Fritzell, Johan Rehnberg, Jennie Bacchus Hertzman, Jennie Blomgren, Jenni
Year of Publication:
LIS Working Paper Series 637
Objectives: We aimed to examine the cross-national and cross-temporal association between poverty and mortality, in particular differentiating the impact of absolute and relative poverty. Methods: We employed pooled cross-sectional time series analysis. Our measure of relative poverty was based upon the standard 60 percent of median income. The measure of absolute, or fixed, poverty we based upon the US poverty threshold. Our analyses were conducted on data for 30 countries between 1978 and 2010, a total of 149 data points. We separately studied infant, child and adult mortality. Results: Our findings highlight the importance of relative poverty for mortality. Especially for infant and child mortality we found that our estimates of fixed poverty is close to zero either in the crude models, or when adjusting for GDP. Conversely, the relative poverty estimates increased when adjusting for confounders. Our results seemed robust to a number of sensitivity tests. Conclusions: If we agree that risk of death is important, the public policy implication of our findings is that relative poverty, which has close associations to overall inequality, should be a major concern also among rich countries.