Author + information
In 2008 the National Quality Forum approved a risk-adjustment method developed by the Agency for Healthcare Research and Quality (AHRQ) to facilitate benchmarking of congenital heart surgical programs. The method uses AHRQ's Statewide Inpatient Database (SID) to provide a reference population for facility-level comparisons of inpatient mortality among children <18 years of age. While the approved methodology includes some features of the Risk Adjustment for Congenital Heart Surgery (RACHS-1) method, some model parameters in the AHRQ method differ from those in the validated RACHS-1 model. As a part of the National Quality Forum endorsement maintenance process, researchers were asked to harmonize the two methodologies, to improved face validity and avoid confusion.
Developers of both methodologies met to examine differences between the two methods, and reached consensus for parameters that differed based on an evaluation of empirical evidence from the SID, as well as the Kids’ Inpatient Database (KID) and Pediatric Health Information System (PHIS). The overall goal was to maximize model performance as well as face validity.
Inclusion criteria for the new method were not changed, and include all discharges <18 years of age with ICD-9-CM procedure codes for congenital heart surgery or non-specific heart surgery and diagnosis codes for congenital heart disease. The final model includes the following parameters: procedural risk group, age (0-28 days, 29-90 days, 91-364 days, 1-17 years), low birth weight (500-2499g), other congenital anomalies (CCS 217 less 758.xx), multiple procedures, and transfer-in. Among 17,525 eligible cases in the SID 2008, the c-statistic for model performance was 0.815. Reliability adjustment was performed to account for random factors; systematic variation of risk-adjusted mortality rates by center ranged from 1.8 to 6.1 per 100 (5th to 95th percentile).
Congenital heart surgery programs can now obtain national benchmarking reports by applying AHRQ Quality Indicators software to their own administrative data, based on the harmonized RACHS method, with excellent discrimination and high face validity.
Poster Sessions, Expo North
Monday, March 11, 2013, 9:45 a.m.-10:30 a.m.
Session Title: Cardiac Surgical Outcomes and Quality of Care
Abstract Category: 28. Quality of Care and Outcomes Assessment
Presentation Number: 1289-107
- 2013 American College of Cardiology Foundation