At follow-up of 751 subjects receiving a brief nurse-administered anti-smoking intervention in general practice, 135 subjects (18%) reported stopping smoking, of whom 44 (6%) reported sustained cessation for one year. The demographic, social and attitudinal characteristics of these subjects were compared with 616 subjects who continued to smoke. The most important predictors of cessation were intention to stop (OR 5.1, 95% CI 2.1-12.0), personal rating of likelihood of cessation (OR 4.9, 95% CI 2.8-8.5), nurse rating of likelihood of cessation (OR 4.0, 95% CI 2.2-7.4), and smoking habit of partner (1.9, 95% CI 1.3-2.9). As practice nurses are able to distinguish likely quitters from those who are not motivated and less likely to succeed, it is important to decide whether it is more cost effective to target support at the motivated or to spend more time encouraging less motivated. The most challenging, but possibly the most rewarding, task is to try to reduce the high proportion of new ex-smokers who relapse. Although 41.1% (95% CI 28.1, 58.0) of those expressing a definite intention to stop smoking gave up, only 17.9% (95% CI 8.9, 30.4) achieved sustained cessation. However, as sustained cessation is strongly predicted by social variables, such as marital status and time spent in the company of smokers, preventing relapse may not be easy to achieve through medical intervention alone.