BMC Public Health (Jun 2016)
Socioeconomic disparities in lung cancer mortality in Belgian men and women (2001-2011): does it matter who you live with?
Abstract
Abstract Background Ample studies have observed an adverse association between individual socioeconomic position (SEP) and lung cancer mortality. Moreover, the presence of a partner has shown to be a crucial determinant of health. Yet, few studies have assessed whether partner’s SEP affects health in addition to individual SEP. This paper will study whether own SEP (education), partner’s SEP (partner’s education) and own and partner’s SEP combined (housing conditions), are associated with lung cancer mortality in Belgium. Methods Data consist of the Belgian 2001 census linked to register data on cause-specific mortality for 2001–2011. The study population includes all married or cohabiting Belgian inhabitants aged 40–84 years. Age-standardized lung cancer mortality rates (direct standardization) and mortality rate ratios (Poisson regression) were computed for the different SEP groups. Results In men, we observed a clear inverse association between all SEP indicators (own and partner’s education, and housing conditions) and lung cancer mortality. Men benefit from having a higher educated partner in terms of lower lung cancer mortality rates. These observations hold for both middle-aged and older men. For women, the picture is less uniform. In middle-aged and older women, housing conditions is inversely associated with lung cancer mortality. As for partner’s education, for middle-aged women, the association is rather weak whereas for older women, there is no such association. Whereas the educational level of middle-aged women is inversely associated with lung cancer mortality, in older women this association disappears in the fully adjusted model. Conclusions Both men and women benefit from being in a relationship with a high-educated partner. It seems that for men, the educational level of their partner is of great importance while for women the housing conditions is more substantial. Both research and policy interventions should allow for the family level as well.
Keywords