Racial/Ethnic Differences in CAHPS Ratings and Reports
AHRQ / PHS
05/01/01 - 04/30/03
Robert J. Weech-Maldonado, Department of Health Policy and Administration
Consumer assessments of health care are increasingly being used as an indicator of the quality of care provided by health plans and providers. Consumer ratings of health care can provide important information about how well health plans and clinicians meet the needs of the people that they serve. The extent to which consumer assessments of health care vary by race/ethnicity is of significant importance in evaluation of health plan performance. Differences in evaluations between subgroups may be due to response bias or to differences in the quality of care received. This study examined the racial and ethnic differences in consumer health plan assessments for adults and children enrolled in managed care plans. Three research questions were investigated in this study: 1) Do the psychometric properties of the consumer reports and ratings of health plans differ by race/ethnicity?; 2) Do consumer reports and ratings of health plans vary by race/ethnicity?; and 3) Do managed care organizational characteristics account for the observed racial/ethnic differences in reports and ratings of health care? The data analyzed were from the National CAHPS Benchmarking Database (NCBD) and consisted of 169,136 commercial adults, 28,420 Medicaid adults, 42,879 commercial children (<18 years of age), and 14,016 Medicaid children (<18 years of age). The sample consisted of consumers receiving care from 599 health plans in 39 states, DC, Guam, and Puerto Rico for the period of 1998-1999. In addition, the 1999 NCQA Quality Compass database was used to identify health plan organizational characteristics. The dependent variables were CAHPS 2.0 ratings and reports of care, and the independent variables included race/ethnicity (White, African American, Asian, Native American, and Hispanic subgroups), case mix variables (gender, age, education, and health status), and health plan characteristics. Equivalence of CAHPS items and composites across race/ethnic subgroups were compared using reliability estimates, confirmatory factor analysis, and item response theory analysis. Regression methods were used to evaluate the associations of race/ethnicity with CAHPS reports and ratings of care and to examine whether managed care organizational characteristics account for the observed racial/ethnic differences in reports and ratings of health care.