Public Reporting and Case Selection in Congenital Heart Surgery: Signals From a Multicenter Observational Study.

Document Type

Article

Publication Date

6-9-2026

Identifier

DOI: 10.1016/j.jacc.2025.11.042

Abstract

BACKGROUND: Public reporting for congenital heart surgery promotes transparency and quality improvement but has the potential for the unintended consequence of risk avoidance. To assess this possibility, it is necessary to understand both changes in case selection and outcomes after public reporting.

OBJECTIVES: This study sought to examine the association between adoption of public reporting with hospital performance and predicted probability of mortality as a measure of case-mix complexity.

METHODS: Using the Inpatient Essentials Database, we identified all congenital heart surgeries performed at centers who elected to participate in the Society of Thoracic Surgeons-Congenital Heart Surgery Database public reporting program 2016 to 2019. Surgeries were stratified into pre- and post-public reporting cases based on the midpoint of the year in which each hospital adopted public reporting. Interrupted time-series with generalized estimating equations was used to quantify associations between public reporting and the outcomes of interest, and supplemented with observed/expected mortality before and after public reporting.

RESULTS: Among 16,401 congenital heart surgeries from 18 centers, 9,409 cases were performed before and 6,992 after public reporting. Overall, there were 500 (3.0%) in-hospital deaths. The unadjusted odds of in-hospital mortality following adoption of public reporting dropped by 35% (95% CI: 15%-50%; P = 0.002; absolute risk: 2.5% vs 3.4%), but this difference was no longer significant after adjusting for case complexity (95% CI: -2.97 to 0.43; P = 0.14). Importantly, the predicted probability of mortality decreased 0.5 percentage points (95% CI: 0.1-0.8; P = 0.006) immediately after hospitals adopted public reporting, suggesting an acute change in case selection. There was higher than expected mortality at the beginning of the study period that declined before public reporting.

CONCLUSIONS: The association of public reporting with mortality is complex. Given no change in risk-standardized mortality, the lower unadjusted mortality after public reporting in these 18 centers appears to reflect both an improvement in observed/expected mortality before reporting, along with a selection of lower-risk cases after public reporting.

Journal Title

Journal of the American College of Cardiology

Volume

87

Issue

22

First Page

3143

Last Page

3153

PubMed ID

41603821

Keywords

congenital heart disease; outcomes; public reporting

Comments

Grants and funding

Link to Full Text
Library Record

Share

COinS