In the risk-truncated survival analysis, determining the proportional hazard status of the risk-truncated factor is critical to arriving at appropriate parameter estimations. Over the years, several methods for proportional hazard status determination have been proposed. In the work where the risk-truncated factor is age, can its distribution be used to determine its proportional hazard status? Three risk-truncated survival datasets were considered in this study. By boxplots, outliers of the factor were identified and histograms without the outliers overlaid with an added line and the normal densities of the factor were plotted to ascertain its distributions. Next, by the Schoenfeld test, the proportional hazard status of the factor was determined for each dataset. Finally, the distribution of the factor of each dataset was compared with its corresponding Schoenfeld test results. When the factor was skewed, it met the proportional hazard assumption in some cases and violated in another. Thus, the distribution of a quantitative variable – the risk-truncated factor – cannot be used to determine its proportional hazard status.

distribution, proportional hazard, Schoenfeld test.

Received: May 8, 2023; Accepted: June 14, 2023; Published: June 17, 2023

How to cite this article: John Darkwah, Can the proportional hazard status of a quantitative variable be determined by its distribution? Far East Journal of Theoretical Statistics 67(2) (2023), 211-216. http://dx.doi.org/10.17654/0972086323011

This Open Access Article is Licensed under Creative Commons Attribution 4.0 International License

References:

[1] P. D. Allison, Event History and Survival Analysis, Routledge, 2018.
[2] J. In and D. K. Lee, Survival analysis: Part I-analysis of time-to-event, Korean J. Anesthesiol. 71(3) (2018), 182-191.
[3] R. B. Geskus, Data Analysis with Competing Risks and Intermediate States, Chapman and Hall/CRC, 2019.
[4] Patrick Schober and Thomas R. Vetter, Survival analysis and interpretation of time-to-event data: the tortoise and the hare, Anesth. Analg. 127 (2018), 792-798. Available from: www.anesthesia-analgesia.org.
[5] L. P. Chen and G. Y. Yi, Semiparametric methods for left-truncated and right-censored survival data with covariate measurement error, Ann. Inst. Stat. Math. 73(3) (2021), 481-517.
Available from: https://doi.org/10.1007/s10463-020-00755-2.
[6] G. Frank, M. Chae and Y. Kim, Additive time-dependent hazard model with doubly truncated data, J. Korean Stat. Soc. 48(2) (2019), 179-193.
Available from: https://doi.org/10.1016/j.jkss.2018.10.005.
[7] R. Sutradhar and P. C. Austin, Relative rates not relative risks: addressing a widespread misinterpretation of hazard ratios, Ann. Epidemiol. 28(1) (2018), 54-57.
[8] F. Emmert-Streib and M. Dehmer, Introduction to survival analysis in practice, Machine Learning and Knowledge Extraction 1(3) (2019), 1013-1038.
Available from: www.mdpi.com/journal/make.
[9] O. Korosteleva, Clinical Statistics: Introducing Clinical Trials, Survival Analysis, and Longitudinal Data Analysis, Jones and Bartlett Publishers, 2008.
[10] M. Martcheva, An Introduction to Mathematical Epidemiology, Springer, Vol. 61, 2015.
[11] M. Wang, W. Li, N. Yehya, G. Keim and N. J. Thomas, Use of time-varying coefficients in a Cox regression model when the proportional hazard assumption is violated, Intensive Care Medicine 44(11) (2018), 2017-2019. Available from: https://doi.org/10.1007/s00134-018-5351-1.
[12] D. G. Kleinbaum and M. Klein, Survival Analysis, Springer, 2010.
[13] W. W. Abeysekera and M. R. Sooriyarachchi, Use of Schoenfeld’s global test to test the proportional hazards assumption in the Cox proportional hazards model: an application to a clinical study, Journal of the National Science Foundation of Sri Lanka 37 (2009), 41-45. DOI:10.4038/jnsfsr.v37i1.456.
[14] C. L. Loprinzi, J. A. Laurie, H. S. Wieand, J. E. Krook, P. J. Novotny, J. W. Kugler, J. Bartel, M. Law, M. Bateman and N. E. Klatt, Prospective evaluation of prognostic variables from patient-completed questionnaires, North Central Cancer Treatment Group, Journal of Clinical Oncology 12(3) (1994), 601-607.
[15] R. A. Kyle, T. M. Therneau, S. V. Rajkumar, J. R. Offord, D. R. Larson, M. F. Plevak and L. J. Melton III, A long-term study of prognosis in monoclonal gammopathy of undetermined significance, New England Journal of Medicine 346(8) (2002), 564-569.
[16] J. W. Osborne and A. Overbay, The power of outliers (and why researchers should always check for them), Practical Assessment, Research, and Evaluation 9(1) (2004), 6.
[17] P. T. Von Hippel, Mean, median, and skew: correcting a textbook rule, Journal of Statistics Education 13(2) (2005).