Updating an Empirically Based Tool for Analyzing Congenital Heart Surgery Mortality

Area of Science:

• Cardiovascular Surgery
• Health Services Research

Background:

• The STAT Mortality Categories, developed in 2009, stratify congenital heart surgery procedures by mortality risk to characterize provider case mix.
• An update was needed to reflect contemporary outcomes and expand the scope of procedures analyzed.

Purpose of the Study:

• To update the STAT Mortality Score and Categories using contemporary data from the Society of Thoracic Surgeons Congenital Heart Surgery Database.
• To empirically derive new risk scores and category assignments for congenital heart surgery procedures, including those with multiple components.

Main Methods:

• Analysis of cardiovascular surgical operations from January 1, 2010, to June 30, 2017.
• Bayesian modeling was used to estimate the mortality risk of individual procedures and combinations, assuming risk is determined by the highest-risk component.
• Risk estimates were rescaled to STAT 2020 Scores (0.1–5.0), and categories were assigned to minimize within-category and maximize between-category variation.

Main Results:

• The STAT 2020 update includes 235 unique STAT Scores for procedure combinations, a 58% increase from the 2009 metrics.
• Mortality risk estimates ranged from 0.2% in Category 1 to 38.7% in Category 5.
• The STAT 2020 Scores and Categories showed good discrimination in an independent validation sample (C-statistic = 0.791 for Score, 0.779 for Category).

Conclusions:

• The updated STAT 2020 metrics accurately reflect current congenital heart surgery practice and outcomes.
• The expanded set of empirically based scores and categories, accounting for multi-component operations, facilitates precise case-mix assessment.
• Regular updates of STAT metrics are crucial for maintaining accurate case-mix characterization in congenital heart surgery.