Question
Asked 10th Apr, 2022
Gabriel Cassinelli Petersen
Gabriel Cassinelli Petersen

What is a good method to select variables for Cox Regression?

Dear biostats community,
I am trying to build a Cox (Proportional Hazards) Regression and have a dataset with several variables. I am trying to decide which variables are the most useful to use as covariates in my model. Which method do you use and recommend for doing this? I thought at first to do a univariate analysis and see which variables don't have a significant survival difference to exclude them but as I understand such procedures have the issue that they don't take into account the interaction between the different variables.
Thank you very much!!
Gabriel
BioStat
Cox Regression
Covariance
Dataset
Survival

Top contributors to discussions in this field

Maria Bettencourt Pires
Maria Bettencourt Pires
José Luis García Vigil
José Luis García Vigil
Arvind Singh
Arvind Singh
Hazim Hashim Tahir
Hazim Hashim Tahir
P.F. Zabrodskii
P.F. Zabrodskii

All Answers (3)

10th Apr, 2022
Bhogaraju Anand
Malla Reddi Institute of Medical Sciences
Hi,
Intuitively selecting based upon already associated covariates from literature. another way is step-wise selection.
This is a reference link for different methods which can be used.:
https://www.diva-portal.org/smash/get/diva2:1067479/FULLTEXT01.pdf
and
Garcia, R. I., Ibrahim, J. G., & Zhu, H. (2010). Variable selection in the cox regression model with covariates missing at random. Biometrics, 66(1), 97–104. https://doi.org/10.1111/j.1541-0420.2009.01274.x
For a theoretical discussion:
Handbook of Survival Analysis by John P. Klein, CRC Press Taylor & Francis Group
Cite
1 Recommendation
10th Apr, 2022
David Eugene Booth
Kent State University
Stepwise would be a terrible choice. I recommend lasso or the adaptive lasso version. See the attached papers. Best wishes, David Booth
Cite

