Pitfalls of Administrative Censoring in Survival Models with Time-Indexed Inputs
2026-07-11 • Machine Learning
Machine Learning
AI summaryⓘ
The authors explain a problem called administrative-cutoff leakage that happens when survival models learn about the date information hidden in clinical data instead of the actual risk of an event like disease progression. This happens because newer records have less follow-up time before the study ends, so models can guess the event risk from how much follow-up time is available rather than true patient risk. They show this problem can falsely improve model performance measures in simulations and real data, and suggest datasets should have enough follow-up time after the last recorded input to avoid this bias. Their work helps improve the reliability of predictions in clinical time-to-event studies.
Survival modelsTime-to-event outcomesAdministrative-cutoff leakageInformative censoringFixed-horizon AUCHarrell's C-indexClinical predictionFollow-up timeRetrospective datasetsBias in prediction models
Authors
Yanqi Xu, Hui Dai, Carlos Fernandez-Granda, Krzysztof J. Geras, Yiqiu Shen
Abstract
Survival models can model time-to-event outcomes using partially observed data. They are widely used in clinical prediction, including cancer risk, disease progression, treatment response, and mortality. Recent models often rely on rich inputs collected at a specific clinical encounter, such as medical images, laboratory tests, electronic health record snapshots, or sensor measurements. In large retrospective datasets, these inputs are usually collected over many calendar years. As a result, they may contain clues about when they were acquired through changes in devices, protocols, documentation, patient mix, or clinical practice. This creates a potential failure mode when outcomes are observed only up to a fixed study end date. More recent records necessarily have less potential follow-up than older records. A model that can infer the record date from the input may therefore learn to predict how much follow-up was available rather than the patient's true risk of experiencing the event. We call this failure mode administrative-cutoff leakage. In this paper, we characterize when this leakage can occur, distinguish it from classical informative censoring and genuine temporal changes in risk, and propose practical ways to detect it. In simulations, we show that administrative-cutoff leakage can inflate fixed-horizon AUC and can also affect Harrell's C-index under realistic follow-up patterns. We then demonstrate the same behavior in a real mammography cohort. These results motivate a simple design principle for survival prediction: for an n-year prediction task, the dataset should provide at least n years of potential follow-up after the latest input date. Otherwise, the models may be subject to bias induced by administrative-cutoff leakage.