Association Between State-Wide Cardiac Quality Improvement Program and Costs Following Intervention for Coronary Artery Disease

Abstract

Background

Since 2010, all non-VA hospitals performing cardiac surgeries and percutaneous interventions in Washington State have participated in the Cardiac Care Outcomes Assessment Program (COAP), a data-driven, physician-led collaborative quality improvement (QI) collaborative. Prior literature has demonstrated QI programs such as COAP can avert avoidable utilization such as hospital readmissions. However, it is unknown whether such improvements translate into economic benefits.

Hypothesis

This study compared downstream healthcare costs between patients undergoing cardiac interventions for coronary artery disease (CAD) at hospitals that were and were not participating in COAP.

Methods

Post hoc analysis of Medicare administrative and claims data examined 2.5 million randomly selected deidentified beneficiaries receiving a percutaneous coronary intervention or coronary artery bypass grafting between 2013 and 2020. Total costs were defined as all reimbursements paid by Medicare for up to 5 years following cardiac intervention. Because all non-VA hospitals in Washington State participated in COAP, we compared respective groups of patients receiving intervention in Washington State with all non-Washington states, adjusting for patient demographics and comorbidity. To model costs, we applied a multipart estimator, which distinguishes the impact of QI program participation due to survival and utilization while accounting for censoring.

Results

Total 5-year downstream costs were $3861 lower (95% confidence interval [CI] = $1794 to $5741) among patients receiving cardiac intervention at COAP-exposed hospitals. Lower costs were largely driven by lower utilization during calendar quarters where death was not observed.

Conclusions

Participation in this state-wide cardiac quality improvement program was associated with economic benefits in patients receiving intervention for CAD.