Municipal socioeconomic status and mortality in Japan: sex and age differences, and trends in 1973-1998

Soc Sci Med. 2004 Dec;59(12):2435-45. doi: 10.1016/j.socscimed.2004.04.012.


The purpose of this study was to examine the sex and age differences and the time trends in the association between municipal socioeconomic status (SES) and all-cause mortality across Japan from 1973 to 1998. Sex-specific mortality of municipalities (N=3319 in 1995) by age groups (total, under 75-year, and over 75-year populations) was linked to municipal SES indicators related to income, education, unemployment and living space, and two SES composite indices formulated by principle component analysis (Index 1 related to lower income and education, and Index 2 related to unemployment and overcrowding). The relation was assessed using mortality gradients by SES quintiles and Bayesian hierarchical Poisson regression. The results showed that a lower SES was related to higher mortality for all SES indicators and composite indices. The mortality gradient was steeper for the under 75-year population than the total and over 75-year populations, and the relation between mortality and income- and education-related indicators/index was stronger for males than for females. The time trend showed an increase in the relation for Index 2, while a decrease for Index 1. This study demonstrated that lower municipal SES had an adverse influence on population health, and the influence was marked for males and premature death. Although a substantial health disadvantage still remained in lower SES areas, the impact of SES factors on geographical health variation changed over time; the association with mortality has weakened for income and education, while it has strengthened for unemployment and living space.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Age Distribution
  • Aged
  • Female
  • Health Status Indicators*
  • Humans
  • Japan / epidemiology
  • Male
  • Mortality / trends*
  • Regression Analysis
  • Sex Distribution
  • Social Class*
  • Urban Health / statistics & numerical data
  • Urban Health / trends*