Socioeconomic position and physical activity among women in Melbourne, Australia: does the use of different socioeconomic indicators matter?

This study aimed to identify which indicators of socioeconomic position (SEP) are independently associated with leisure-time physical activity among women. In 2005, women (n = 1166; 18-65 years) from Melbourne, Australia, reported their own (education, occupation, income), their partner's (education, occupation), their household (home ownership, ability to cope with income), and their neighbourhood (area-level) SEP, and leisure-time physical activity. Multinomial logistic regression examined associations between SEP indicators and leisure-time physical activity categorized as: none (no min/week; reference group), insufficient (1-149 min/week), and sufficient (≥150 min/week). In the fully adjusted model, lower education, lower partner's education (where applicable), and non-home ownership were independently associated with between 33% and 50% lower odds of sufficient physical activity, while lower income and lower area-level SEP were associated with 40% lower odds of insufficient physical activity. Understandings of socioeconomic inequalities in physical activity among women may be enhanced if a range of SEP indicators are used, particularly education, partner's education and home ownership.