New York Risk Model and Simplified Risk Score for In-Hospital/30-Day Mortality for Percutaneous Coronary...

Area of Science:

• Cardiology
• Health Services Research
• Medical Informatics

Background:

• Existing risk models for percutaneous coronary interventions (PCIs) require updates due to evolving patient populations and procedures.
• There is a lack of contemporary risk models for predicting in-hospital or 30-day mortality following PCI.

Purpose of the Study:

• To develop and validate an updated risk model and a simplified risk score for predicting in-hospital/30-day mortality after PCI.
• To address the need for accurate risk prediction using current data in PCI procedures.

Main Methods:

• Utilized data from New York's Percutaneous Coronary Intervention Reporting System (2019).
• Developed a logistic regression model and a simplified risk score using 12 risk factors and 27 categories.
• Validated both models using prior year New York PCI data.

Main Results:

• The developed models demonstrated excellent discrimination (0.894-0.896) in both development and validation datasets.
• The full logistic model showed superior calibration, especially for high-risk patients, compared to the simplified score.
• Both models accurately predict mortality risk following PCI procedures.

Conclusions:

• The updated PCI risk model and its simplified score offer robust prediction of in-hospital/30-day mortality.
• The full logistic model provides slightly better calibration for high-risk individuals.
• These tools can aid in assessing patient risk and improving outcomes in PCI.