2015 年 25 巻 2 号 p. 2_313-2_325
Over 1.1 million injuries and accidents happen in schools every year in Japan. Most of them happen during sports activities such as physical education or sports club activities. In order to prevent accidents it is important to elucidate the contributing factors by focusing on the occurrence. The purpose of this study was to elucidate the characteristics and factors contributing to sudden death during sports activities in schools, by using text-mining analysis. Data was collected from the database of death and disability cases publicized by the Japan Sports Council. There were 127 cases of sudden death during sports activities from 2005 to 2012. The results by morphological analysis and co-occurrence network analysis revealed two characteristics. The first point was that the ′act of running′ had a close relationship with the occurrence of sudden death in schools. The second was that not a few cases of sudden death happened even though the children had done warm-up exercises before the accident. Our results suggest that the relationship with the act of running, warm-up exercise and the occurrence of sudden death should be kept in mind when sports activities in schools are held. In addition, our study shows that data-mining analysis has a potential to objectively analyze more than one million injuries and accidents in schools.