Resource utilization in coronary artery bypass operation: does surgical risk predict cost?

Area of Science:

• Cardiovascular Surgery
• Health Economics
• Medical Informatics

Background:

• Healthcare cost containment is a growing concern in cardiac surgery.
• Clinical risk stratification is often used to predict healthcare costs, with length of stay (LOS) being a key factor.
• Direct correlation between patient-specific surgical risk and actual operative costs is not well-established.

Purpose of the Study:

• To investigate the correlation between The Society of Thoracic Surgeons (STS) predicted mortality risk and the actual direct costs of coronary artery bypass grafting (CABG).
• To assess the predictive accuracy of STS risk and LOS for individual patient costs in CABG procedures.

Main Methods:

• Analysis of variable direct costs, LOS, and STS risk in 628 consecutive CABG patients.
• Correlation analysis between STS risk, LOS, and direct costs for individual patients.
• Evaluation of predictive accuracy when patients were grouped into risk-stratified cohorts.

Main Results:

• Individual patient costs were well correlated with LOS (R2 = 0.48) but poorly correlated with STS risk (R2 = 0.12).
• STS risk poorly predicted LOS (R2 = 0.09).
• When grouped into risk cohorts, STS risk, LOS, and cost showed high correlations (R2 = 0.97–0.99).

Conclusions:

• Surgical risk models can accurately predict CABG costs in large patient cohorts.
• Preoperative risk data are poor predictors of cost for individual patients.
• Risk models are best suited for analyzing cost trends in cardiac surgery, not for individual financial risk assessment in clinical decision-making.