Disparities in Diabetes by Education and Race/Ethnicity in the U.S., 1973-2012.

Abstract

Diabetes mellitus incidence has more than doubled in the U.S. over the past 2 decades. Not all sectors of the population have experienced the increase proportionally. The goal of this study was to determine if disparities in diabetes by education and race/ethnicity have increased over time, and if there are differences by gender and birth cohort. Repeated cross-sectional data were used from the 1973-2012 National Health Interview Survey of adults aged 25-84 years. Logistic regression models were run and predicted probabilities were calculated to determine if disparities in self-reported diabetes by education and race/ethnicity changed over time, by gender and birth cohort (birth before 1946, 1946-1970, 1971 or after). Analyses were conducted in 2014-2015. Relationships between education or race/ethnicity and diabetes were modified by time for people born before 1971, with stronger effect modification for women than men. Inequalities in diabetes prevalence grew over time, although the magnitude of disparities was smaller for the 1946-1970 cohort. For example, in 2005-2012, the gap in diabetes prevalence for women with the highest and lowest levels of education was 12.7% for pre-1946 versus 7.9% for 1946-1970. Similar trends were seen for differences between non-Hispanic whites and non-Hispanic blacks or Hispanics. Results were inconclusive for the youngest cohort. Diabetes disparities are evident. Smaller differences in later cohorts may indicate that large structural changes in society (e.g., Civil Rights movement, increased educational and economic opportunities) have benefited later generations.