PT - JOURNAL ARTICLE AU - Gavin Turrell AU - Anne Kavanagh AU - Glenn Draper AU - S V Subramanian TI - Do places affect the probability of death in Australia? A multilevel study of area-level disadvantage, individual-level socioeconomic position and all-cause mortality, 1998–2000 AID - 10.1136/jech.2006.046094 DP - 2007 Jan 01 TA - Journal of Epidemiology and Community Health PG - 13--19 VI - 61 IP - 1 4099 - http://jech.bmj.com/content/61/1/13.short 4100 - http://jech.bmj.com/content/61/1/13.full SO - J Epidemiol Community Health2007 Jan 01; 61 AB - Background: In Australia, studies finding an association between area-level socioeconomic disadvantage and mortality are often based on aggregate-ecological designs which confound area-level and individual-level sources of socioeconomic variation. Area-level socioeconomic differences in mortality therefore may be an artefact of varying population compositions and not the characteristics of areas as such. Objective: To examine the associations between area-level disadvantage and all-cause mortality before and after adjustment for within-area variation in individual-level socioeconomic position (SEP) using unlinked census and mortality-register data in a multilevel context. Setting, participants and design: The study covers the total Australian continent for the period 1998–2000 and is based on decedents aged 25–64 years (n = 43 257). The socioeconomic characteristics of statistical local areas (SLA, n = 1317) were measured using an index of relative socioeconomic disadvantage, and individual-level SEP was measured by occupation. Results: Living in a disadvantaged SLA was associated with higher all-cause mortality after adjustment for within-SLA variation in occupation. Death rates were highest for blue-collar workers and lowest among white-collar employees. Cross-level interactions showed no convincing evidence that SLA disadvantage modified the extent of inequality in mortality between the occupation groups. Conclusions: Multilevel analysis can be used to examine area variation in mortality using unlinked census and mortality data, therefore making it less necessary to use aggregate-ecological designs. In Australia, area-level and individual-level socioeconomic factors make an independent contribution to the probability of premature mortality. Policies and interventions to improve population health and reduce mortality inequalities should focus on places as well as people.