A0752
Title: Debiased lasso for stratified Cox models with application to the national kidney transplant data
Authors: Lu Xia - Michigan State University (United States) [presenting]
Bin Nan - University of California at Irvine (United States)
Yi Li - University of Michigan (United States)
Abstract: The scientific registry of transplant recipients (SRTR) system has become a rich resource for understanding the complex mechanisms of graft failure after kidney transplant, a crucial step for allocating organs effectively and implementing appropriate care. As transplant centers that treated patients might strongly confound graft failures, Cox models stratified by centers can eliminate their confounding effects. Also, since recipient age is a proven nonmodifiable risk factor, a common practice is to fit models separately by recipient age groups. The moderate sample sizes, relative to the number of covariates, in some age groups, may lead to biased maximum stratified partial likelihood estimates and unreliable confidence intervals, even when samples still outnumber covariates. To draw reliable inferences on a comprehensive list of risk factors measured from both donors and recipients in SRTR, a debiased lasso approach is proposed via quadratic programming for fitting stratified Cox models. Asymptotic properties are established, and the method is verified via simulations to produce consistent estimates and confidence intervals with nominal coverage probabilities. By accounting for nearly 100 confounders in SRTR, results from the proposed method may inform the refinement of donor-recipient matching criteria for stakeholders.
