Cox Regression - Science topic

The Analyse for Pilot study should be based on the descriptive stat and ideally not involved inferential stat.

what about exploratory study? can I do inferential stat like cox regression or logistic regression for exploratory study?

Thanks

From my understanding, the baseline hazard or baseline survival function is unkown because cox regression is semi-parametric model. So why and how can we use it as a prediction model, for example using it to predict the 10 years survival probability.

I am working on a meta-analysis where i have extracted the data directly from KM curves via web plot digitizer to calculate HRs for the studies that reported only KM Curves. One of the study has three curves and web plot digitizer would give me a total of three groups. I was wondering if it is appropriate to combine the data for two of those groups and calculate an overall HR for a meta-analysis? Keeping in view that it is a time-event data and there's censoring too. I tried using the method elaborated by Cochrane but it gave me a really wide confidence Interval.

Anyone having any lead of how to deal with this?

I'm planning to use a cox regression in a future study exploring time-to-event or survival analysis comparing a control with an experimental group. I've seen sample size calculated through several packages, but I prefer G*Power and wanted to know if anyone's done this. Any resources would be appreciated.

Hi,

If there is a binary/nominal dependent variable and one wants to do a logistic regression analysis. In that case, for a given categorical predictor, how many minimum observations are required per category to conduct analysis? Should I include independent variables that have low frequencies in some categories of response?

For eg, if dependent variable is treatment received on time (1=yes, 0=no) and a categorical predictor of education status has 3 categories ( illiterate=0, primary=1, >secondary=2). In this case, if there were only 12 illiterate women who were treated on time but the rest categories across the DV had a decent sample (>=30). Can we still use education as a predictor for this analysis? Please share any references of rules/guidelines.

Dear Fellow Researchers,

We have analyzed the effect of various parameters on bladder cancer recurrence. Among included parameters, we had various continuous parameters.

In the literature, continuous parameters are often split into two groups based on the cut-off values. What is the advantage of doing this instead of keeping these parameters continuous?

For now, we determined cut-off values and repeated cox regression analysis for three parameters. The results in univariate and multivariate analyses are similar to when these parameters were continuous. However, we still have around 20 parameters only analyzed as continuous. Should we stay with continuous parameters or change to categorical?

I am looking forward to your suggestions,

Best,

Malgorzata

Greetings Fellow Researchers,

I am a newbie in using survival analysis (Cox Regression). In my data-set 10-40% cases have missing values (Depending on the variable I include in my analysis). Based on this I have two questions,

1- there are any recommendations on accepted percentage of cases dropped (missing values) from the analysis?

2- Should I impute the missing values of all the cases that were dropped (lets say maximum of 40%).

Thank you so much for your time and kind consideration.

Best,

Sarang

Dear RG members.

I conducted cox regression analysis and unfortunately the HR of some of the variables turned out extremely large, like 1.17e+09, 1.31e+10, and extremely low for some others (1.87e-21).

FYI: 24 variables were included in the regression, and before running the regression, I have checked interaction and around three variables were excluded because of this. So, why I am encountering this problem and any solution please!

Thank you in advance.

Hi, I am currently conducting a survival study to investigate the role of several potential biomarkers as prognostic factors in certain cancer. First, I perform Kaplan-Meier analysis for all the biomarkers and other relevant clinicopathologic data. However, only one biomarker fulfilled the proportional hazard criteria from the Kaplan-Meier curve. Other biomarkers and clinicopathologic variables do not fulfill the criteria.

I am wondering, do I still need to proceed to Cox Regression analysis? Can I include the other biomarkers and relevant clinicopathologic data in Cox Regression, even though they do not fulfill proportional hazard criteria during Kaplan-Meier analysis? Thank you.

I need to perform cox regression hazards ratio analysis, currently only having graphpad v8.0

Can anyone kindly help me to sort this issue?

#coxregressionanalysis #statistics

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?

Hi,

I developed probability of default model using cox PH approach. I use survival package in R. I have panel data with time-varying covariates. I made prediction with following code: predict(fit, newdata, type='survival'). However, predicted survival probability is not decreasing over time for each individual (see picture).

I wonder if prediction is marginal survival probability or cumulative survival probability?

If this is cumulative survival probability, why are it not decreasing across time?

Thanks

Which of the two models is better to analyze factors that influence the appearance of a certain event when the data is not censored? Cox Regression or Logistic Regression? Let´s add that time to the event is not really relevant.

