Cox Regression in Survival Analysis: Practical Insights for Clinicians
DOI:
https://doi.org/10.20344/amp.23078Keywords:
Investigative Techniques, Models, Statistical, Proportional Hazards Models, Regression Analysis, Survival AnalysisAbstract
Survival analysis is a fundamental tool in clinical research for evaluating time-to-event outcomes. While the Kaplan-Meier method remains a widely used univariable approach for estimating survival probabilities and comparing groups, it does not account for multiple risk factors simultaneously. To address this limitation, multivariable regression models are employed, with the Cox proportional hazards model (Cox regression) being the most commonly used. This paper provides a practical guide to Cox regression for clinicians, emphasizing its application in survival analysis rather than focusing on mathematical derivations. We discuss key concepts, including hazard ratios, model assumptions, variable selection, and interpretation of results. Additionally, we explore essential methodological considerations, such as assessing proportional hazards assumptions, handling missing data, and avoiding overfitting. By offering a step-by-step approach to implementing Cox regression in clinical research, this article aims to enhance understanding and improve the quality of survival analysis in medical studies. Practical examples illustrate how to interpret Cox regression results and their relevance in clinical decision-making.
Downloads
References
Gomes AP, Costa B, Marques R, Nunes V, Coelho C. Kaplan-Meier survival analysis: practical insights for clinicians. Acta Med Port. 2024;37:280-5. DOI: https://doi.org/10.20344/amp.21080
Bugnard F, Ducrot C, Calavas D. Advantages and inconveniences of the Cox model compared with the logistic model: application to a study of risk factors of nursing cow infertility. Vet Res. 1994;25: 134-9.
Haushona N, Esterhuizen TM, Thabane L, Machekano R. An empirical comparison of time-to-event models to analyse a composite outcome in the presence of death as a competing risk. Contemp Clin Trials Commun. 2020;19:100639. DOI: https://doi.org/10.1016/j.conctc.2020.100639
Swindell WR. Accelerated failure time models provide a useful statistical framework for aging research. Exp Gerontol. 2009;44:190-200. DOI: https://doi.org/10.1016/j.exger.2008.10.005
Shen Y, Yu S, Zhang Y, Wang S, Chen J, Xu Y, et al. The exploration of surgery and survival prediction in patients with peritoneal metastasis from gastric adenocarcinoma based on the SEER database. J Gastrointest Oncol. 2024;15:597-611. DOI: https://doi.org/10.21037/jgo-23-886
Cox DR. Regression models and life-tables. J R Stat Soc B. 1972;34:187-202. DOI: https://doi.org/10.1111/j.2517-6161.1972.tb00899.x
Kleinbaum DG, Klein M. Survival analysis: a self-learning text. Berlin: Springer Nature; 2012. DOI: https://doi.org/10.1007/978-1-4419-6646-9
Akinyemi JO, Afolabi RF, Awolude OA. Semi-parametric model for timing of first childbirth after HIV diagnosis among women of childbearing age in Ibadan, Nigeria. PLoS One. 2020;15:e0240247. DOI: https://doi.org/10.1371/journal.pone.0240247
Chen J, Cheng Z, Yao Y, Wang S. Variation of all-cause mortality with fat-free mass index (FFMI) and fat mass index (FMI) in individuals with asthma: results from the NHANES retrospective cohort study. J Epidemiol Glob Health. 2024;14:1555-68. DOI: https://doi.org/10.1007/s44197-024-00307-4
Park SH, David J. Reassessing Schoenfeld residual tests of proportional hazards in political science event history analyses. Am J Polit Sci. 2015;59:1072-87. DOI: https://doi.org/10.1111/ajps.12176
Grambsch PM, Therneau TM. Proportional hazards tests and diagnostics based on weighted residuals. Biometrika. 1994;81:515-26. DOI: https://doi.org/10.1093/biomet/81.3.515
Beyersmann J, Schumacher M. Time-dependent covariates in the proportional hazards model: background, structure, and interpretation. Lifetime Data Anal. 2007;13:401-18.
