Skip navigation

Zyban for smoking cessation


Is this new Zyban stuff really any good for stopping people smoking? That's a recent question put to Bandolier by several readers. Not an easy one to answer until someone draws your attention to the fact that bupropion (Zyban) has been included in a Cochrane review on the use of anxiolytics and antidepressants for smoking cessation [1].

There are apparently two reasons to believe antidepressants and anxiolytics might help in smoking. First, anxiety and depression are symptoms of nicotine withdrawal, and smoking cessation sometimes precipitates depression. Second, smoking appears to be due, in part, to deficits in dopamine, serotonin and noradrenaline, all of which are increased by anxiolytics and antidepressants.


This involved a typically thorough Cochrane search, using the trial register maintained by the Cochrane Tobacco Addiction Group. Randomised trials compared anxiolytic or antidepressant drugs to placebo or to an alternative therapeutic control for smoking cessation and with at least six months follow-up, and using the most rigorous definition of abstinence.


For most anxiolytics and antidepressants there was only a single trial, but two of nortriptyline and four of bupropion. Only these latter two drugs increased smoking cessation.

For bupropion, four randomised studies compared 12-month cessation rates with placebo. For bupropion 300 mg a day for seven to 12 weeks, 105/518 (20%) smokers had stopped at 12 months, compared with 34/430 (8%) with placebo (Figure). The relative benefit was 2.6 (95% confidence interval 1.8 to 3.8) and the number needed to treat was 8.2 (6.1 to 13).

Figure: Smoking cessation at 12 months with bupropion and placebo


Bupropion looks better than nicotine replacement therapy. The placebo response rate is the same, but cessation rates are higher, and the NNT, at 8, is lower than that for most nicotine replacement therapies, at about 12-15 ( Bandolier 54 ). Given that the smoking of cigarettes has such major health consequences ( Bandolier 9 , 14 ), clearly there will be benefits for some people. Smokers don't have to be depressed to use bupropion. There are adverse effects, and the review [1] quotes a risk of seizures estimated at 1 in 1000, as well as minor adverse effects like nausea and insomnia, but without numbers.

The other comment is just how useful the Cochrane Library is becoming. The latest issue has over 800 reviews, and it will soon top 1000. Quick literature searches elsewhere reveal that the total number of systematic reviews is probably of the order of 5,000 to 10,000, many of which are poor or out of date, so Cochrane is fast coming to house a significant proportion of all worthwhile reviews. The Library also houses many sources of other reviews as well. There may be some way to go to make the Library less clunky and the reviews more user friendly, but that will come.

What we could do with is a single source of all known systematic reviews. This has been done for particular areas, but knowing that there was a single place to search for reviews, good or bad, would be a decided advantage, and not a particularly difficult job.


  1. JR Hughes, LF Stead, T Lancaster. Anxiolytics and antidepressants for smoking cessation (Cochrane Review). In: The Cochrane Library, Issue 3, 2000. Oxford: Update Software.

previous or next story in this issue