Hi, I am currently performing survival analysis using SPSS. I wanted to determine which of the following factors: (expression of several proteins from immunohistochemistry and several clinicopathologic parameters) influence the survival of cancer patients. However, during univariate analysis, most of the factors did not fulfill the proportional hazard assumption (the Kaplan-Meier curve have multiple overlaps). Questions:

1) Can I still include the factors (especially the protein expressions) in Cox Regression analysis? It is a fairly small study with about 30 patients.

2) If not allowed, are there any potential solutions? Some of the curves overlap at multiple time points. (I have attached an example of the curve)

3) During Cox Regression, which method is preferred? Can I use 'Enter' or 'Backward LR'? If I use Backward LR, do I still need to perform univariate analysis before and filter the variable to be included into the regression? How about the 'Enter' method?

4) When I tried including all factors into Cox Regression, there are a couple of variables where the hazard ratio is < 1 during univariate analysis and non-significant, but becomes > 1 and significant during Cox regression. What are the possible explanations?

Thank you for your help.

Automated statistical inference in medical research meets the criterion of artificial intelligence. John McCarthy, widely recognized as the father of Artificial Intelligence, defined the term AI as “the science and engineering of making intelligent machines”. These would be computer tools, modeled on the functioning of the human mind, as well as technologies and research in the field of fuzzy logic, evolutionary computing, neural networks or artificial life, robotics.

In the field of clinical medicine the concept of AI is widely used , e.g. computer-assisted diagnosis, computer consultation systems (Stanford University, a leading center of AI in medicine), pattern recognition, machine learning, electronic health record EHR, clinical guidelines in a computerized form, diagnosis of the individual diseases and can be further listed.

We have carried out research focused on statistical elaboration of clinical data and some software solutions that automate statistical works.

In research projects defined as epidemiological and statistical, two people (or groups) are involved: an epidemiologist specifying the subject and scope of the analysis, the research plan, and statistician translating the research assumptions into the language of statistics, including the selection of statistical analyzes, finding libraries, correct execution and interpretation of the results analyzes. In the AI approach, an "intelligent computer program" works similarly to a human statistician. The epidemiologist plans the analyzes, saves the assumptions in metadata files for computer aims. Specialized computer program manages the statistical program in every step . First, it creates rational command texts based metafile , as if a programmer would do it using a screen interface and his mind (knowledge, skills, experience). Second, it "uses" a statistical package to execute prepared scripts. Third, he transforms the results of statistical program into a text for epidemiologist in plain tekst.. Statistical program is hidden from the user.

The project currently covers GLM, GLMM, one- and two-level linear and logistic regression models, Cox regression , KM, as well as preliminary data analysis.

If you would like to read the computer implementation of this issue I write in the discussion "AI Statistical Analysis of Clinical Data for Non-Statisticians - part 2: computer implementation"

I seem to be getting really exponentially high HR on multivariate analysis by cox-regression. I only have 70 or so patients and 5 poor outcomes. Any ideas why my results may have turned out this way?

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

Dear Colleagues,

In the Stata manual for Lasso, they came up with a new way to do inferential models using Lasso for linear and logistic regression.

It helps calculating standard error, thereby calculating odds ratio and confidence interval

The equation is as follows:

where d is the covariate of interest.

The double-selection solution Double selection is the easiest of the three to explain. Its algorithm is the following:

1. Run a lasso of d on x.

2. Run a lasso of y on x.

3. Let xe be the union of the selected covariates from steps 1 and 2.

4. Regress y on d and xe.

I wonder if the same equation can be applied on cox regression as cox regression is a special type of linear regression?

I wonder if you guys tell me your opinion ?does any one have reference for this?

or is it just theoretical thoughts?

Hi,

I have been performing survival analysis using Cox regression models, and I encountered situation when after adding time-varying effect for a variable X (X*time; variable violated the PH assumption), the added interaction with time was significant in the model, but the main effect of variable X was not, as illustarted below:

Model without interaction with time:

coef exp(coef) se(coef) z Pr(>|z|)

factor(X)1 0.4633 1.5894 0.1625 2.852 0.004 **

Model with interaction between X and time:

coef exp(coef) se(coef) z Pr(>|z|)

factor(X)1 -0.3978 0.6718 0.4444 -0.895 0.371

tt(factor(X)) 0.6230 1.8645 0.2816 2.212 0.027 *

In the study we are interested in the effect of the X variable on the survival outcome, and after inclusion of the time-varying effect X*time, I am no longer sure about the value of the variable X in describing the risk of the outcome, as the main effect is now not significant.

