Regional variation in the survival and health of older Australian women: a prospective cohort study

Objective: Older people may act as sensitive indicators of the effectiveness of health systems. Our objective is to distinguish between the effects of socio-economic and behavioural factors and use of health services on urban-rural differences in mortality and health of elderly women.
Methods: Baseline and longitudinal analysis of data from a prospective cohort study. Participants were a community-based random sample of women (n=12778) aged 70-75 years when recruited in 1996 to the Australian Longitudinal Study on Women's Health. Measures used were: urban or rural residence in Australian States and Territories, socio-demographic characteristics, health related behaviour, survival up to 1 October 2006, physical and mental health scores and use of medical services.
Results: Mortality was higher in rural than in urban women (hazard ratio, HR 1.14; 95% CI, 1.03,-1.26) but there were no differences between States and Territories. There were no consistent baseline or longitudinal differences between women for physical or mental health, with or without adjustment for socio-demographic and behavioural factors. Rural women had fewer visits to general practitioners (odds ratio, OR=0.54; 95% CI, 0.48-0.61) and medical specialists (OR=0.60; 95% CI, 0.55-0.65).
Conclusions: Differences in use of health services are a more plausible explanation for higher mortality in rural than urban areas than differences in other factors.
Implications: Older people may be the 'grey canaries' of the health system and may thus provide an 'early warning system' to policy makers and governments.