Bayesian Semi-Parametric Modeling of Colon Cancer Survival: Assessing the Impact of Socio-Demographic, Clinical, and Healthcare Access Factors Using BART

DOI

10.2991/978-94-6463-906-3_16How to use a DOI?

Keywords

Colon Cancer; Survival Analysis; Bayesian Additive Regression Trees (BART)

Abstract

Colon cancer remains one of the leading causes of cancer-related mortality worldwide. Identifying key prog-nostic factors and understanding their interactions are crucial for improving patient outcomes. In this study, Bayesian Addi-tive Regression Trees (BART) approach for modeling cancer survival presented by, focusing on a cohort of 18,296 colon cancer patients diagnosed in England in 2012.This study examines the influence of socio-demographic factors (age, depriva-tion level) and clinical characteristics (tumor stage, commodities, emergency presentation) on survival probabilities over a seven-year follow-up period. The analysis indicates that emergency presentation (EP) serves as a strong predictor of lower survival, with diagnoses through emergency routes being associated with significantly worse prognoses across various cancer stages. Age at diagnosis exhibits a significant, nonlinear effect on survival, where younger individuals tend to have better outcomes, though this effect diminishes over time. Stage at diagnosis emerges as the most influential factor, particularly in the period following diagnosis, while deprivation further contributes to survival disparities, especially in advanced stages. A comparison between the COXNPH-BART model and the traditional COXPHBART model demonstrates that the former pro-vides a substantially better fit to the data, offering deeper insights into the complex interactions among prognostic factors. The findings emphasize the importance of early detection and healthcare access in improving cancer survival, highlighting potential areas for intervention. Additionally, results suggest that tailored treatment approaches based on age and stage may enhance survival outcomes while addressing socio-economic disparities could help reduce survival inequalities. This work demonstrates the potential of BART in cancer survival modeling, offering a flexible framework to incorporate time-varying and non-linear relationships between multiple risk factors. These insights have important implications for clinical practice, policymaking, and future research into cancer survival disparities.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Younus Mohiuddin Mohammad
PY  - 2025
DA  - 2025/12/12
TI  - Bayesian Semi-Parametric Modeling of Colon Cancer Survival: Assessing the Impact of Socio-Demographic, Clinical, and Healthcare Access Factors Using BART
BT  - Proceedings of Botho University International Research Conference (BUIRC 2025)
PB  - Atlantis Press
SP  - 284
EP  - 295
SN  - 3005-155X
UR  - https://doi.org/10.2991/978-94-6463-906-3_16
DO  - 10.2991/978-94-6463-906-3_16
ID  - Mohammad2025
ER  -