Adjusting for Congenital Heart Surgery Risk Using Administrative Data.

Area of Science:

• Pediatric Cardiac Surgery
• Health Services Research
• Biostatistics

Background:

• Congenital heart surgery (CHS) involves diverse patients and procedures, necessitating risk standardization for comparative studies.
• Existing models may not fully capture the complexity of CHS patient populations and surgical interventions.

Purpose of the Study:

• To develop and validate a risk-adjustment model for CHS using the Risk Stratification for Congenital Heart Surgery for ICD-10 Administrative Data (RACHS-2) methodology.
• To identify key patient and procedural characteristics associated with in-hospital mortality in CHS.

Main Methods:

• Utilized the Kids' Inpatient Database 2019 to identify CHS cases with assigned RACHS-2 scores.
• Employed hierarchical logistic regression to analyze factors linked to in-hospital mortality.
• Validated the model using data from 24 State Inpatient Databases from 2017.

Main Results:

• The RACHS-2 score alone achieved a C-statistic of 0.81 for mortality prediction.
• Inclusion of age, payer, and complex chronic conditions improved model discrimination to 0.87.
• The model demonstrated strong discrimination in the validation cohort (C-statistic = 0.83).

Conclusions:

• A validated risk-adjustment model for CHS, incorporating RACHS-2 and other administrative data, has been developed.
• This model accurately accounts for patient and procedural factors influencing in-hospital mortality.
• The developed risk model is crucial for advancing health services research and quality improvement in CHS.