Keywords:-
Article Content:-
Abstract
The Cox proportional hazards (CPH) model and the exponential Cox model are foundational tools in survival analysis, yet they impose structural assumptions; proportional hazards and constant baseline hazard, respectively that frequently fail in practice. This paper proposes and evaluates pooled logistic regression (PLR) applied to survival-piled (stacked) data as an alternative discrete-time survival estimator. Right-censored survival data are restructured into a long-format person-time binary classification dataset, and three models are compared: the exponential Cox model, the classical semi-parametric CPH model, and PLR. Monte Carlo simulations with 1,000 replications at each of six sample sizes (n = 25, 50, 100, 150, 200, 250) are conducted to assess stability, discrimination, and goodness-of-fit. Performance is measured by Harrell's concordance index (C-index) and the Akaike information criterion (AIC). Simulation results show that PLR maintains consistently high and stable C-index values (0.742–0.784) across all sample sizes, whereas the CPH model exhibits high variability (0.575–0.884) and the exponential Cox model underperforms throughout (0.532–0.671). AIC comparisons further demonstrate PLR's competitive goodness-of-fit at moderate and large sample sizes. Real-data validation using the NCCTG lung cancer dataset corroborates these findings. PLR's theoretical connection to the Cox partial likelihood is established formally, and its compatibility with machine-learning classification pipelines is discussed. The findings support PLR as a robust, flexible, and interpretable alternative for discrete-time survival analysis.
References:-
References
Klein JP, Goel PK (1992) Survival analysis: state of the art. Kluwer Academic, Boston
Cleves MA et al. (2008) An empirical basis for standardizing adherence measures derived from administrative claims data among diabetic patients. Med Care 46:1125–1133
Cox DR (1972) Regression models and life-tables. J R Stat Soc Ser B 34(2):187–220
Cox DR (1975) Partial likelihood. Biometrika 62(2):269–276
Efron B (1977) The efficiency of Cox's likelihood function for censored data. J Am Stat Assoc 72:557–565
Hess KR (1995) Graphical methods for assessing violations of the proportional hazards assumption in Cox regression. Stat Med 14(15):1707–1723
Pourhoseingholi MA et al. (2007) Comparing Cox regression and parametric models for survival of patients with gastric carcinoma. Asian Pac J Cancer Prev 8(3):412–416
Ramin R et al. (2011) Comparison of the results of Cox proportional hazards model and parametric models in the study of length of stay in a tertiary teaching hospital in Tehran, Iran. Acta Med Iran 49(10):650–658
Luck M et al. (2017) Deep learning for patient-specific kidney graft survival analysis. arXiv:1705.10245
Alonso E et al. (2024) Survival stacking ensemble model for lung cancer risk prediction. Stud Health Technol Inform 321:155–159
D'Agostino RB et al. (1990) Relation of pooled logistic regression to time dependent Cox regression analysis: The Framingham Heart Study. Stat Med 9(12):1501–1515
Andrew W, John C, Kyle R (2013) Combining parametric, semi-parametric, and non-parametric survival models with stacked survival models. Biostatistics 16(3):537–549
James HM et al. (2022) Stacked survival models for residual lifetime data. BMC Med Res Methodol 22(10):1–10
Bender R, Augustin T, Blettner M (2005) Generating survival times to simulate Cox proportional hazards models. Stat Med 24(11):1713–1723
Hosseini M et al. (2007) Comparison of survival models in studying breastfeeding duration. Hakim Health Syst Res J 10(1):66–71
Cox C et al. (2007) Parametric survival analysis and taxonomy of hazard functions for the generalized gamma distribution. Stat Med 26(23):4352–4374
Ginestet PG, Gabriel EE, Sachs MC (2022) Survival stacking with multiple data types using pseudo-observation-based AUC loss. J Biopharm Stat 32(6):858–870
Baek E-T et al. (2021) Survival time prediction by integrating Cox proportional hazards network and distribution function network. BMC Bioinformatics 22(1):192–206
Karthik D, Nader E (2011) A semi-parametric generalization of the Cox proportional hazards regression model. Comput Stat Data Anal 55(1):667–676
Wegbom AI, Victor AK, Isaac DE (2019) Comparison between semi-parametric Cox and parametric survival models in estimating the determinants of under-five mortality in Nigeria. Afr J Math Stat Stud 2(2):1–12
Collett D (2003) Modelling survival data in medical research, 2nd edn. Chapman and Hall/CRC, Boca Raton
Bland JM, Altman DG (2004) The logrank test. BMJ 328(7447):1073
Burnham KP, Anderson DR (2002) Model selection and multimodel inference, 2nd edn. Springer, New York
R Core Team (2023) R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna