The editorial by Butler & Malone ^{Reference Butler and Malone1} focused on non-suicidal self-injury (NSSI), proposed as a new category in DSM-5, and described the criteria of this item. The authors also provided information about NSSI among young people. In Japan in August 2012, the Japanese Cabinet decided to enhance suicide prevention measures for younger people, given the increasing suicide rate among students, partly as a result of bullying. ² Thus, suicide prevention measures that are effective among younger age groups in Japan, and particularly among teenagers, must be designed promptly.

There were 58 509 suicides in Japan in 2011 and 2012 combined, according to the National Police Agency (www.npa.go.jp). This number included 1209 suicides of young people under the age of 19 (2.1%). The causative factors were classified as ‘family problems’, ‘health problems’, ‘economic and life problems’, ‘work problems’, ‘problems of relations between the sexes’, ‘problems in school’ or ‘others’. The most common causative factor for suicide among young people under the age of 19 was ‘problems in school’ (33.0%), followed by ‘health problems’ (23.7%); therefore, determining the detailed aspects of ‘problems in school’ for teenagers will contribute to effective suicide prevention measures for that age group.

‘Problems in school’ included ‘worrying about entrance exams’, ‘worrying about one's academic progress’, ‘poor academic performance’, ‘poor relationships with teachers’, ‘bullying’, ‘discord with schoolmates’ and ‘others’. Of these, ‘poor academic performance’ (31.1%) was the most prevalent causative factor, followed by ‘worrying about one's academic progress’ (25.6%). The incidence of ‘bullying’ (which has been taken into account in currently used prevention strategies) and measures and responses to deal with ‘poor academic performance’ and ‘worrying about one's academic progress’ must be explored in conjunction with the results of the current study.

It is necessary to discuss specific and effective suicide prevention measures from various viewpoints. ^{Reference Hirayasu, Kawanishi, Yonemoto, Ishizuka, Okubo and Sakai3,Reference Tochigi, Nishida, Shimodera, Oshima, Inoue and Okazaki4} We therefore propose that understanding the criteria of NSSI in DSM-5 is important, as noted by Butler & Malone.