The public health hazards of risk avoidance associated with public reporting of risk-adjusted outcomes in coronary intervention

Abstract

Public reporting of risk-adjusted outcomes for percutaneous coronary intervention (PCI) procedures has been mandated in New York State for more than a decade. During that time there has been a significant decline in the unadjusted mortality after such procedures. Massachusetts joined New York in 2003 as only the second state to require case level reporting of every coronary interventional procedure performed. In this review, we explore the differences in the populations reported by the 2 states and consider possible risks of public reporting of clinical outcomes after PCI procedures, including the risk of increasing conservatism in the treatment of the sickest patients. We offer a conceptual framework to understand the potential risk-averse behavior of interventional cardiologists subject to public reporting, and offer several proposals to counteract this potential deleterious effect of reporting programs.

Figure 1. Rates of PCI for Cardiogenic Shock in New York and Massachusetts in 2003

Revascularization rates for New York and Massachusetts for cardiogenic shock in 2003 (per million of population). Massachusetts physicians appear to be three times as likely to undertake PCI for the treatment of cardiogenic shock as their counterparts in New York State.

Figure 2. Proportion of Patients Undergoing PCI in Massachusetts for Cardiogenic Shock 2003–2005

Decline in proportion of patients undergoing PCI in Massachusetts for cardiogenic shock: 2003 through 2005. The proportion of PCI cases that were performed in the setting of cardiogenic shock has declined nearly 47% in the three years since public reporting of outcomes was implemented.

Figure 3. Map of PCI risk versus clinical benefit

The vertical axis denotes the risk of the procedure represented as the likelihood of survival to hospital discharge. The horizontal axis denotes the patient benefit represented as the incremental health benefit of having the procedure performed.

Figure 4. Potential for “risk avoidance creep”

The map of PCI risk versus clinical benefit is shown (see figure 3) with illustrative example scenarios shown. Green ovals indicate scenarios where clinical benefit is high, purple indicate intermediate risk and red ovals indicate scenarios in which incremental clinical benefit is negligible. The red dashed arrow indicates the “risk avoidance creep” toward lower risk cases in the face of public reporting.