Should smokers be offered assistance with stopping? Many governments are actively considering whether and how to provide their population with assistance with smoking cessation. Arguments have been raised against this, but these are often based on fallacies (e.g. most smokers stop without help so assistance is unnecessary). This editorial counters these fallacies so that a constructive debate can be had about the role of cessation assistance in the tobacco control strategies for a given population. Article 14 of the Framework Convention on Tobacco Control [1] states that every country should provide smoking cessation assistance, and implementation of this is now being considered by many countries. However, it has been argued that offer of assistance with stopping should not be a major part of a country’s tobacco control strategy e.g. [2]. This viewpoint has been restated recently and gained some media attention (e.g. [3]). It is therefore important to draw attention to fallacies in the arguments that are often used against providing assis- tance with stopping smoking so that discussion on its role within a comprehensive tobacco control strategy can be based upon evidence and logic rather than misconceptions. FALLACY 1: MOST SMOKERS STOP WITHOUT HELP SO PROVIDING ASSISTANCE IS UNNECESSARY The fact that most smokers who stop do so without assis- tance does not mean that this is the most effective method of stopping. It merely reflects the fact that the numbers attempting to stop without assistance are greater than those trying to stop with it. To illustrate this point, if 1000 smokers try to stop without assistance and have a 5% chance of success this will create 50 ex-smokers; if 100 smokers try to stop with assistance and have a 20% chance of success this will create 20 ex-smokers. So in this example, more than twice as many smokers will have stopped without assistance as with it, despite the fact that doing it this way was four times less effective. To argue that unassisted cessation is more effective purely from the numbers stopping is to misunderstand this arithmetic. In fact, most smokers do not stop before smoking has cost them years of life expectancy. In England only 37% of those who have ever smoked for at least a year manage to stop by the age of 35 years [4]. After that age, each year that stopping is delayed costs an average of 3 months of life [5], and smoking at all ages causes substantial harm to others, particularly children. It is therefore vital for smokers to stop at the earliest possible opportunity, and for every quit attempt to have the best possible chance of success. Of course, smokers differ in the extent to which they are dependent and the ease with which they can quit. In many countries smoking is concentrated in more disad- vantaged groups and, on average, dependence is greater in these smokers [6,7]. To implement tobacco control strategies such as increasing taxation, and mass media campaigns that marginalize smokers with social eco- nomic and health disadvantage without providing assis- tance with stopping smoking, is to make them doubly disadvantaged, and must be morally indefensible. FALLACY 2: PROMOTING HELP WITH STOPPING IS COUNTERPRODUCTIVE BECAUSE IT MAKES SMOKERS THINK THEY ARE ADDICTED AND SO FEWER TRY TO STOP This is supposition. If it were true, then smokers who had taken on board the message that they were ‘addicted’ would be less likely to try to stop; yet in a recent study, smokers who believed they were addicted were actually more likely to make quit attempts than other smokers [4]. It is true that smokers in the United Kingdom, which has placed greater emphasis on assisting smokers to stop than other countries, recall making fewer quit attempts than those in other countries [8]. However, there may be many reasons for this and as methods to aid cessation have become promoted more widely in England, no decrease in the proportion making quit attempts has been observed [9,10]. FALLACY 3:THE RESULTS OF RESEARCH INTO ASSISTED CESSATION DO NOT APPLY TO THE ‘REAL WORLD’ One argument used here is that randomized controlled trial evidence of smoking cessation support can be dis- counted, as it has not proved possible to blind participants as to their allocation to active versus control treatments. Therefore, the results are attributable to participants in the ‘active’ treatment group expecting to do better. If the effects of support were merely expectancy or placebo effects, any medication with side effects and any plausible behavioural support package should show benefits. However, there are many examples of putative pharma- cological smoking cessation treatments with obvious side effects that have failed to show benefits (e.g. [11,12]), and EDITORIAL doi:10.1111/j.1360-0443.2010.03111.x © 2010 The Authors, Addiction © 2010 Society for the Study of Addiction Addiction, 105, 1867–1869