CMS Reimbursement Reform and the Incidence of Hospital-Acquired Pulmonary Embolism or Deep Vein Thrombosis

Abstract

BACKGROUND: In October 2008, the Centers for Medicare & Medicaid Services (CMS) stopped reimbursing hospitals for the marginal cost of treating certain preventable hospital-acquired conditions. OBJECTIVE: This study evaluates whether CMS’s refusal to pay for hospital-acquired pulmonary embolism (PE) or deep vein thrombosis (DVT) resulted in a lower incidence of these conditions. DESIGN: We employ difference-in-differences modeling using 2007–2009 data from the Nationwide Inpatient Sample, an all-payer database of inpatient discharges in the U.S. Discharges between 1 January 2007 and 30 September 2008 were considered “before payment reform;” discharges between 1 October 2008 and 31 December 2009 were considered “after payment reform.” Hierarchical regression models were fit to account for clustering of observations within hospitals. PARTICIPANTS: The “before payment reform” and “after payment reform” incidences of PE or DVT among 65–69-year-old Medicare recipients were compared with three different control groups of: a) 60–64-year-old non-Medicare patients; b) 65–69-year-old non-Medicare patients; and c) 65–69-year-old privately insured patients. Hospital reimbursements for the control groups were not affected by payment reform. INTERVENTION: CMS payment reform for hospital-based reimbursement of patients with hip and knee replacement surgeries. MAIN MEASURES: The outcome was the incidence proportion of hip and knee replacement surgery admissions that developed pulmonary embolism or deep vein thrombosis. KEY RESULTS: At baseline, pulmonary embolism or deep vein thrombosis were present in 0.81 % of all hip or knee replacement surgeries for Medicare patients aged 65–69 years old. CMS payment reform resulted in a 35 % lower incidence of hospital-acquired pulmonary embolism or deep vein thrombosis in these patients (p = 0.015). Results were robust to sensitivity analyses. CONCLUSION: CMS’s refusal to pay for hospital-acquired conditions resulted in a lower incidence of hospital-acquired pulmonary embolism or deep vein thrombosis after hip or knee replacement surgery. Payment reform had the desired direction of effect.