Fisher LD, Lin DY. Time-dependent covariates in the Cox proportional-hazards regression model. Annu Rev Public Health. 1999;20:145-57. DOI: https://doi.org/10.1146/annurev.publhealth.20.1.145
Therneau TM, Grambsch PM. Modeling survival data: extending the Cox model. Berlin: Springer Nature; 2000. DOI: https://doi.org/10.1007/978-1-4757-3294-8
Suissa S. Immortal time bias in observational studies of drug effects. Pharmacoepidemiol Drug Saf. 2007;16:241-7. DOI: https://doi.org/10.1002/pds.1357
Andersen PK, Gill RD. Cox’s regression model for counting processes: a large-sample study. Ann Stat. 1982;10:1100-20. DOI: https://doi.org/10.1214/aos/1176345976
Arslan J, Benke KK. Application of machine learning to ranking predictors of anti-VEGF response. Life. 2022;12:1176060. DOI: https://doi.org/10.3390/life12111926
Jin M. Imputation methods for informative censoring in survival analysis with time-dependent covariates. Contemp Clin Trials. 2024;136:107401. DOI: https://doi.org/10.1016/j.cct.2023.107401
Ogundimu EO, Altman DG, Collins GS. Adequate sample size for developing prediction models is not simply related to events per variable. J Clin Epidemiol. 2016;76:175-82. DOI: https://doi.org/10.1016/j.jclinepi.2016.02.031
Peduzzi P, Concato J, Feinstein AR, Holford TR. Importance of events per independent variable in proportional hazards regression analysis: II. Accuracy and precision of regression estimates. J Clin Epidemiol. 1995;48:1503-10. DOI: https://doi.org/10.1016/0895-4356(95)00048-8
Vittinghoff E, McCulloch CE. Relaxing the rule of ten events per variable in logistic and Cox regression. Am J Epidemiol. 2007;165:710-8. DOI: https://doi.org/10.1093/aje/kwk052
Hsieh FY, Lavori PW. Sample-size calculations for the Cox proportional hazards regression model with nonbinary covariates. Control Clin Trials. 2000;21:552-60. DOI: https://doi.org/10.1016/S0197-2456(00)00104-5
Riley RD, Snell KI, Ensor J, Burke DL, Harrell FE Jr, Moons KG, et al. Minimum sample size for developing a multivariable prediction model: part II – binary and time-to-event outcomes. Stat Med. 2019;38:1276-96. DOI: https://doi.org/10.1002/sim.7992
Schoenfeld DA. Sample-size formula for the proportional-hazards regression model. Biometrics. 1983;39:499-503. DOI: https://doi.org/10.2307/2531021
Postuma RB, Iranzo A, Hu M, Högl B, Boeve BF, Manni R, et al. Risk and predictors of dementia and parkinsonism in idiopathic REM sleep behaviour disorder: a multicentre study. Brain. 2019;142:744-59. DOI: https://doi.org/10.1093/brain/awz030
Spreafico M, Hazewinkel AD, van de Sande MA, Gelderblom H, Fiocco M. Machine learning versus Cox models for predicting overall survival in osteosarcoma: analysis of the EURAMOS-1 trial. Cancers. 2024;16:2880. DOI: https://doi.org/10.3390/cancers16162880
Kantidakis G, Hazewinkel AD, Fiocco M. Neural networks for survival prediction in medicine: a review and critical appraisal. Comput Math Methods Med. 2022;2022:1176060. DOI: https://doi.org/10.1155/2022/1176060
Wiegrebe S, Kopper P, Sonabend R, Bischl B, Bender A. Deep learning for survival analysis: a review. Artif Intell Rev. 2024;57:65. DOI: https://doi.org/10.1007/s10462-023-10681-3
Hazewinkel AD, Gelderblom H, Fiocco M. Prediction models with survival data: a comparison between machine learning and the Cox model. medRxiv. 2022:2022.03.29.22273112. DOI: https://doi.org/10.1101/2022.03.29.22273112
Moncada-Torres A, van Maaren MC, Hendriks MP, Siesling S, Geleijnse G. Explainable machine learning can outperform Cox regression and provide insights in breast cancer survival. Sci Rep. 2021;11:6968. DOI: https://doi.org/10.1038/s41598-021-86327-7
Wu Z, Cheng C, Sun X, Wang J, Guo D, Chen S, et al. The synergistic effect of the triglyceride-glucose index and serum uric acid on major adverse cardiovascular events after CABG: a multicentre cohort study. Cardiovasc Diabetol. 2023;22:103. DOI: https://doi.org/10.1186/s12933-023-01838-z
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Acta Médica Portuguesa
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
All the articles published in the AMP are open access and comply with the requirements of funding agencies or academic institutions. The AMP is governed by the terms of the Creative Commons ‘Attribution – Non-Commercial Use - (CC-BY-NC)’ license, regarding the use by third parties.
It is the author’s responsibility to obtain approval for the reproduction of figures, tables, etc. from other publications.
Upon acceptance of an article for publication, the authors will be asked to complete the ICMJE “Copyright Liability and Copyright Sharing Statement “(http://www.actamedicaportuguesa.com/info/AMP-NormasPublicacao.pdf) and the “Declaration of Potential Conflicts of Interest” (http:// www.icmje.org/conflicts-of-interest). An e-mail will be sent to the corresponding author to acknowledge receipt of the manuscript.
After publication, the authors are authorised to make their articles available in repositories of their institutions of origin, as long as they always mention where they were published and according to the Creative Commons license.
