Predicting Waiting Time and Quality of Kidney Offers for Kidney Transplant Candidates

In Canada, more than 3,500 patients are currently waiting for a kidney transplantation from a deceased donor. When a patient is registered on the transplant waiting list, both the expected waiting time before the first offer and its expected quality are unknown. This information is nonetheless crucial for transplant candidates in order to better manage their renal replacement therapy options, including accepting deceased donors with expanded criteria or identifying a living donor if expected waiting time is deemed too long. In this paper, we describe a novel method to estimate the arrival of future kidney offers and their quality for kidney transplant candidates. The method is based on the construction of a pseudo history of offers (occurrences and quality) for the current candidate using the past data of kidney donors and transplant candidates. The pseudo history is modeled with a marked Poisson process. The expected waiting time before the first offer and its expected quality for a given candidate can then be estimated. By providing such reliable quantitative estimates of time to and quality of future offers personalized to the transplant candidate, the proposed approach has the potential to guide and empower transplant candidates in managing their health condition.