Outcome differences across transplant centers: comparison of two methods for public reporting

Area of Science:

• Nephrology
• Transplant Surgery
• Biostatistics

Background:

• The Scientific Registry of Transplant Recipients (SRTR) uses observed-to-expected (OE) ratios to report transplant center performance.
• The accuracy of the current OE method for standardized reporting has not been previously evaluated.
• Public reporting of transplant outcomes aims to incentivize centers to improve patient and graft survival rates.

Purpose of the Study:

• To compare the accuracy of the generalized mixed-effect (ME) method against the OE method for standardizing transplant center outcomes.
• To evaluate the association between public reporting initiatives and improvements in transplant center performance.

Main Methods:

• The study analyzed kidney transplant data from the United States Renal Data System (USRDS) between January 1, 1996, and September 30, 2009.
• Accuracy was quantified using the root mean square error (RMSE) comparing standardized rates between different time periods.
• A generalized mixed-effect (ME) model was employed as an alternative to the traditional OE ratio method.

Main Results:

• The ME method demonstrated a 0.5% to 4.5% lower RMSE compared to the OE method, indicating higher accuracy.
• The ME method resulted in a narrower range between the 5th and 95th percentile standardized rates for 3-year graft survival (7.5% vs. 10.5%) and patient survival (4.7% vs. 7.9%).
• Significant outcome variations across centers were observed, with differences accounting for 33% of deaths and 29% of graft failures within 3 years post-transplant.

Conclusions:

• The generalized mixed-effect (ME) method provides a more accurate approach for public reporting of transplant center outcomes.
• Further investigation is needed to understand why public reporting has not yet led to a reduction in outcome disparities among transplant centers.