Health plan report cards and managed care enrollment.

This dissertation examines the relationship between health plan ratings and 1995 managed care (MC) enrollment decisions for employees and retirees of a Fortune 100 firm. The study takes advantage of an initiative at the firm to collect data from the Health Plan Employer Data and Information Set (HEDIS), and to construct, and disseminate health plan ratings to employees and retirees based on the data. Conditional logit, nested logit, and market share regression models are estimated for the probability of plan enrollment as a function of five health plan ratings (physician quality, medical treatment, surgical care, preventive care, and employee satisfaction), out-of-pocket price, health plan model type. and the ratio of physicians to plan enrollees. Models are estimated by coverage category for eight samples, including union and non-union employees, Medicare and non-Medicare retirees, new employees, and employees switching to MC plans between 1994 and 1995. The findings suggest that the relationship between managed care enrollment and plan ratings is mixed. The most robust findings are that enrollment is correlated positively with the rating labeled by the firm as 'preventive care', and correlated inversely with the rating labeled 'enrollee satisfaction.' The preventive care estimates suggest that employees and retirees are between 23 and 90 percent more likely to enroll in plans rated superior on preventive care, relative to otherwise identical plans not rated superior. The counterintuitive finding for the employee satisfaction rating may be due to positive correlations between the measures that comprise the satisfaction rating (e.g., appointment waiting times) and measures that are unobserved by the researcher, but valued by consumers (e.g., physician popularity). The findings also suggest that health plan enrollment is inversely related to out-of-pocket price, consistent with what other studies have found. Estimates suggest that employees and retirees are between 2 and 14 percent less likely to enroll in health plans with monthly premiums that are five dollars more expensive than otherwise identical plans. The mixed results for the report card ratings, coupled with the consistent results for out-of-pocket price, suggest that price incentives may be the best way to increase managed care enrollment in insured populations.Ph.D.EconomicsHealth and Environmental SciencesHealth care managementPublic healthSocial SciencesUniversity of Michigan, Horace H. Rackham School of Graduate Studieshttp://deepblue.lib.umich.edu/bitstream/2027.42/131088/2/9825337.pd