Development of donor yield models

Increasing donor yield, or the number of organs transplanted per donor, has been a focus of the transplant community in recent years. However, an exclusive focus on observed yield, unadjusted for the donor characteristics, ignores important differences between donors and donor case mixes in donation service areas (DSAs). We analyzed deceased donor registry data from the Organ Procurement and Transplantation Network/Scientific Registry of Transplant Recipients from January 2006 to December 2009 (N = 32 116 donors). Overall yields and kidney yields were modeled using ordinal logistic regression, and logistic regression was used to model heart, lung, pancreas and liver yields. Donor characteristics, including demographics, historical information and positive serology were related to overall and organ-specific yield. This study shows the potential value of the yield models as evaluation metrics and as tools that can inform DSA-wide practices in donor management and can improve organ utilization.