Educational disparities in mortality among patients with type 2 diabetes in the Netherlands (ZODIAC-23)

Background: Relative mortality differences between educational level in mortality have been reported among diabetic as well as among non-diabetic subjects in Europe, but data on absolute differences are lacking. We studied the effect of educational disparities on mortality in a Dutch prospective cohort of type 2 diabetes mellitus (T2DM) patients. Methods: This study was part of the ZODIAC study, a prospective observational study of patients with T2DM. Data on educational level were first collected on 19 May 1998, and from this date on, 858 patients were included in 1998; educational level was known for 656 patients. Vital status was assessed in 2009. The relationship between mortality and educational level was studied using a Cox proportional hazard model, the relative index of inequality (RII), slope index of inequality (SII) and the population attributable risk (PAR). Educational level was divided into four categories; the highest educational level was used as reference. Results: After a median follow-up time of 9.7 years, 365 out of 858 patients had died. The hazard ratio of primary education for total mortality was 3.02 (95% CI 1.44-6.34). The RII was 2.85 (95% CI 1.21-6.67), the absolute difference in the risk for mortality (SII) was 384 deaths (95% CI 49-719) per 10,000 follow-up years. PAR for patients with the lowest level of education was 51.4%. Conclusions: A low educational level had a higher impact on mortality than having a macrovascular complication. Given the substantial differences in mortality between educational levels in T2DM, more understanding of underlying (modifiable) mechanisms is necessary.