Is the significance of time-varying effect of variable X enough to assume that the variable X is significant for the outcome risk, even though the main effect is no longer significant in such scenario?

Or, do both of them, the main effect of X and the time-varying effect of X have to be significant in the model to be able to say that X is significant for the outcome?

Any help in interpreting these is very welcome.

I have performed a Cox regression model on patients' overall survival for both univariate and multivariate. A variable I interested most is only significant in univariate but not in multivariate. 1) How can I explain this result in my article? This variable is also significant in log-rank test.

2) Is there any other methods for me to try to make it more valuable? I am using R software.

Thanks a lot.

Dear Colleagues

I hope all is well.

I am interested in doing penalized cox regression using PYTHON.

I came across the following way to do penalized cox regression:

Penalized Cox Models — scikit-survival 0.17.1

The code blocks 10 till 13 are doing the following:

Choosing the penalty strength alpha by :calculating the best value of penalty strength among all the values of penalty strength (alpha) by reducing the value by 1% in each iteration

then it perform 5 fold cross-validation to estimate the performance – in terms of concordance index – for each αα.

I am confused and dont fully understand the codeblock 13. I found that after codeblock 13, it executes the model that has the best alpha and calculate the coefficients. Is this right?

I wonder in codeblock 13, the how was the model executed? is it by doing gridsearch?

What is the name or type of this model? Is it elastic net regression?

I found in statistical books that to verify the linear assumption of a Cox model I need to plot Martingale residuals.

However, I cannot find any explanation about interpretation of the plot!

So, if I plot predicted values versus Martingale residuals what have I to expect if linearity is satisfied?

Thank you in advance... please help me!

Greetings,

We have been conducting a retrospective cohort study. The variables we are examining are assumed to be very age-dependent and the exposed population is very small (~40 patients), therefore we have considered matching for age and sex at a 1:2 or 1:3 ratio to increase statistical power and limit confounding.

Which statistical test would be most appropriate for calculating risk ratios for dichotomous categorical variables?

This article ( https://academic.oup.com/epirev/article/25/1/43/718675 ) suggests conditional Poisson regression, which I have attempted in Stata, but it appears to work only for 1:1 matched pairs.

It also suggests an adjustment of Cox regression so as to yield the same results as conditional Poisson regression (" if the time to death or censoring is set to some arbitrary constant value and if the Breslow or Efron methods are used to account for tied survival times, the results will be the same as those from conditional Poisson regression, as the likelihoods for these methods are identical when the data come only from matched pairs ").

I have recently attempted a similar adjustment (as described here: https://www.ibm.com/support/pages/conditional-logistic-regression-using-coxreg ) to yield the same results as conditional logistic regression (odds ratio) for a 1:N matched case-control study using Cox regression.

If such an adjustment is possible, how exactly could it be implemented in SPSS? If not, what other alternatives are available to us in this juncture?

Thank you in advance.

Example 1: i want to test if Diabetes is a predictor of myocardial infarction. The result is this:

Covariate    b         SE        Wald       P      Exp(b) 95% CI of Exp(b)

Diabetes 1,1624 0,3164 13,4996 0,0002 3,1976 1,7254 to 5,9257

How can i interpret this result? the p is less than 0,05 but i don't understand if it is in favor of patients with diabetes or without diabetes.

Example 2: And with continuos variables, for example:

Covariate      b       SE        Wald        P       Exp(b) 95% CI of Exp(b)

RVD      -1,0549 0,1800 34,3351 <0,0001 0,3482 0,2451 to 0,4947

how can interpret the results?

Could someone help me, please?

Dear All,

I hope all is well.

I am trying to do a penalized cox regression analysis in Python.

However, in order for the function to work, the "event" and "time"

need to be stored as a structured array

similar to:

array([( True, 72.), ( True, 411.), ( True, 228.),.........,dtype=[('event', '?'), ('time', '<f8')])

I am struggling to do a structured array of the event and time.

I wonder if you guys can help me with that?

Thank you very much

Hello!

Has anyone tried plotting survival probabilities for a cox regression using Firth's penalized likelihood? I am using the coxphf() function in R. Any advice would be appreciated!

Hello

I have 10,000 data related to Covid patients who have 2,000 blood pressure and 8,000 do not have high blood pressure.

My main goal is to study the risk factors for death in patients with high blood pressure with Covid 19. On the other hand, I want to find the risk factors in patients who do not have high blood pressure and compare it with the group that has high blood pressure. As in the table below.

