EIC

ESRD Network 6 (what is this anyway, in context of OPTN regions?)
GA, NC, SC
Isn’t patient mix of interest?
Can’t rely on downstream analytic fixes in context of complex network of effects
That is what you wish to study!!!
So imputing the data doesn’t make any sense Ecology: referral rate: E-STAR; network 6 (faustian bargain vs. islamic finance); upcoming efforts Symbiolotology: patterns Teleology: not clear what is being optimized, but if its the “upstrea” patient mix (clinical and demographic attributes), then clearly this is more speculation than study

EIC#

This manuscript addresses an important issue but fundamentally misdiagnoses the problem. The proposed statistical fixes risk reinforcing existing biases rather than revealing true referral patterns. Major revisions are necessary, particularly in integrating patient mix considerations and reconsidering the reliance on imputation. If refined, the study could provide valuable insights for transplant policy and practice. But it is unlikely that the authors have these data.

Authors#

Overall Assessment: This letter to the editor attempts to refine kidney transplant referral rate estimates when transplant center participation is incomplete. The authors employ the Early Steps to Transplant Access Registry (E-STAR) in ESRD Network 6, using statistical imputation methods to adjust denominators. While the topic is relevant, the methodological approach raises concerns—particularly the reliance on imputation within a complex system where referral patterns are inherently linked to patient mix, geographic disparities, and center-specific factors. The approach risks mistaking statistical fixes for structural realities, potentially obscuring rather than clarifying referral dynamics.

Strengths:

Timely and Relevant: The issue of incomplete transplant data is pressing, especially with increasing national data collection efforts under OPTN and IOTA.
Clear Structure: The manuscript is well-organized and presents findings efficiently within the constraints of a letter format.
Technical Execution: The simulation framework systematically tests multiple adjustment methods.

Major Concerns:

Weak Biological and Clinical Integration The study approaches referral rates purely as a statistical problem, overlooking the biological and clinical heterogeneity of patients across transplant centers. ESRD Network 6 (GA, NC, SC) has a specific demographic and clinical patient mix—without accounting for these variables, the proposed denominator adjustments are potentially misleading. Referral rates are not just about missing data but about who is missing from the data.
Misplaced Reliance on Imputation The authors propose imputation methods that assume referral behavior follows predictable, generalizable patterns. However, in a network where patient characteristics, socioeconomic factors, and provider incentives create nonlinear effects, such imputation is unlikely to be reliable. The very phenomenon they wish to study—referral dynamics—should not be “adjusted away.”
Lack of Clear Optimization Goal It is unclear what the authors are optimizing—if the goal is to estimate “true” referral rates, then the methodology needs a more explicit justification of why denominator adjustments accurately represent reality (this journal and a letter format may not be suited). If the aim is to create a practical proxy for national referral data collection, then broader policy implications should be discussed.

Minor Issues:

Results Interpretation: Why do some adjustments overestimate the referral rate at low participation? A clearer explanation would improve accessibility for a clinical audience.
Policy Linkage: While the OPTN directive and IOTA model are mentioned, the real-world application of these findings is underdeveloped—especially in terms of how to increase center participation rather than just adjust for its absence.

flick 20250414180445-2t3Z#

flick 20250414201855-GyCJ#

flick 20250414205057-gLSW#

flick 20250414205331-D0sY#