Educational gain in cause-specific mortality: accounting for confounders

For many causes of death a negative educational gradient has been found. This association may be partly explained by confounding factors that affect both education attainment and mortality. We correct the cause-specific educational gradient for observed individual background and unobserved family factors, using an innovative method based on months lost due to a specific cause of death re-weighted by the probability to attain a higher education level. We use men with brothers in the Swedish Military Conscription Registry (1951-1983), linked to administrative Swedish registers. These data, comprising 700,000 men, allow us to distinguish five education levels and many causes of death. The empirical results reveal that improving education from primary to higher education would lead to 20 months longer survival between 18 and 63. The reduction in death due to external causes when improving education is attributable to most of these gains. Ignoring confounding would underestimate the educational gains, especially for the low educated. Implied by our results is that if 50,000 men from the 1951 cohort had had the 1983 education distribution they would have saved 22% of the person-years between ages 18 and 63 that were lost to death.