Does enhancing standard registry models with prehospital delay and hemodynamic measures improve the prediction of clinical outcomes and revascularization delays in STEMI patients undergoing primary PCI?
Incorporating prehospital delay and initial hemodynamic measures into standard clinical registry models improves the prediction of short-term mortality and readmission, providing a practical tool for assessing hospital quality in STEMI care.
Acute myocardial infarction (AMI) is one of the leading contributors to the high mortality associated with cardiovascular disease. AMI outcomes vary substantially across hospitals, and quality assurance activities are necessary to standardize care practices and reduce the disease burden. The purpose of this research was to develop and validate hospital performance indicators for the inpatient care of ST-elevation myocardial infarction, which is associated with higher mortality. This dissertation comprises a series of retrospective cohort analyses conducted to identify independent predictors of clinical outcomes and revascularization delays among patients undergoing primary percutaneous coronary intervention (PCI) in New York State from December 1, 2016, to November 30, 2019, and to develop practical tools for assessing hospital quality and equity. The base clinical registry model included standard registry risk factors. Enhancing the base registry model with prehospital delay and initial hemodynamic measures significantly improved the prediction of in-hospital/30-day mortality outcome for primary PCI, making this modestly improved model a reliable and practical tool for assessing individual hospital quality and equity. The modestly improved model also proved to be a reliable and practical method for evaluating hospital quality and equity based on hospital-risk-adjusted 30-day readmission outcomes. Compared with 30-day outcome metrics, 6-month metrics based on modestly improved models were less valid and practical for evaluating individual hospital quality and equity. Finally, incorporating the mode of arrival and prehospital delay variables significantly enhanced the base model performance in detecting door-to-balloon delay, making the modestly improved model valid and practical for evaluating individual hospital performance.
Zaza Samadashvili (Wed,) studied this question.