RAND Study Finds Patients' Ratings of Their Medical Care Do Not Reflect the Technical Quality of Their Care
May 1, 2006
Published in: Annals of Internal Medicine, v. 144, no. 9, May 2, 2006, p. 665-672, W-149-W-157
Posted on RAND.org on December 31, 2005
BACKGROUND: Patient global ratings of care are commonly used to assess health care. However, the extent to which these assessments of care are related to the technical quality of care received is not well understood. OBJECTIVE: To investigate the relationship between patient-reported global ratings of health care and the quality of providers' communication and technical quality of care. DESIGN: Observational cohort study. SETTING: 2 managed care organizations. PATIENTS: Vulnerable older patients identified by brief interviews of a random sample of community-dwelling adults 65 years of age or older who received care in 2 managed care organizations during a 13-month period. MEASUREMENTS: Survey questions from the second stage of the Consumer Assessment of Healthcare Providers and Systems program were used to determine patients' global rating of health care and provider communication. A set of 236 quality indicators, defined by the Assessing Care of Vulnerable Elders project, were used to measure technical quality of care given for 22 clinical conditions; 207 quality indicators were evaluated by using data from chart abstraction or patient interview. RESULTS: Data on the global rating item, communication scale, and technical quality of care score were available for 236 vulnerable older patients. In a multivariate logistic regression model that included patient and clinical factors, better communication was associated with higher global ratings of health care. Technical quality of care was not significantly associated with the global rating of care. LIMITATIONS: Findings were limited to vulnerable elders who were enrolled in managed care organizations and may not be generalizable to other age groups or types of insurance coverage. CONCLUSIONS: Vulnerable elders' global ratings of care should not be used as a marker of technical quality of care. Assessments of quality of care should include both patient evaluations and independent assessments of technical quality.