Suicide, Socio-economic Inequalities, Gender, and Psychiatric Disorders
Commentary: Educational Levels and Risk of Suicide in Japan: The Japan Public Health Center Study (JPHC) Cohort I

This issue of the Journal of Epidemiology includes a report titled “Educational Levels and Risk of Suicide in Japan: The Japan Public Health Center Study (JPHC) Cohort I” by Kimura and colleagues.1 These authors investigated the association between education level and suicide risk in a population-based cohort in Japan. They considered educational level to serve as a surrogate marker for socioeconomic status. In total, 21 829 males and 24 327 females were included in the analysis reported in this article, and 218 males and 81 females died by suicide during a median follow-up period of 21.6 years. Their results revealed that, compared with junior high school graduates as a reference category, the hazard ratios in men were 0.91 (95% confidence interval [CI], 0.67–1.22) for high school graduates, 1.01 (95% CI, 0.54–1.90) for junior/career college graduates, and 0.47 (95% CI, 0.24–0.94) for university graduates or those with higher education. For women, the hazard ratios were 0.44 (95% CI, 0.24–0.79) for high school graduates, 0.56 (95% CI, 0.22–1.43) for junior/career college graduates, and 0.55 (95% CI, 0.06–4.03) for university graduates or those with higher education. The authors concluded that high educational levels among Japanese men and women were associated with a reduced risk of suicide, suggesting that higher educational levels have a protective role against suicide risk in the Japanese society. This is the first large-scale population-based prospective study in Japan to examine the association between educational levels and the risk of suicide after controlling for individual lifestyle and socioeconomic factors. The study is organized well and is important for future suicide-prevention implications in Japan. The results for males in the study are generally consistent with those in previous studies from other countries or regions. However, I think that we need to interpret the results for the female participants more carefully because only a small number of female participants, particularly those with a higher level of education, died by suicide during the follow-up period. The European Union (EU) Working Group on SocioEconomic Inequalities in Health previously reported a European overview of socioeconomic inequalities in mortality arising from suicide among men and women.2 The EU working group used a prospective follow-up of censuses matched with vital statistics in 10 European populations in the early 1990s. The mean follow-up period was 4 years. In this earlier article, level of education and housing tenure were used as socioeconomic indicators. The main findings of this EU study were that socioeconomic inequalities in suicide were pervasive in almost all male populations and that inequalities were far less pronounced (or even reversed, in some cases) in women, particularly when educational status was considered.2 The authors of the EU study suggested that this gender difference could be explained by gender differences in healthrelated and life-threatening behaviors, such as alcohol or drug misuse, which were known risk factors for suicide and were more prevalent among men and among people in lower socioeconomic groups. In addition, it was reported that suicide risks in women resembled those in men more closely where house ownership was concerned rather than education. This may be because house ownership was mostly an attribute of the household and was thus shared by both spouses, whereas education is an individual attribute and thus may be more sensitive to differences in gender. As is the case with educational level, it has been reported that school performance also affects suicidal behavior differently between males and females. For example, Gunnell et al examined the association of school performance scores at the ages of 16 years and 18/19 years with suicide in a record