Evaluation of Accepting Kidneys of Varying Quality for Transplantation or Expedited Placement With Decision Trees

Vikram Kilambi, Kevin Bui, Gordon B. Hazen, John J. Friedewald, Daniela P. Ladner, Bruce Kaplan, Sanjay Mehrotra

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

BACKGROUND: Underutilization of marginal-quality kidneys for transplantation produced ideas of expediting kidney placement for populations with decreased opportunities of receiving transplants. Such policies can be less efficacious for specific individuals and should be scrutinized until the decision-making for accepting marginal-quality organs, which has relied on experiential judgment, is better understood at the individual level. There exist rigorous tools promoting personalized decisions with useful and objective information. METHODS: This article introduces a decision-tree methodology that analyzes a patient's dilemma: to accept a kidney offer now or reject it. The methodology calculates the survival benefit of accepting a kidney given a certain quality now and the survival benefit of rejecting it. Survival benefit calculation accounts for patients' and donors' characteristics and transplant centers' and organ procurement organizations' performances and incorporates patients' perceived transplant and dialysis utilities. Valuations of rejecting an offer are contingent on future opportunities and subject to uncertainty in the timing of successive kidney offers and their quality and donor characteristics. RESULTS: The decision tree was applied to a realistic patient profile as a demonstration. The tool was tested on 1000 deceased-donor kidney offers in 2016. Evaluating up to 1 year of future offers, the tool attains 61% accuracy, with transplant utility of 1.0 and dialysis utility of 0.5. The accuracy reveals potential bias in kidney offer acceptance/rejection at transplant centers. CONCLUSIONS: The decision-tree tool presented could aid personalized transplant decision-making in the future by providing patients with calculated, individualized survival benefits between accepting and rejecting a kidney offer.

Original languageEnglish (US)
Pages (from-to)980-989
Number of pages10
JournalTransplantation
Volume103
Issue number5
DOIs
StatePublished - May 1 2019

Fingerprint

Decision Trees
Transplantation
Kidney
Transplants
Survival
Tissue Donors
Dialysis
Decision Making
Tissue and Organ Procurement
Graft Rejection
Kidney Transplantation
Uncertainty
Organizations
Population

ASJC Scopus subject areas

  • Transplantation

Cite this

@article{743e03d1aaef48248d8cfff501181936,
title = "Evaluation of Accepting Kidneys of Varying Quality for Transplantation or Expedited Placement With Decision Trees",
abstract = "BACKGROUND: Underutilization of marginal-quality kidneys for transplantation produced ideas of expediting kidney placement for populations with decreased opportunities of receiving transplants. Such policies can be less efficacious for specific individuals and should be scrutinized until the decision-making for accepting marginal-quality organs, which has relied on experiential judgment, is better understood at the individual level. There exist rigorous tools promoting personalized decisions with useful and objective information. METHODS: This article introduces a decision-tree methodology that analyzes a patient's dilemma: to accept a kidney offer now or reject it. The methodology calculates the survival benefit of accepting a kidney given a certain quality now and the survival benefit of rejecting it. Survival benefit calculation accounts for patients' and donors' characteristics and transplant centers' and organ procurement organizations' performances and incorporates patients' perceived transplant and dialysis utilities. Valuations of rejecting an offer are contingent on future opportunities and subject to uncertainty in the timing of successive kidney offers and their quality and donor characteristics. RESULTS: The decision tree was applied to a realistic patient profile as a demonstration. The tool was tested on 1000 deceased-donor kidney offers in 2016. Evaluating up to 1 year of future offers, the tool attains 61{\%} accuracy, with transplant utility of 1.0 and dialysis utility of 0.5. The accuracy reveals potential bias in kidney offer acceptance/rejection at transplant centers. CONCLUSIONS: The decision-tree tool presented could aid personalized transplant decision-making in the future by providing patients with calculated, individualized survival benefits between accepting and rejecting a kidney offer.",
author = "Vikram Kilambi and Kevin Bui and Hazen, {Gordon B.} and Friedewald, {John J.} and Ladner, {Daniela P.} and Bruce Kaplan and Sanjay Mehrotra",
year = "2019",
month = "5",
day = "1",
doi = "10.1097/TP.0000000000002585",
language = "English (US)",
volume = "103",
pages = "980--989",
journal = "Transplantation",
issn = "0041-1337",
publisher = "Lippincott Williams and Wilkins",
number = "5",

}

Evaluation of Accepting Kidneys of Varying Quality for Transplantation or Expedited Placement With Decision Trees. / Kilambi, Vikram; Bui, Kevin; Hazen, Gordon B.; Friedewald, John J.; Ladner, Daniela P.; Kaplan, Bruce; Mehrotra, Sanjay.

