Background Mortality amenable to healthcare interventions has increasingly been used as an indirect indicator of the effect of healthcare on health inequalities. Studies have consistently shown socioeconomic differences in amenable mortality, but evidence on the joint effects of multiple socioeconomic and demographic factors is limited. We examined whether income and living arrangements have an independent effect on amenable mortality taking into account other dimensions of social position.

Methods The longitudinal and yearly updated individual level data were derived from different administrative registers and obtained from Statistics Finland. The data set includes an 11% random sample of all individuals aged 25–74 years at the end of 1999 and an 80% oversample of deaths in the follow-up period between 2000 and 2007. We used Cox proportional hazard regression with appropriate weights.

Results We found twofold to threefold differences in amenable mortality between the top and bottom income quintiles. These differences were found to be largely attenuated by economic activity and living arrangements. We also found differences in amenable mortality by living arrangements suggesting that those living alone, as well as lone parents and those cohabiting have higher amenable mortality. These differences were largely independent of our indicators of socioeconomic position and economic activity.

Conclusions While our results give indirect support to the hypothesis that income differences in amenable mortality may be at least partially due to barriers in access to care, the large independent effects of living arrangements on amenable mortality suggest that seeking care may also have an impact.