I want to make two separate coxs for both data. Do you think this comparison are correct? Do I need to make adjustments to compare the final risk factors so that I can make the right comparison?

(I can not match here.)

The main independent variable is A, but I want to adjust the model with a covariate B. When I check the PH assumption, A holds but B does not. Do I need to run a Proportional odds or an AFT model for that?

There is a problem I faced computing multivariable risk factor analysis using Cox regression in SPSS. I have 5 variables that are significant in univariate Cox regression with time-dependent variable. I do not know how can I run multivariable analysis using these 5 time-dependent variables in a multivariable Cox regression simultaneously.

Huge thanks in advance for solutions and advices.

Hi everyone, I would like to run a Cox regression progressively including potential confounding factors in the models (Model 0: no confounding factors; Model 1: 1 confounding factor; Model 2: 2 confounding factors; ...)

Since I never did it on my own, I am wondering if you could suggest a practical statistics software for this purpose.

PS I usually use Stata or GraphPad Prism.

Thank you for your collaboration and time.

I am getting this error"flat region resulting in a missing likelihood r(430)"while running cox regression in STATA. I have tried different settings for changing ties but I still get the same error. How can I fix this error using STATA. Any help will be useful. Thanks in advance!

Hello advisors. I am glad to be here in this Community. If you allow me, I would ask if anyone of the members here know about Databases of cutting tools. Nowadays, I am working on a project and the main goal is to research about lifetime of cutting tools using the Cox proportional hazards model. At the moment, I just have one Database that contains cutting speed, feed rate, tool failure time and depth of cut. My research question is: Which of these variables are the most representative that give us the time of failure? Any kind of help, comments and questions are welcome.

Dear Colleagues,

I hope all is well. I am sorry for very Naive question as I am very new to using R package.

I have imported a dataset from STATA to R package, installed the package for survival as follows:

library(haven)

data <- read_dta("C:/Users/user/Desktop/data.dta")

View(data)

install.packages(c("survival", "survminer"))

library (caret)

library (glmnet)

library (mlbench)

library (psych)

library("survival")

library("survminer")

I changed some of the varaibles in my dataset so to be recognised by R as factor variables (some has 2 levels, some has more than 2 levels)The factor variables are: ethnicity, sex, diabetes, HLA mismatches, steroids, donor type..

Examples:

data$DIAB=as.factor(data$DIAB)

data$sex=as.factor(data$sex)

data$ETHCAT=as.factor(data$ETHCAT)

I managed to define my time and event varaiables as follows:

time=data$finaltime

event=data$GSTATUS_DTHCNS_KI

I combined the time and event in one matrix, also combined all the independent variables in another matrix

as follows:

Y=cbind ( time, event)

X=cbind (data$sex, data$BMI_TCR, data$COLD_ISCH_KI, data$SERUM_CREAT, data$finalpra, data$AGE, data$STEROIDS_MAINT, data$induction, data$DIAB, data$dgf, data$timeondialysis, data$ETHCAT, data$KDPI, data$finalcmv)

I ran a cox regression model using the following syntax:

coxph<-coxph(Surv(time,event) ~X, method="breslow")

However, I found that R treated all the variables as continuous numeric variables. (even the ones I identified as factor variables !). I checked on these variables again and found that R recognise them as factor variables !

I wonder if any of you guys can help me how to fix this problem.

The second question:

I tried to fit a glmnet model , however I got error:

> Fit=glmnet (X,Y, family=”cox”)

Error: unexpected input in "Fit=glmnet (X,Y, family=”"

I am not sure why it is giving me error. I wonder if anyone can help me with this

Thank you very much

Kind regards

I would like to identify the predictors of seizure recurrence after a first seizure. Should I use a Cox regression model yielding hazard ratios or should I use a log-binomial regression yielding risk ratios?

From what I understand, I shouldn't use logistic regression because it yields odds ratios, which often overestimate risk ratios.

Hi there. I'm a medical student and now I'm working on survival prediction models. But I encountered some difficult problems about feature selection. The original sequencing data was huge, so that I relied on univariate COX regression to obtain a subset (subset A), and I'd like to perform Lasso regression to further select features for the final survival prediction model construction (using multi-variate COX regression). However, the subset A was still huge than I expected. Can I further obtain subsets by limiting the range of Hazard Ratio (HR) for Lasso regression? Or, could I perform Random Survival Forest to obtain subsets from subset A, for final survival prediction model construction? Is there anything I need to pay special attention to during these processes?

