132 Annual Meeting Logo - Go to APHA Meeting Page  
APHA Logo - Go to APHA Home Page

Disparities in Medicaid use between American Indian and Whites: An application of generalized linear model and risk adjustment in health disparity study

Chi Kao, PhD1, Carol C. Korenbrot, PhD1, Sabrina T. Wong, RN, PhD2, and James Crouch, MPH3. (1) Institute for Health Policy Studies, University of California San Francisco, Box 0936, San Francisco, CA 94143-0936, (415) 514-2952, chikao@itsa.ucsf.edu, (2) Medical Effectiveness Research Center for Diverse Populations, University of California San Francisco, Box 0856, 3333 California Street, Suite 335, San Francisco, CA 94143, (3) Executive Director, California Rural Indian Health Board, 1451 River Park Drive, Suite 220, Sacramento, CA 95815

Background: Disparities in Medicaid service use and payments for American Indians and Alaska Natives (AI/AN) entitled to care through the Indian Health Service (IHS) raise issues of whether AI/AN are receiving the Medicaid services they need. Objective: We examined disparities in health resource use adjusting for health risks using generalized linear models with the most appropriate distribution for each outcome variable. Method: Using Medicaid paid claims from California in 1996, when Medicaid managed care had made little penetration of rural areas of California, we compared service use and costs of 7,910 AI/AN to a matched group of 15,075 Whites in the same counties. Linear rank statistics using Wilcoxon scores were used to test differences of location and scale of resource use between two groups. Generalized linear models were then constructed in order to test whether the differences in resource use were significant after adjustment for health risk. Risk was measured using Adjusted Clinical Groups (ACG). ACG weights generated using both California data and benchmarks (J.P. Weiner, C. Abrams, et al.) were both examined for risk adjustment. Benchmark weights were rescaled to using the White reference group. Results: Benchmark weights offered more reliable results. AI/AN have lower service use than Whites regardless of whether or not the service is also available through the IHS, and lower total payments regardless of whether or not the dominant source of ambulatory care is IHS funded. Lower Medicaid service use and costs for AI/AN are consistent with barriers in receiving Medicaid-paid services.

Learning Objectives:

Keywords: Health Care Utilization, Health Disparities

Presenting author's disclosure statement:
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.

Biostatistical Methods in Public Health Research and Practice

The 132nd Annual Meeting (November 6-10, 2004) of APHA