Health literacy in a population-based sample of Australian women: a cross-sectional profile of the Geelong Osteoporosis Study

BACKGROUND: The term health literacy refers to the abilities and resources required to find, understand and use health information in managing health. This definition is reflected in the recent development of multidimensional health literacy tools that measure multiple facets of health literacy. The aim of this study was to determine the health literacy profile of a randomly selected, population-based sample of Australian women using a multidimensional tool, the Health Literacy Questionnaire (HLQ). A second aim was to investigate associations between independent HLQ scales, sociodemographic characteristics and lifestyle and anthropometric risk factors for chronic disease. METHODS: We surveyed women involved in the Geelong Osteoporosis Study (GOS), a longitudinal, population-based study. We included demographic data, lifestyle information and anthropometric measures as well as the HLQ. The HLQ has 44 items, scored on either 4- or 5-point scales, within nine conceptually distinct scales. Means for each scale were calculated, and HLQ scales were regressed on educational level and socioeconomic status. Risk factors for chronic disease were investigated using analysis of variance (ANOVA) and calculation of effect sizes. RESULTS: Higher mean scores were seen for the scales 'Feeling understood and supported by healthcare professionals' (mean 3.20, ± SD 0.52) and 'Understanding health information well enough to know what to do' (mean 4.28, ±SD 0.54), and lower mean scores were seen for 'Appraisal of health information' (mean 2.81, ±SD 0.48) and 'Navigating the healthcare system' (mean 4.09, ± SD 0.57). Associations were also seen between lower HLQ scores and poor health behaviours including smoking and being more sedentary, in addition to greater body mass index and waist circumference. Positive gradients were seen between several HLQ scales and education level, as well as SES. For some HLQ scales, these associations were non-linear. CONCLUSIONS: The profile of this populat