Low-dose CT scan screening greatly improves the likelihood of detecting small nodules and, thus, of detecting lung cancer at a potentially more curable stage. To evaluate the cost-effectiveness of a single baseline low-dose CT scan for lung cancer screening in high-risk individuals, data from the Early Lung Cancer Action Project (ELCAP) was incorporated into a decision analysis model comparing low-dose CT scan screening of high-risk individuals (ie, those > or = 60 years with at least 10 pack-years of cigarette smoking and no other malignancies) to observation without screening. Cost-effectiveness was expressed as the incremental cost per year of life saved. The analysis adopted the perspectives of the health-care system. The probability of the different outcomes following the decision either to screen or not to screen an individual at risk was based on data from ELCAP and the Surveillance, Epidemiology, and End Results Registry or published data, respectively. The cost of the screening and treatment of patients with lung cancer was established based on data from the New York Presbyterian Hospital's financial system. The base-case analysis was conducted under the assumption of similar aggressiveness of screen-detected and incidentally discovered lung cancers and then was followed by multiple sensitivity analyses to relax these assumptions. The incremental cost-effectiveness ratio of a single baseline low-dose CT scan was 2,500 US dollars per year of life saved. The base-case analysis showed that screening would be expected to increase survival by 0.1 year at an incremental cost of approximately 230 US dollars. Only when the likelihood of overdiagnosis was > 50% did the cost effectiveness ratio exceed 50,000 US dollars per year of life saved. The cost-effectiveness ratios were also relatively insensitive to estimates of the potential lead-time bias. A baseline low-dose CT scan for lung cancer screening is potentially highly cost-effective and compares favorably to the cost-effectiveness ratios of other screening programs.