Interpreting educational inequalities in site-specific cancer mortality: statistical significance versus public health relevance

I read with interest the recent study by Tanaka et al1 examining ]educational inequalities in site-specific cancer mortality in Japan. The authors provide nationwide census-linked analyses across 23 cancer sites using age-standardised mortality rates (ASMRs), the Relative Index of Inequality (RII) and the Slope Index of Inequality (SII). Their work highlights the scale and complexity of educational disparities in cancer outcomes. I wish to extend their findings by emphasising the distinction between relative and absolute measures of inequality and its implications for public health prioritisation.

While the statistical analyses are robust, one interpretative consideration deserves attention. The very large study population allows detection of very small differences between educational groups. For example, among men, pancreatic cancer ASMRs were 36.5, 36.7 and 38.3 per 100 000 person-years in the high-education, middle-education and low-education groups, respectively, with a reported RII of 1.02. However, the simple absolute difference between the high-education and low-education…

Read the full article ›