Reproducible Survival Analysis Code for One-Year Outcomes After Acute-Care Hospitalization in Uterine, Tubo-Ovarian, and Cervical Malignancies
Year: 2026
Type: software
Abstract: This software archive provides the reproducible analysis code supporting the study “One-Year Survival After Acute-Care Hospitalization for Uterine, Tubo-Ovarian, and Cervical Malignancies: Cox Regression, Restricted Mean Survival Time, and Hierarchical Bayesian Time-Varying Survival Modeling.”The workflow implements ICD-based gynecologic malignancy phenotype ascertainment, one-patient/one-index-admission cohort construction, Kaplan–Meier estimation, restricted mean survival time, multivariable Cox proportional-hazards regression with robust covariance, proportional-hazards diagnostics, empirical standardization, hierarchical Bayesian piecewise-exponential time-varying survival modeling, semantic missingness auditing, survival-informed fully conditional specification multiple imputation, Rubin pooling, and prespecified sensitivity analyses.The archive contains software and documentation only. It contains no patient-level MIMIC-IV data, MIMIC-derived analytic dataset or model object, posterior draws, bootstrap resamples, or restricted source files. Independent reproduction requires credentialed access to MIMIC-IV v3.1 through PhysioNet under the applicable training, licensing, and data-use requirements.The primary analytic estimand is the association between gynecologic malignancy phenotype at the first qualifying acute-care hospitalization and all-cause mortality through 365 days. The analyses are prognostic and associational rather than causal.
Source: Figshare
Author: Sunday Adetunji
Language: English
Subfield: Oncology
Field: Medicine
Domain: Health Sciences
Open Access status: green
OpenAlex ID: W7204476470
