Educational inequalities in all-cause and cause-specific mortality in Japan: national census-linked mortality data for 2010-15.

Abstract

Due to the lack of a national mortality inequality monitoring framework, the overall picture in Japan remains unclear. Here, we investigated educational inequalities in mortality and their cause-specific contribution in Japan. Data were obtained by linking the 2010 Japanese population census and death records between 1 October 2010 and 30 September 2015. We included 7 984 451 Japanese people aged 30-79 years who had a unique 'matching key' generated by sex, birth year/month, address (municipality), marital status and age of spouse (9.9% of the total census population). We computed population-weighted all-cause and cause-specific age-standardized mortality rates (ASMRs) by education level. In addition, we calculated the slope index of inequality (SII), relative index inequality (RII) by education level, and population attributable fraction (PAF) referenced with the highest education (e.g. university graduation). Individuals with less education had higher all-cause and cause-specific ASMRs than highly educated individuals. All-cause SII (per 100 000 person-years) values were 433 (95% CI: 410-457) for men and 235 (95% CI: 217-252) for women. RII values were 1.48 (95% CI: 1.45-1.51) for men and 1.47 (95% CI: 1.43-1.51) for women. Estimated PAFs, excess premature deaths caused by educational inequalities, were 11.6% for men and 16.3% for women, respectively. Cerebrovascular diseases, ischaemic heart diseases and lung cancer were the major contributors to mortality inequalities for both sexes. This first census-based comprehensive report on cause-specific educational mortality inequalities suggested that differences in unfavourable health risk factors by educational background might be associated with these inequalities in Japan.