Does health service use in a diabetes management program contribute to health disparities at a facility level? Optimizing resources with demographic predictors.

The objective of this study was to determine if demographic variation in the use of health service resources among type 2 diabetes patients contributes to health disparities. A prospective cohort design was used to analyze differences in health care utilization among 315 adults registered in primary care internal medicine clinics of an academic medical center. Patients were cared for by interdisciplinary teams of internal medicine residents, nurse practitioner students, and pharmacy students supervised by interdisciplinary faculty. A post hoc multivariate repeated measures analysis, using generalized estimating equation (GEE) statistical modeling, was used to determine if age, sex, race, ethnicity, marital status, primary language, and insurance predicted use of health care services (ie, primary care, acute care, emergency department [ED], hospitalization). Medicare/Medicaid-insured patients had an average of 2.49 primary care visits per month (P < .0001) and 75% more ED visits (P < .001) during the study than patients with other insurance types. ED visits for Hispanics grew by a factor of 3.3 compared to non-Hispanics (P < .0001). Females had 52% more hospitalizations than males (P < .05), and Hispanics had 44% fewer hospitalizations than non-Hispanics (P < .05). Analysis of selected health status indicators showed no significant differences for HbA1c, significantly greater likelihood of blood pressure >130/80 with every 5-year increase in age, and significantly greater likelihood of low-density lipoprotein >100 among Medicare/Medicaid-insured patients. Sociodemographic characteristics are predictive of health care services use and suggest that, although equally available to all participants, the use of health care resources vary at the facility level and are independent of diabetes health status outcomes.