Accurate staging of non-small cell lung cancer (NSCLC) is critical for optimal management. Minimally invasive pathologic assessment of mediastinal lymphadenopathy is increasingly being performed. The cost-benefit (minimization of health care costs) of such approaches, in comparison with traditional surgical methods, is yet to be established. Decision-tree analysis was applied to compare downstream costs of endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA), conventional TBNA, and surgical mediastinoscopy. Calculations were based on real costs derived from actual patient data at a major teaching hospital in Melbourne, Australia. One- and two-way sensitivity analyses were undertaken to account for potential variation in input parameter values. For the base-case analysis, initial evaluation with EBUS-TBNA (with negative results being surgically confirmed) was the most cost-beneficial approach (AU$2961) in comparison with EBUS-TBNA (negative results not surgically confirmed) ($3344), conventional TBNA ($3754), and mediastinoscopy ($8859). The sensitivity of EBUS-TBNA for detecting disease had the largest impact on cost, whereas the prevalence of mediastinal lymph node metastases determined whether surgical confirmation of negative EBUS-TBNA results remained cost-beneficial. Our study confirms that minimally invasive staging of NSCLC is cost-beneficial in comparison with traditional surgical techniques. EBUS-TBNA was the most cost-beneficial approach for mediastinal staging of patients with NSCLC across all studied parameters.