BACKGROUND: Pneumothorax is one of the leading causes of preventable death in trauma patients. Chest radiograph has a lower sensitivity than a computed tomography (CT) scan for the diagnosis of pneumothorax. OBJECTIVES: The objective of this study is to assess the accuracy of ultrasound (US) in diagnosing post-traumatic pneumothorax using a simplified diagnostic algorithm. METHODS: This prospective study was carried out in two academic emergency departments with a combined annual emergency department census of 70 000 visits per year by convenient sampling. Emergency physicians underwent a 2-h training course using a two-step diagnostic algorithm. The trauma patients included suspected of having post-traumatic pneumothorax were evaluated by thoracic US and chest CT scan. The accuracy of US in identifying pneumothorax was measured by calculating the sensitivity, specificity, positive predictor value, and negative predictor value using CT scan as the gold standard. The performance of US was compared with chest radiograph, which was also done as a part of standard trauma care for all patients. RESULTS: From June 2009 until July 2009, a total of 153 patients were included. US had a sensitivity of 86.4%, a specificity of 100%, a positive predictor value of 100%, and a negative predictor value of 95.6%. Chest radiograph showed a sensitivity of 48.6%, a specificity of 100%, a positive predictor value of 100%, and a negative predictor value of 85.1%. The mean time to perform chest radiograph was 12 min, which was significantly higher than US, with a mean time of 2 min. All missed pneumothoraces in US evaluation were small in size. CONCLUSION: After just a 2-h training course, emergency physicians showed a good success rate in finding pneumothoraces. Thoracic US can be an easy to learn and an accurate diagnostic modality for the detection of traumatic pneumothorax in emergency departments.