linical approaches to smoking cessation are based on a goal of increasing the odds that a given person will quit smoking. Current treatment recommendations are based on metaanalyses of randomized clinical trials conducted over short intervals. An underlying assumption is that nicotine is the primary reason it is so difficult to quit and that pharmacotherapies are the “best” way to treat the underlying dependency. Data on the relative risk of developing tobacco related disease is used as the basis for selecting priority sub-populations (e.g., those who smoke the most). But should these data be used as the foundation for creating national strategies for the treatment of smokers? The presenter will use data from Canada to offer a different view. He will argue that national strategies should not be simple extrapolations of clinical treatments or relative risk ratios. He will argue that national strategies must employ different goals and outcome measures. For example, our aim should be to use available resources to maximally reduce the health and economic burden of smoking. Policies that create supportive environments, communication campaigns that build self-awareness and efficacy, combined with a triage system that matches smokers to different types of treatment should be the foundation of our approach. Traditional clinical approaches (especially pharmacotherapy) should be used selectively rather than universally.