In: Transplantation, Vol. 103, No. 5, 01.05.2019, p. 980-989.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Evaluation of Accepting Kidneys of Varying Quality for Transplantation or Expedited Placement With Decision Trees

AU - Kilambi, Vikram

AU - Bui, Kevin

AU - Hazen, Gordon B.

AU - Friedewald, John J.

AU - Ladner, Daniela P.

AU - Kaplan, Bruce

AU - Mehrotra, Sanjay

PY - 2019/5/1

Y1 - 2019/5/1

N2 - BACKGROUND: Underutilization of marginal-quality kidneys for transplantation produced ideas of expediting kidney placement for populations with decreased opportunities of receiving transplants. Such policies can be less efficacious for specific individuals and should be scrutinized until the decision-making for accepting marginal-quality organs, which has relied on experiential judgment, is better understood at the individual level. There exist rigorous tools promoting personalized decisions with useful and objective information. METHODS: This article introduces a decision-tree methodology that analyzes a patient's dilemma: to accept a kidney offer now or reject it. The methodology calculates the survival benefit of accepting a kidney given a certain quality now and the survival benefit of rejecting it. Survival benefit calculation accounts for patients' and donors' characteristics and transplant centers' and organ procurement organizations' performances and incorporates patients' perceived transplant and dialysis utilities. Valuations of rejecting an offer are contingent on future opportunities and subject to uncertainty in the timing of successive kidney offers and their quality and donor characteristics. RESULTS: The decision tree was applied to a realistic patient profile as a demonstration. The tool was tested on 1000 deceased-donor kidney offers in 2016. Evaluating up to 1 year of future offers, the tool attains 61% accuracy, with transplant utility of 1.0 and dialysis utility of 0.5. The accuracy reveals potential bias in kidney offer acceptance/rejection at transplant centers. CONCLUSIONS: The decision-tree tool presented could aid personalized transplant decision-making in the future by providing patients with calculated, individualized survival benefits between accepting and rejecting a kidney offer.

AB - BACKGROUND: Underutilization of marginal-quality kidneys for transplantation produced ideas of expediting kidney placement for populations with decreased opportunities of receiving transplants. Such policies can be less efficacious for specific individuals and should be scrutinized until the decision-making for accepting marginal-quality organs, which has relied on experiential judgment, is better understood at the individual level. There exist rigorous tools promoting personalized decisions with useful and objective information. METHODS: This article introduces a decision-tree methodology that analyzes a patient's dilemma: to accept a kidney offer now or reject it. The methodology calculates the survival benefit of accepting a kidney given a certain quality now and the survival benefit of rejecting it. Survival benefit calculation accounts for patients' and donors' characteristics and transplant centers' and organ procurement organizations' performances and incorporates patients' perceived transplant and dialysis utilities. Valuations of rejecting an offer are contingent on future opportunities and subject to uncertainty in the timing of successive kidney offers and their quality and donor characteristics. RESULTS: The decision tree was applied to a realistic patient profile as a demonstration. The tool was tested on 1000 deceased-donor kidney offers in 2016. Evaluating up to 1 year of future offers, the tool attains 61% accuracy, with transplant utility of 1.0 and dialysis utility of 0.5. The accuracy reveals potential bias in kidney offer acceptance/rejection at transplant centers. CONCLUSIONS: The decision-tree tool presented could aid personalized transplant decision-making in the future by providing patients with calculated, individualized survival benefits between accepting and rejecting a kidney offer.

UR - http://www.scopus.com/inward/record.url?scp=85065432926&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85065432926&partnerID=8YFLogxK

U2 - 10.1097/TP.0000000000002585

DO - 10.1097/TP.0000000000002585

M3 - Article

C2 - 30720682

AN - SCOPUS:85065432926

VL - 103

SP - 980

EP - 989

JO - Transplantation

JF - Transplantation

SN - 0041-1337

IS - 5

ER -