some serious health effects. The information we have for other treatments does not provide clear evidence of serious harms. For all treatments, findings suggest very few people experience serious harms when using them.

How confident are we in our results?

We are confident that e-cigarettes, cytisine, varenicline, nicotine replacement therapy and bupropion help people stop smoking. We do not expect more evidence will change these results. However, more evidence on how these treatments compare to one another, particularly in relation to harms, would be useful. Due to the nature of our analyses we were not able to judge our confidence in the evidence for combination nicotine replacement therapy (two types used together). We are moderately confident that nortriptyline also helps people to stop smoking, but are less confident in our results for non-nicotine e-cigarettes and for potential harms of most of the treatments. We still need more evidence on potential harms and hope more studies will report on these in future; however, nicotine replacement therapy has been used since the 1980s with no evidence of serious harms.