Article Title

Disparities in Glycemic Control Among Hispanic Adults With Diabetes

Publication Date



disparities, glycemic control


Background/Aims: Poor glycemic control is associated with increased morbidity and mortality for adults with diabetes mellitus (DM). Little research has examined disparities in glycemic control among Hispanics with DM compared to whites. The objective of this work was to determine: 1) whether disparities in glycemic control exist among Hispanics versus whites; and 2) whether demographics, socioeconomic status, disease characteristics, health care utilization (primary care, specialty care, care management services) and treatment characteristics (oral hypoglycemic medications, insulin use) explain differences in glycemic control.

Methods: Using an observational study design, we studied 29,825 adults on the Kaiser Permanente Northwest DM registry as of January 1, 2013, with a valid HbA1c test during calendar year 2013. Good glycemic control was defined as HbA1c 30 (vs. body mass index < 30), Charlson comorbidity score (continuous), primary care utilization in 2013 (1+ visits vs. none), specialty care utilization (1+ visits vs. none), use of DM care management services (1+ services vs. none), use of oral hypoglycemic medications (1+ medications fills vs. none) and insulin use (any insulin use vs. none). Seven logistic models were constructed: model 1 (race/ethnicity), model 2 (model 1 + demographics), model 3 (model 2 + SES), model 4 (model 3 + disease characteristics), model 5 (model 4 + health care utilization) and model 6 (model 5 + treatment characteristics).

Results: Hispanics were less likely to have good glycemic control in unadjusted models (odds ratio: 0.56, 95% confidence interval [CI]: 0.51–0.61; P<0.0001). This point estimate remained consisted across all logistic models examined, even after adjusting for covariate measures (odds ratio: 0.69, 95% CI: 0.62–0.77; P<0.0001).

Discussion: Our findings suggest that disparities in glycemic control among Hispanics compared to whites remain even after adjusting for critical covariate measures. More work is needed to understand whether lifestyle choices and other factors explain differences and whether targeted interventions can reduce these disparities.