Validating the Clinical Outcomes Assessment Program risk model for percutaneous coronary intervention.

Area of Science:

• Cardiovascular Medicine
• Health Services Research
• Biostatistics

Background:

• A risk model for hospital mortality after percutaneous coronary intervention (PCI) was developed by the Clinical Outcomes Assessment Program (COAP).
• This COAP model was previously validated in northeastern US hospitals.
• Further validation was conducted using data from Providence Health System (PHS) hospitals in western US states.

Purpose of the Study:

• To validate the COAP risk model for hospital mortality after PCI using an independent dataset from PHS hospitals.
• To assess the model's discrimination and calibration in a different patient population and geographic region.

Main Methods:

• A total of 13,124 consecutive PCI procedures from 6 PHS hospitals (July 2001-June 2004) were analyzed.
• Model discrimination was tested using the c-index.
• Model calibration was assessed using the Hosmer-Lemeshow test, le Cessie-van Houwelingen-Copas test, and cumulative sum method.

Main Results:

• Patient demographics and overall mortality (1.4% in PHS vs. 1.6% in COAP) were similar between the COAP and PHS datasets.
• The COAP model demonstrated excellent discrimination with a c-index of 0.893 (95% CI: 0.859-0.928) when applied to PHS data.
• Good global goodness of fit was observed, indicating successful calibration of the model.

Conclusions:

• The COAP risk model for PCI-related hospital mortality was successfully validated using PHS data.
• Periodic updates of PCI risk models are essential to reflect advancements in technology and changes in patient profiles.
• Objective, clearly defined, and easily obtainable predictors are crucial for the transportability of PCI risk models.