Work

NHANES-CVD-competing-risk: quantifying and predicting competing-risk bias in cardiovascular mortality

Year: 2026

Type: software

Abstract: Code and result tables for a study quantifying how much the conventionalKaplan-Meier estimator overestimates cardiovascular mortality whennon-cardiovascular death is treated as censoring. Contents: the NHANES 2005-2018 analysis pipeline (R), the Monte Carlosimulation in three variants (constant hazard, Weibull hazards with shape0.5-2.0, non-proportional and piecewise time-varying competing hazards;78 scenarios), and the derivation and validation of a reporting thresholdR = q x Lambda / 2, where q is the competing share of deaths andLambda = -ln S_all. The pipeline reproduces every reported number and table byte-for-byte(eight independent end-to-end reruns, 25 result CSV tables MD5-identical).All random seeds are fixed. Raw NHANES files are not redistributed; theexact CDC URLs are listed in 00_raw/urls.txt.

Source: Zenodo (CERN European Organization for Nuclear Research)

DOI: 10.5281/zenodo.23565195

Authors: Cui, Likun, Lan, Qingsu, Shen, Zimeng, Miao, Qingrun, Han, Wenqiang, Zhong, Jingquan

Institution: Qilu Hospital of Shandong University

Language: English

Topic: Statistical Methods in Epidemiology

Subfield: Statistics and Probability

Field: Mathematics

Domain: Physical Sciences

SDG: Good health and well-being

Open Access status: green

OpenAlex ID: W7223223284

Cites:

Cited by:

Related to: 10