Variation in Inpatient Care

This work examines the influence of financial incentives and physician practice style on variation in inpatient care for both Medicare beneficiaries and the privately insured. I develop a model of physician behavior in the presence of multiple insurers that results in three testable predictions: 1) due to increased variation in hospital reimbursements, privately insured patients will exhibit greater variation in inpatient care than Medicare beneficiaries; 2) due to the effective employment of utilization restrictions by private insurers, Medicare beneficiaries will exhibit greater variation in inpatient care than the privately insured; 3) utilization restrictions will have a greater impact on conditions characterized by treatments that are highly dependent on physician discretion. This model of physician behavior motivates an empirical analysis comparing the magnitude of variations in hospital resource utilization for Medicare and privately insured patients. Results suggest that both Medicare recipients and the privately insured exhibit large variations in care and that these variations are present both between- and within-hospitals. For admissions with standardized treatment guidelines, there is little difference in the magnitude of variation between Medicare beneficiaries and the privately insured. However, for admissions characterized by competing treatments, the variation for Medicare patients exceeds that of the privately insured by as much as 6% for length of inpatient stay, by as much as 11% for the average number of inpatient procedures, and by as much as 7% for the probability of a surgical intervention. To quantify the health effects of these variations in treatment, I then conduct an analysis of the effect of additional treatment intensity on inpatient mortality for Medicare and privately insured patients hospitalized with a diagnosis of ischemic heart disease (IHD). Results of this analysis indicate that additional inpatient treatment intensity is not associated with a decline in mortality for Medicare patients, but is associated with a significant reduction in mortality among the privately insured. Increasing total inpatient charges for privately insured IHD patients by 10% (approximately $5,000, on average) results in a reduction in inpatient mortality of 12.5%.