Objectives: We developed a probabilistic decision model of cost-effectiveness for lung cancer (LC) screening with helical computed tomography (hCT) compared with chest x-ray (CXR) and no screening (NS) given uncertain efficacy and risks of screening in practice. Methods: Markov model comparing NS to CXR and hCT screening in 60-year old current smokers screened annually until age 75 as base case using published literature, Surveillance Epidemiology and End Results database, and Mayo Clinic data. In the base case, we assumed stage shifts observed with screening translate into survival benefits. Sensitivity analyses evaluated cohort ages for starting and stopping screening, screening compliance, smoking status, positive nodule management and treatment costs. An expected value of perfect information analysis (EVPI) was estimated to determine the value of further research to reduce current uncertainty. Results: In the base case, CXR cost $51,245/QALY vs. NS and hCT was dominated by CXR. The probability that CXR is cost-effective was 94.3% at a maximum acceptable ratio of $100,000/QALY (4.8% and 0.9% for NS and hCT, respectively). EVPI analysis suggested that at a maximum ICER of $100,000/QALY, further research would potentially be worth $55 million for the US population over ten years. When it was assumed that hCT screening did not result in any false positives necessitating invasive surgery, hCT cost $119,571/QALY vs. CXR. CXR cost $137,652 vs. NS for former smokers and was dominated by NS for never smokers; hCT was dominated in both these analyses. Results were sensitive to age at annual screening initiation and termination. Conclusions: Assuming stage shifts observed with LC screening translate into survival benefits, hCT was, as expected, most efficacious, but also had the highest false positive rate. The associated detrimental cost and quality of life effects resulted in hCT being dominated by CXR (less efficacious but more specific).