Abstract P225: Creating a Platform for National Longitudinal Quality Assessment for Percutaneous Coronary Intervention (PCI)

Abstract

Background: Current quality and performance measures are generally limited to in-hospital care and outcomes due to a lack of longitudinal data. We examined the feasibility of linking detailed clinical data from the ACC NCDR CathPCI Registry to claims data sources from UnitedHealthCare (UHC) and Center for Medicare and Medicaid Services (CMS) to allow for longitudinal quality of care assessment for patients undergoing PCI. Methods: Using a two-step probabilistic matching strategy, we linked patients in ACC NCDR CathPCI with both UHC and CMS claims data. This matching strategy was used since these databases were limited datasets with encrypted patient identifiers. An ACC expert panel identified several measures of quality based on a literature review to test the feasibility of longitudinal care assessment. Results: Between July 1, 2006, and June 30, 2008, 471,641 CathPCI records and 38,018 UHC PCI records (patients age 18-64 years) and 1,553,481 CMS PCI records (patients age ≥ 65 years) were included in the matching strategy. Among linked sites, 81% and 89% of UHC and CMS records, respectively, were linked to CathPCI records. The final linked dataset included 16,485 unique records for the CathPCI-UHC matched dataset and 886,648 unique records for the CathPCI-CMS matched dataset. The Table displays initial CathPCI-UHC test performance results. Conclusions: For the first time, we have linked two separate claims data sources to ACC NCDR CathPCI to create a unique resource for assessing longitudinal quality of care following PCI. Future work is required to determine whether these data can used as feedback to providers to stimulate improved care quality for PCI patients. Table: Preliminary data for potential performance metrics from linked datasets. Potential Performance Metric NCDR CathPCI + UHC Patients with staged PCI (repeat PCI within 30 days on non-target lesion) 5.2% Patients with at least 1 lipid profile within 1 year of discharge 14% Patients on statin therapy at discharge with at least 1 liver function test within 1 year of discharge 16% Patients with at least 1 non-invasive stress testing with 6 months of discharge 29% Patients requiring repeat target lesion revascularization within 15 months of index PCI 12%