Similar questions and discussions

Which method to select for fitting cox regression method?
Question
5 answers
  • Asked 12th Feb, 2016
  • Jignasa SathwaraJignasa Sathwara
My goal is to fit a cox regression model in SPSS. As I am still new to regression methods, I would appreciate a little of your help. 
First, I checked all variables in a univariate regression and select those, which have a p-value less than 0.05. The next step is that these significant variables (p-value < 0.05) are considered in a model with backward conditional. 
My question now is, whether I can find out, why one variable is removed in the backward selection or which variable rather explains the influence? I am surprised, that this variable drops out, although it is very significant in the univariate regression. 
Statistical significance of variables between univariate vs multivariate analysis using COX model?
Question
7 answers
  • Asked 13th Aug, 2019
  • Ahmed Amine ZergounAhmed Amine Zergoun
Please, when I run a multivariate analysis using COX model, some of variables become statistically significant while they are not in the univariate analysis. How could I interpret these results?
Thank you,
How do I calculate the best cutoff for ROC curves?
Question
15 answers
  • Asked 11th Aug, 2015
  • Tatsiana LeclairTatsiana Leclair
I ran a ROC curve on SPSS. The « Coordinates of the curve » table on my output gives me a footnote saying «All the other cutoff values are the averages of two consecutive ordered observed test values».
How can I know the sensitivity and specificity for each of my observed values?
Thanks!
How to do Regression analysis with two variables in different sample size?
Question
2 answers
  • Asked 7th Jan, 2023
  • Muhammad Abdou Tamer KhattabMuhammad Abdou Tamer Khattab
We want to analyze the relationship and impact between two variables in the education sector, The first variable is the independent variable (intellectual capital), which is measured by a sample of workers and leaders, size 150, and the second variable is the dependent variable (quality of service provided), and it is measured by a sample of Students and parents, its size 330
Stratified cox regression
Discussion
2 replies
  • Asked 19th May, 2022
  • Hatem AliHatem Ali
Dear Colleagues
I am struggling to understand stratified cox regression ?
In Google it says: The “stratified Cox model” is a modification of the Cox proportional hazards (PH) model that allows for control by “stratification” of a predictor that does not satisfy the PH assumption
However, I dont understand what stratification do?
My understanding is that the variable we stratify on is not included in the cox Re model? Is this correct?
So what happens then? Is a cox regression runs when this variable we stratify on is negative and run once again when it is positive ? And then the hazard ratios are averages of all the models?
If this is wrong, so what happens?
What if the variable we stratify on is really important factor for predicting survival?
How to interpret it in this case?
How many covariates can I use for propensity score matching to analyze the impact of a specific drug on death outcomes among COVID-19 patients?
Discussion
6 replies
  • Asked 1st Apr, 2022
  • Mehdi Azizmohammad LoohaMehdi Azizmohammad Looha
In my study, I am using propensity score matching to balance the effects of covariates on the impact of prednisolone on death outcomes among COVID-19 patients. 92 covariates have been considered, including demographic factors, signs and symptoms, other drugs, laboratory tests, and vital indicators. A problem arises when I remove one of the variables (the ALT). This changes the final results significantly.
How can I ensure whether I should remove that variable?
How to match 1:1 case:control in Rstudio?
Question
1 answer
  • Asked 4th Apr, 2022
  • Laia Gutierrez TorderaLaia Gutierrez Tordera
Hi,
I have a database with 300 patients, some of them controls and some cases. I want to randomly select and match 96 cases with 96 controls. I want to find a match 1:1, that is, for each control just one case. Without repeating them, only using them one time per pair. I want to match them exactly for Gender and in a range for Age (I am not looking for an exact match, but the most similar pair).
I want to do so in R, but I don't know how. I tried a package called ccoptimalmatch, but the match is not 1:1, but k:1. Moreover, I found it difficult to understand.
After matching them, I will run some experiments in the lab and for the analysis I wil compare both groups, so I need to match them.
Thanks! :)
Kaplan Meier Survival Analysis - Continuous Or Categorical Data?
Question
6 answers
  • Asked 18th Dec, 2020
  • Kevin WinstonKevin Winston
Greetings,
I would like to ask a dumb question if I may. Suppose in SPSS, I have data on the outcome for mortality (dead versus alive) and raw data for age. Should I just leave the data for age as continuous variable or should I dichotomize it ( <65 years and >= 65 years) ?
I have heard that we are allowed to use continuous data for Kaplan Meier analysis such as for age, weight, leukocyte county, etc. I wonder why so many papers analyze the Kaplan Meier survival with categorical data instead?
Many thanks
My status has changed from "Awaiting Reviewer Score" to Awaiting AE Decision". Could the Associate Editor reject the manuscript at this stage?
Question
30 answers
  • Asked 19th Nov, 2020
  • Bilal AhmadBilal Ahmad
Otherwise what this status means. If the Editor has received the comments from reviewer then why he is not sending it to Authors. As 10 days already spent till the status changed.??

Related Publications

Nonparametric estimator of survival curve under dependent censoring
Article
  • Oct 1989
Assume that conditional on a set of covariates, the survival and censoring times are independent. Under this particular dependent censorship model, nonparametric estimators of the marginal hazard and survival function are investigated when some of the covariates are continuous. Consistency of the estimators is established by proving the weak conver...
View
Scalable Sparse Cox's Regression for Large-Scale Survival Data via Broken Adaptive Ridge
Article
  • Dec 2017
This paper develops a new sparse Cox regression method for high-dimensional massive sample size survival data. Our method is an $L_0$-based iteratively reweighted $L_2$-penalized Cox regression, which inherits some appealing properties of both $L_0$ and $L_2$-penalized Cox regression while overcoming their limitations. We establish that it has an o...
View
View
Got a technical question?
Get high-quality answers from experts.
Ask a question