Widening educational inequalities in type 2 diabetes risk and mediators across successive generations: a prospective cohort study

Background Educational attainment is a pivotal socioeconomic determinant of metabolic health ①. However, whether education-related inequalities extend from T2DM incidence to disease progression, and how they evolve across successive birth cohorts, remain unclear. Purpose To investigate intergenerational differences in educational inequalities regarding the incidence and progression of T2DM, and to identify the driving mediators. Methods This study included participants from the UK Biobank who were free of diabetes at baseline. Participants were stratified into three birth cohorts (pre-1949, 1950–1959, and post-1960) to investigate intergenerational differences. We employed Fine-Gray proportional subdistribution hazards models and multi-state competing risks models to evaluate the impact of education on the risk of incident T2DM and its subsequent progression to cardiovascular disease (CVD) and mortality. The Relative Index of Inequality (RII) was calculated to quantify the magnitude of disparities. A sensitivity analysis was performed by redefining the cohort entry at age 30. Mediation analysis was conducted to estimate the proportion of the association mediated by socioeconomic, behavioural, and metabolic factors. Results During a median follow-up of 16.5 years, 19,931 incident cases of T2DM were documented among the 460,797 included participants. The proportion of primary education or less dropped from 26.7% in the pre-1949 birth cohort to 11.7% and 5.1% in the 1950–1959 and post-1960 cohorts. Educational level was inversely associated with the risk of incident T2DM across all birth cohorts. Compared with college education, primary education or less was associated with increased risks of incident T2DM (HR: 1.92; 95% CI: 1.85–1.99), progression to CVD (HR: 1.33; 95% CI: 1.30–1.37), and mortality (HR: 1.53; 95% CI: 1.49–1.57), and these disparities widened across successive birth cohorts. Overall, the adjusted RII was 2.36 (95% CI: 2.24–2.49), showing an increasing trend from 2.37 (2.25–2.50) in the pre-1949 cohort to 2.53 (2.29–2.79) in the 1950–1959 cohort, and 2.53 (2.18–2.93) in the post-1960 cohort. Based on the reconstructed cohort with follow-up starting at age 30, the risk of incident T2DM increased across successive birth cohorts. Furthermore, a widening of education-related inequalities was consistently observed across these cohorts. Substantial attenuation of the HRs and RII was observed after adjusting for mediators, either individually or simultaneously. Socioeconomic, metabolic, and behavioural factors mediated a substantial proportion of the educational inequality, accounting for 42.2% (39.2%–45.2%), 42.6% (38.2%–47.3%), and 20.5% (18.9%–22.2%), respectively. Conclusion Education-related inequalities extend from T2DM incidence to subsequent progression and have widened across successive birth cohorts. Socioeconomic, behavioural, and metabolic factors mediated a substantial proportion of these disparities.