Prior research identified variations in care experiences across Medicare health plans (Medicare Advantage [MA]), but the relative amount of variation in MA and traditional fee-for-service (FFS) Medicare is unknown.
Compare variation and correlations of beneficiary reports of care experiences across geographic areas in MA and FFS.
Using the 2001 to 2004 Medicare CAHPS surveys, we analyzed 14 measures of care experiences and preventive services reported by 433,092 MA beneficiaries (82% response rate) and 244,731 in FFS (69% response rate). We estimated hierarchical regression models with random effects for state, hospital referral region, and plan, adjusting for respondent characteristics. We examined the relative variation in FFS and MA scores across areas and among individual MA plans, the correlation between FFS and MA scores across areas, and variability relative to average MA-FFS differences in scores.
Although MA and FFS scores are highly correlated, variation is greater in MA than FFS across states and local areas for almost all measures. MA plan variation within areas accounts for 25% to 50% of explained MA variation. MA-FFS differences are smaller than the standard deviations of differences across areas for 10 of 14 measures.
Relative performance between MA and FFS may differ across areas and locally between individual plans and FFS. Quality improvement initiatives should address local system factors that affect both MA and FFS, and identify organizational factors that make some MA plans more successful in improving quality.
From the *Yale School of Public Health, Yale School of Medicine, New Haven, CT; †Department of Health Care Policy, Harvard Medical School, Boston, MA; and ‡Division of General Medicine and Primary Care, Beth Israel Deaconess Medical Center, Boston, MA.
Supported by a contract (99–200–700–47) from the Centers for Medicare & Medicaid Services (CMS) to the RAND Corporation. CMS funded the collection of CAHPS data, approved the manuscript topic, and reviewed the manuscript before submission for nondisclosure of personally identifiable information.
Reprints: Patricia S. Keenan, PhD, Yale School of Public Health, Yale School of Medicine; 60 College St #300C New Haven, CT 06520. E-mail: firstname.lastname@example.org.