The innovative Advanced Automatic Collision Notification (AACN) uses the URGENCY algorithm, which is studied and developed using the National Automotive Sampling System Crashworthiness Data System (NASS‐CDS). The objectives of this study are to clarify the adaptation of the URGENCY algorithm to Japan and to the U.S. and to ascertain the important issues in predicting injuries. Accident data for 2000–2009 were obtained from the Institute for Traffic Accident Research and Data Analysis in‐depth accident investigation database in Japan, and the NASS‐CDS in the U.S. Crashes are classified as head‐on crashes, frontal single‐vehicle crashes, nearside (driver‐side) crashes, and farside crashes.

The results of the present study show that the performance for frontal crashes is almost the same for the two countries;​ however, the performance for side‐impact crashes is different, and the balance between the sensitivity and the rate of overtriage for Japan is worse than that for the U.S. The under‐ and overtriage cases include some unusual accidents that indicate a need to improve the injury prediction algorithm for both Japan and the U.S. The cutoff value of the injury prediction algorithm is important for determining the overall balance of field triage decisions, and optimization must consider the accident conditions.