I'm using cox regression model to study the association of variable A with coronary heart disease. One of my covariates (variable B) has a clear non-linear relationship with predictor (Variable A), assessed by scatter plots; linear regression. My question is that does this cause any violation if the Schoenfeld residuals of the cox-model are fine? I'm using R.

Hello,

I am trying to show if there is any relationship between the different variables (weight, age, risk score) and freedom of re-intervention after a stent placement. Freedom of reinterevntion is a time depending value, so the more I think about it I think I should use cox's regression model. But on the other hand I am thinking weight/age/risk score and freedom of re-intervention are both continuous values so if I want to show a correlation between the two I should use the pearson/spearman? What do you think? Also if I am reporting the cox model, should I just report the hazard values and the p-values or may be more?

Thank you in Advance

I'm using a cox proportional hazards regression to analyze latency data, i.e. time until behavior x occurs. The model I'm running is fitted with the "coxme" package in R (which is a wrapper to the "survival" package if I'm not mistaken) because I wanted to fit a mixed model for the survival analysis. The model converges without problems, however, when I'm trying to test the assumptions of the model I get a fatal error, and my R session crashes. Specifically, when I try to test the "proportional hazard" assumption of the model using the "cox.zph" call, the R session crashes and I don't know why, because the function is supposed to work with both a mixed model (coxme) and a standard, non-mixed, model (which is a "coxph" object in the package terminology). I've tried the non-mixed version of my model and it provides the desired output, but it won't work for my intended model. I've also tried updating RStudio, so I have the latest version, but it didn't help. Finally, I've tried to manually increase the working memory dedicated to RStudio, in case the function was memory demanding, but it didn't help. Looking around at different forums has provided no answers either, both with general search parameters like "causes for R session crash" and more specific, like "cox.zph cause R session crash", but I could not find any help.

Has anyone experienced this error? Were you able to solve it, and if so, how?

I appreciate any advise I can get on this issue.

Dear Collegaues,

I hope all is well.

Thank you very much for your help

I am used to STATA and very new to using R package.

STATA has lasso inference for linear and logistic regression. However, it doesnt have LASSO features for cox regression.

I wonder if I can use R to do LASSO inference for cox regression model?

I am literally very new to R and would appreciate if you can help me do syntax in R for my model.

I am sorry that I am very Naive in R

If I am using STATA, I would do the following to produce the cox model:

1)stset PTIME, failure(PSTATUS)

2)stcox i.sex BMI_TCR COLD_ISCH_KI SERUM_CREAT END_CPRA i.ETHCAT AGE AMIS BMIS i.STEROIDS_MAINT AGE_DON i.DIAB i.dgf

3)estat phtest (to test proportional hazard assumptions)

I wonder what is the syntax to do the same in R ?

Also, what are the syntaxes to use this model to perform LASSO inference for cox harazrd regression?

Finally how to do the post-estimation tests after fitting in the LASSO inference for cox harard regression?

Thank you vey much for your help

Looking forward to hearing back from you

Dear research gate members, would you please let me know the specific assumptions and conditions to use these phrases?

Hello,

I built cox proportional hazard model with time-dependent covariates. when I predict survival probability, it wasn't monotonically decreasing for each ID. what is problem and how to handle it?

thanks

How to add all variables of ONE group (as categorical variables ) in Cox regression analysis without losing a single group. SPSS considers the first group as a reference/ control variable , because of this out 5 variables hazard ratio I am getting only 4 categorical variables answered.

Please suggest ?

Hi everyone!

I am wondering whether the prentice-weighted cox regression I need to do for a specific study design (case-cohort) can be done within the framework of multilevel modelling?

Is there any R packages for mixed-effects Cox model with prentice-weighting?

Dear all,

I am working on gene expression and Kaplan Meier curve dividing the patients in " high" and "low" using SPSS, then I want to do the cox proportional hazard analysis combining i.e the mutational status of one gene. I am naive in using SPSS Software How I can set up the analysis in order to find the Hazard ratio of specific combinations i.e X gene(high) and Y (gene) "mut" or "Wt".

Is it possible to merge effect sizes expressed as hazard ratio and odds ratio in the same dataset? Of course, the construct underlying the outcome of interest is the same, but in some articles was explored with logistic regression and in other articles was explored with Cox regression.

Hello everybody.

