181323 Racial and ethnic disparities in Medicaid expenditures for hypertension, heart disease, stroke, congestive heart failure, diabetes, and cancer: Evidence from four states – United States, 1999-2001

Sunday, October 26, 2008

Isaac Nwaise, MA , Division for Heart Disease and Stroke Prevention, Centers for Disease Control and Prevention, Atlanta, GA
Susan Haber, ScD , Division for Health Services and Social Policy Research, RTI International, Waltham, MA
Kumiko Imai, PhD , Division of Diabetes Translation, Centers for Disease Control and Prevention, Atlanta, GA
Florence K.L. Tangka, PhD , Epidemiology and Applied Research Branch, Centers for Disease Control and Prevention, Chamblee, GA
Jyoti Aggarwal, MHS , Division for Health Services and Social Policy Research, RTI International, Waltham, MA
Diane Orenstein, PhD , Division for Heart Disease and Stroke Prevention, Centers for Disease Control and Prevention, Atlanta, GA
This study examines whether there are racial/ethnic disparities in expenditures for treating chronic illnesses within an insured population of Medicaid enrollees and whether disparities are explained by differences in expenditures for specific types of service. We use Medicaid Analytic Extract files for 1999-2001 to develop estimates by race/ethnicity (white, black, Hispanic) of the prevalences of treatment and per capita Medicaid expenditures attributable to hypertension, heart disease, stroke, congestive heart failure, diabetes, and cancer in Illinois (n=2,295,098), Indiana (n=995,320), Kansas (n= 377,086), and Louisiana (n=1,124,635). Multivariate regression is used to estimate annual Medicaid expenditures (total and by type of service – inpatient, outpatient hospital including ER, office visits, prescription drugs, long term care, and other services) associated with a condition after controlling for age, gender, and comorbid conditions.

The prevalence of all six conditions is significantly higher among whites compared to blacks and Hispanics. For example, the prevalence of treated hypertension is 18-21% among whites in the study states, 13-15% among blacks and 5-6% among Hispanics. Expenditures for these chronic illnesses are generally lower for whites than for racial/ethnic minorities. In Illinois, whites have lower per capita expenditures attributed to hypertension than blacks (34% lower, p<.0001) and Hispanics (43% lower, p<.0001). Differences in hypertension treatment expenditures were smaller in the other states (10-17% lower for whites compared to blacks and 20-25% lower compared to Hispanics), but still statistically significant. Whites have lower Medicaid expenditures for most of the services examined, although there is variation across illnesses and states. For example, inpatient expenditures attributed to hypertension range from 33% to almost 60% lower for whites compared to blacks and 50-60% lower compared to Hispanics. However, in a number of cases whites have significantly higher prescription drug expenditures. Prescription drug expenditures attributable to hypertension are about 50% higher for whites compared to blacks in Illinois and 30% higher in Indiana and Kansas. Differences between whites and Hispanics in prescription drug expenditures for hypertension are smaller.

Although racial/ethnic minorities are at higher risk for a number of the study conditions, our analyses show a lower prevalence of treated chronic illnesses among racial/ethnic minorities in the Medicaid population. The lower prevalence suggests that racial/ethnic minorities are sicker and have more extensive service needs when they receive treatment. This greater severity could explain their higher per capita treatment expenditures compared to whites. Lower prescription drug utilization may indicate chronic illnesses are less well controlled among racial/ethnic minorities.

Learning Objectives:
1. Identify racial and ethnic variation within the Medicaid population in prevalence of treatment and expenditures for treating chronic illnesses. 2. Analyze sources of racial and ethnic variation in treatment expenditures for chronic illnesses. 3. Apply econometric methods for estimating cost of illness using claims data.

Keywords: Chronic Diseases, Health Disparities

Presenting author's disclosure statement:

Qualified on the content I am responsible for because: I contributed to conceptualizing, conducting, and interpreting the findings of the analyses being presented
Any relevant financial relationships? No

I agree to comply with the American Public Health Association Conflict of Interest and Commercial Support Guidelines, and to disclose to the participants any off-label or experimental uses of a commercial product or service discussed in my presentation.