RT Journal Article SR Electronic T1 Individual and area-level socioeconomic associations with fast food purchasing JF Journal of Epidemiology and Community Health JO J Epidemiol Community Health FD BMJ Publishing Group Ltd SP 873 OP 880 DO 10.1136/jech.2009.099614 VO 65 IS 10 A1 Thornton, Lukar E A1 Bentley, Rebecca J A1 Kavanagh, Anne M YR 2011 UL http://jech.bmj.com/content/65/10/873.abstract AB Background It has been suggested that those with lower socioeconomic characteristics would be more likely to seek energy-dense food options such as fast food because of cheaper prices; however, to date the evidence has been inconsistent. This study examines both individual- and area-level socioeconomic characteristics and their independent associations with chain-brand fast food purchasing.Methods Data from the 2003 Victorian Lifestyle and Neighbourhood Environments Study (VicLANES); a multilevel study of 2547 adults from 49 small-areas in Melbourne, Australia, were used. Multilevel multinomial models adjusted for confounders were used to assess associations between individual socioeconomic position (education, occupation and income) and area socioeconomic characteristics in relation to fast food purchasing from five major fast food chains with outcome categories: never, at least monthly and at least weekly. The study finally assessed whether any potential area-level associations were mediated by fast food access.Results Increased fast food purchasing was independently associated with lower education, being a blue-collar employee and decreased household income. Results for area-level disadvantage were marginally insignificant after adjustment for individual-level characteristics, although they were suggestive that living in an area with greater levels of disadvantage increased an individual's odds of more frequent fast food purchasing. This effect was further attenuated when measures of fast food restaurant access were included in the models.Conclusion Independent effects of lower individual-level socioeconomic characteristics and more frequent fast food purchasing for home consumption are demonstrated. Although evidence was suggestive of an independent association with area-level disadvantage this did not reach statistical significance.