Accuracy of Medicare’s HACRP to identify outlier hospitals for SSIs

Editor's Note

Medicare’s Hospital Acquired Condition Reduction Program’s (HACRP) measure for surgical site infections (SSIs) after hysterectomy and colectomy unfairly places hospitals that are not statistical outliers at risk for financial penalties, this study from the University of Michigan finds.

SSIs occurred in 1.1% of hysterectomies and 4.8% of colectomies analyzed. The HACRP model for hysterectomy had a C-statistic significantly lower than the National Healthcare Safety Network (NHSN) (0.55 vs 0.61) or Michigan Surgical Quality Collaborative (MSQC) (0.55 vs 0.77) models. For colectomy, C-statistics were 0.57 for HACRP, 0.66 for NHSN, and 0.73 for MSQC models.

For both procedures, there were five high-outlier hospitals using HACRP, but fewer using the other models. Most hospitals in the bottom quartile were not statistical outliers, but would be flagged under HACRP. 

The researchers concluded that policy makers need to weigh the burden of data collection and the accuracy needed to identify which hospitals should receive financial penalties.