Primary Submission Category: Health equity
Differential returns to education on diabetes risk by race-by-sex subgroup: results from the US Health and Retirement Study
Authors:Â Michelle A DeVost Anusha M Vable
Presenting Author: Michelle A DeVost*
Introduction
Education is an early-life exposure that influences trajectories of health across the life course. Higher educational attainment is associated with lower risk of diabetes in later life, but how this relationship varies by the interaction of race/ethnicity and sex remains unclear. We investigated whether the association between education and diabetes risk varies by race-by-sex subgroups.
Methods
We used data from the Health and Retirement Study, a population-based cohort study of US adults over 50 from 1992 to 2016 (n=32,294). With Cox proportional hazards models, we estimated the risk of diabetes by educational attainment in each subgroup. The base model included years of education, race-by-sex, birth year, birthplace, and parents’ education. To test for differential relationships by demographic group, we included interactions between education and race-by-sex subgroup (ref=White men).
Results
In the base model, the risk of incident diabetes was 3% lower for each additional year of education (HR=0.97, 95%CI: 0.96-0.97). In the interaction model, at 12 years of education, White women had lower risk than White men, while Black and Latinx men and women had higher risk than White men. For each additional year of education compared to White men, White women had lower risk of diabetes (education * subgroup interaction HR: 0.97, 95% CI: 0.95-0.99), and Latinx men (interaction HR: 1.03, 95% CI: 1.01-1.05) and Black men had higher risk of diabetes (interaction HR: 1.06, 95% CI: 1.04-1.09). Subgroup-specific estimates of the effect of education appeared protective for every race-by-sex subgroup besides Black men.
Conclusions
Literature supports the finding that Black men experience worse health outcomes despite educational attainment, which appears protective in other populations. If causal, this suggests that education may reduce social inequities in diabetes for Black and Latinx women but increase inequities for Black men.