FCR-09
Hospital-Level Percutaneous Coronary Intervention Performance with Simulated Risk Avoidance Strategies
Presenter
Ashwin Nathan, M.D., Hospital of the University of Pennsylvania, Philadelphia, PA
Ashwin Nathan, M.D.1, Pratik Manandhar, MS2, Daniel Wojdyla, PhD2, Sameed Khatana, MD, MPH3, Elias Dayoub, MD, MPP3, Lauren Eberly, MD, MPH1, Paul Fiorilli, M.D.4, Stephen W Waldo, MD5, Robert W. Yeh, M.D., FSCAI6, Sunil V. Rao, MD, FSCAI7, Adam James Nelson, MBBS8, Alexander Craig Fanaroff, MD, MHS9, Peter Groeneveld, MD, MS10, Tracy Yu-Ping Wang, M.D.2 and Jay S. Giri, MD, FSCAI1, (1)Hospital of the University of Pennsylvania, Philadelphia, PA, (2)Duke Clinical Research Institute, Durham, NC, (3)Hospital of the University of Pennsylvania, Durham, NC, (4)Hospital of the University of Pennsylvania, Haddonfield, NJ, (5)University of Colorado Anschutz Medical Campus, Denver, CO, (6)Beth Israel Deaconess Medical Center, Boston, MA, (7)NYU Langone Health, New York, NY, (8)Royal Adelaide Hospital, Durham, NC, (9)Hospital of the University of Pennsylvania, Gladwyne, PA, (10)University of Pennsylvania, Philadelphia, PA
Keywords: Cath Lab Administration, Cath Lab Leadership Boot Camp, Coronary and Quality
Background: Percutaneous coronary intervention (PCI) operators may avoid high-risk cases so as to limit the effect of these cases on performance metrics, but the effect of such a strategy on individual hospital performance metrics is uncertain. We simulated the effects of systematic risk avoidance on risk-adjusted post-PCI mortality.
Methods: We identified all adult patients who underwent PCI at a participating hospital in the National Cardiovascular Data Registry CathPCI registry between 01/01/17 and 12/31/17. Risk-adjusted mortality rates were calculated for each hospital. We then simulated a systematic risk-avoidance strategy of eliminating the highest predicted risk cases (top 10%) for each hospital, then recalculating risk-adjusted mortality rates.
Results: 1,565 hospitals were included in the analysis. 883 (56.4%) hospitals reduced their risk-adjusted mortality rate, but 610 (39.0%) hospitals increased their risk-adjusted mortality rate (Figure). Hospitals changed their risk-adjusted mortality rate by -0.14% on average with this strategy. There were no significant differences in the patient or procedural characteristics among hospitals which improved compared with those that worsened.
Conclusions: For any individual hospital seeking to improve their performance, there is no guarantee that practicing systematic risk-avoidance will improve the measured quality of a PCI program, and may in fact worsen it. These data support offering PCI to appropriate candidates regardless of calculated pre-procedural risk.
