Abstract
BACKGROUND: Many deceased donor kidneys go unused despite growing demand for transplantation. Early identification of organs at high risk of nonuse can facilitate effective allocation interventions, ensuring these organs are offered to patients who could potentially benefit from them. While several machine learning models have been developed to predict nonuse risk, the complexity of these models compromises their practical implementation. METHODS: We propose simplified, implementable nonuse risk prediction models that combine the Kidney Donor Risk Index (KDRI) with a small set of variables selected through machine learning or transplantation expert input. Our approach also account for Organ Procurement Organization (OPO) level factors affecting kidney disposition. RESULTS: The proposed models demonstrate competitive performance compared to more complex models that involve a large number of variables while maintaining interpretability and ease of use. CONCLUSION: Our models provide accurate, interpretable risk predictions and highlight key drivers of kidney nonuse, including variation across OPOs. These findings can inform the design of effective organ allocation interventions, increasing the likelihood of transplantation for hard-to-place kidneys.