Abstract

The risk of mortality in old age is associated with marital status and living arrangements. There is still little knowledge about this in Latin America. Our objectives are to examine the association between marital status, living arrangements and mortality of older adults (>60 years) in Chile, and to test whether this association varies when demographic, socioeconomic and health factors are included. We used data from the Social Protection Survey, and mortality data were linked to the Civil Registry. We estimate a series of Poisson regression models. Our results show a clear association between marriage and longevity, since even controlling for demographic, socioeconomic and health factors, we found that separated or divorced, widowed, and unmarried people showed higher relative mortality compared to married people (IRR1.24, IRR1.33, IRR1.35, respectively). Considering only living arrangements, the results show that living alone, alone with children, with children and other relatives or in other arrangements is associated with higher mortality (IRR1.22, IRR1.27, IRR1.35, IRR1.35, respectively) compared to those living with their partners and children. However, considering marital status and living arrangements together, we find that survival among older adults was strongly associated with marital status. Marital status continues to be a direct measure of living arrangements among older adults in Chile.