Discussion: A MEDLINE Search revealed two meta-analyses of controlled trials. These trials and others are analysed. A meta-analysis of 39 trials before nicotine patch therapy became widely available found on average 8.4% fewer smokers at six months and 5.8% fewer at 12 months in the experimental than the control groups (Kottke, Battista, DeFriese, Brekke, JAMA 1988;259(19):2883-9). Factors related to more quitting were: both individual and group counselling (average 22% reduction compared to controls); more than 8 contacts with the programme (14% reduction); more than two interventions (12% reduction); and counselling by both physicians and non-physicians (8% reductions). A meta-analysis of 53 trials of nicotine therapy showed that 19% who used nicotine therapy and 11% of controls had stopped smoking 6-12 months later, a 71% increase in the odds of abstinence (Silagy, Mant, Fowler, Lodge, Lancet 1994;343:139-141). The odds ratios of not smoking 6 - 12 months later were 19.7% in trials which counselled more than 30 minutes at the first assessment or more than 2 reinforcement visits; and 10.5% in trials which did not. Relapse rates were high and it was estimated that 10 patients would need to be treated with nicotine patches and high intensity support to provide one abstainer.