OBJECTIVES: Smoking cessation models have typically evaluated the impact of a single quit attempt on long term outcomes in smokers but smoking cessation is characterized by multiple quit attempts.  A DES was developed to simulate lifetime smoking patterns and the impact on smoking-related complications and costs.  This study evaluates the effect on predictions when modeling single versus multiple quit attempts. METHODS: Using data from trials, surveys, and the literature, the DES simulates individuals’ lifetime smoking behaviors and their impact on outcomes.  The simulation assigns and reassigns the initial outcomes of each quit attempt, time between quit attempts, relapses, and interventions used in each attempt (varenicline, bupropion, nicotine replacement, behavioral modification, unassisted).  Comorbidities include myocardial infarction, stroke, COPD, and lung cancer.  Market survey data are used to assign the initial intervention for quit attempts. Only direct costs (2010 $US) are considered.  All outcomes are discounted at 3%/year. RESULTS: When analyses are restricted to a single quit attempt, mean life expectancy in the population is 15.8 years, and QALYs 13.2; the lifetime costs of treatment and smoking related comorbidities average $55,925. Allowing for multiple quit attempts (average 7.6 attempts/smoker) increases the average time individuals spend abstinent by 8.8 years.  Consequently, predicted life expectancy increases by 1.1 years; QALYs by 0.9. Despite increased smoking intervention costs, total lifetime costs fall by $3300/smoker.   Analyses comparing initial varenicline treatment to mixed initial treatments and allowing multiple versus single quit attempts reduces varenicline-related predicted health gains and cost offsets, although both groups have better outcomes with multiple quit attempts. The reduction is apparent because individuals initially on less effective treatments are able to quit smoking in subsequent attempts. Nevertheless, varenicline is dominant or highly cost-effective in both scenarios.  CONCLUSIONS: Allowing multiple rather than single quit attempts in simulating outcomes for smokers provides better information for decision making.