Do socioeconomic gradients in women's health widen over time and with age?

Soc Sci Med. 2004 May;58(9):1585-95. doi: 10.1016/S0277-9536(03)00368-X.


A population-based study was conducted to investigate changes over time in women's well-being and health service use by socio-economic status and whether these varied by age. Data from 12,328 mid-age women (aged 45-50 years in 1996) and 10,430 older women (aged 70-75 years) from the Australian Longitudinal Study on Women's Health were analysed. The main outcome measures were changes in the eight dimensions of the Short Form General Health Survey (SF-36) adjusted for baseline scores, lifestyle and behavioural factors; health care utilisation at Survey 2; and rate of deaths (older cohort only). Cross-sectional analyses showed clear socioeconomic differentials in well-being for both cohorts. Differential changes in health across tertiles of socioeconomic status (SES) were more evident in the mid-age cohort than in the older cohort. For the mid-aged women in the low SES tertile, declines in physical functioning (adjusted mean change of -2.4, standard error (SE) 1.1) and general health perceptions (-1.5, SE 1.1) were larger than the high SES group (physical functioning -0.8 SE 1.1, general health perceptions -0.8 SE 1.2). In the older cohort, changes in SF-36 scores over time were similar for all SES groups but women in the high SES group had lower death rates than women in the low SES group (relative risk: 0.79, 95% confidence interval 0.64-0.98). Findings suggest that SES differentials in physical health seem to widen during women's mid-adult years but narrow in older age. Nevertheless, SES remains an important predictor of health, health service use and mortality in older Australian women.

Publication types

  • Comparative Study
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Age Distribution
  • Age Factors
  • Aged
  • Aging / physiology
  • Ambulatory Care / statistics & numerical data
  • Australia / epidemiology
  • Cohort Effect
  • Female
  • Health Status Indicators*
  • Humans
  • Longitudinal Studies
  • Middle Aged
  • Regression Analysis
  • Socioeconomic Factors*
  • Women's Health Services / statistics & numerical data
  • Women's Health*