The derailment accident occurred on 25 April 2005. 562 passengers were injured and 107 people died. Based on our periodic training in mass disaster triage, we accepted 113 wounded victims regardless of the severity of their condition. Initially, Simple Triage and Rapid Treatment was supposed to be performed by an emergency physician as a triage officer but START is not the most effective method for this case because patients of different severity levels were admitted to our hospital. Therefore, we performed triage by inspection and palpation based on our experience and intuition. Patient's facial color, breathing patterns, and trauma were processed for inspection. Furthermore, palpation was carried on without counting pulse and we diagnosed victims based upon strength of pulse and sensation of cold. From inspirational and sensuous diagnosis, the severity was determined. Comparing START with experiential and intuitive triage, differences between level of consciousness, the severity of trauma, and vital sign were examined. In addition, which components were necessary for primary survey was also reviewed.

Severe or moderate, which were evaluated from ISS, vital sign, and Revised Trauma Score: RTS, Probability of survival: PS, were classified and then agreement rate, sensitivity, and specificity between START (S model) and triage (D model) were assessed. For START, data was extracted from clinical records.

Agreement rate between the S model and the D model was 76.9%. Moreover, at the percentage of questions answered, the S model showed 84.2% and the D model showed 68.4%. Thus the experiment showed significant difference. 5 over triage cases appeared when the D model was operated. RTS, vital sign, and PS did not show great differences.

It could be difficult and insufficient to use START when a mass disaster occurs, however experiential and intuitive triage is also effective as START.