I am facing problem in selecting reference category of independent variable (categorical) in cox regression analysis using SPSS. There are option to select first or last category as reference. However, when I want to select first category as reference category, it ultimately become last one as reference category. I am not sure whether the last category is set as reference category by default in SPSS. I want to select the first category as reference. Would anyone help to solve this problem?. Regards

We are trying to find if there is an association between postoperative pulmonary complications (PPCs) and overall survival in a cohort of 417 patients. In Kaplan-Meier there is a significant difference in overall survival between patients with and without PPCs. After testing the proportional hazards assumption in cox regression (both through visual analysis and through log minus log plot) we found that our data failed to meet the assumptions. The way I interpret this, it means that the hazard of dying due to a postoperative pulmonary complication varies over time? I'm trying to figure out how to perform a survival analysis now that I can't use the standard cox regression.

From what I understand I could use time dependent coefficients (the same as variables in this example?) but I don't really understand what is meant by that or how I would do it in SPSS. Does it mean I turn my PPCs variable into a time dependent variable and then run the cox regression analysis the way I would if my data would have met the assumptions or how do I do it?

I would be really thankful for guidance or corrections if I have misunderstood something! I'm a PhD student and I don't have much experience in statistics so explain it to me like I'm 5 (a smart 5-year-old!)

We are interested in calculating the readmission rate after 12 months, and uncertain how much data to include in our analysis. Since it is a 12 months readmission rate that we are interested in, should we then include data 12 months after each hospitalisation, so they have a chance to get a readmitted within 12 months?

Hello everyone

I am using the patients' survival times in an open cohort study of cancer patients as a measure of the life expectancy

I would like to know the most important predictors for life expectancy considering clinical covariates and genomic covariates. I did a genomic analysis of my patients and I have all genes up-regulated and down-regulated before and after treatment.

Does anyone know what the best fit model in this case considering Overal Survival and the Event? Both target, not just one.

DEAR members of RG, I need your kindly statistical suggestion!

I was aiming to run Cox-regression, but I doubt the possibility given that the event to sample size ratio is <1:10. In my case, the events are 11, while sample size is 130. My question is, is it possible to run cox-regression under such circumstance?

Waiting for your statistical suggestion!

Thank you in advance.

#KevlynJones#AdrianEsterman#RogerJelliffe#SafaAounti

I have data from an experiment where we looked at the time-to-death of pseudoscorpions under three treatment conditions: control, heat, and submersion underwater. Both the control and heat groups were checked daily and those that survived were right-censored. However, the pseudoscorpions in the underwater group were unresponsive while submerged and would only become active again when they were removed from the water and allowed to dry. Therefore, we removed 10 individuals per day and checked how many of the 10 were dead. All individuals removed from the water on a given day were also removed from the study on that day i.e. we did not re-submerge individuals that were observed to be alive after they were removed from the water the first time. The underwater group is therefore left-censored. We have run Kaplan-Meier curves for all three groups and the underwater group has a much steeper curve and non-overlapping 95% confidence intervals compared to the control and heat groups.

Is there a way to further analyze all three groups in one model given that one level of the treatment is left-censored and the other two levels are both right-censored? Can a Cox regression be run on the left-censored group by itself to produce a hazard rate with 95% CI for the rate? I am a biologist so try to make your answer intelligible to a statistical novice.

SPSS, Data-anaylsis

I fitted a Cox proportional hazard model and checked the proportionality assumption using R's cox.zph function, with the following output:

chisq df p

Var1 0.0324 1 0.857

Var2 0.1972 1 0.657

log(var3) 4.1552 1 0.042

Var4 4.6903 1 0.030

Var5 0.6472 1 0.421

Var6 1.2257 1 0.268

Var7 4.9311 1 0.026

Var8 0.3684 1 0.544

Var9 2.0905 1 0.148

Var10 0.0319 1 0.858

Var11 4.0771 1 0.043

GLOBAL 14.2625 11 0.219

In the study, Variable 1 and 2 are the ones I'm actually interested in, whereas the others are only controls. As you can see, the PH assumptions seems to hold for these two covariates, but most prominently not for Var3. Can I still interpret my findings on Var1 and 2 and talking about Var3 add, that this is displays the average effect over the study-time? Or will my coefficients for Var1 and 2 be biased/incorrect?

Thanks in advance!

I fitted a Cox PH Model and upon examination of the Schoenfeld residuals it became apparent that the proportionality assumption was violated by several variables. Following Box-Steffensmeier & Jones (2004) I included interactions with time for these covariates. However I'm not sure how to interpret this. So its obvious that this indicates time-dependency of the covariate in the sense that the effect inflates/deflates over some function of time, but I work with sociological data and my theory indicates no time-dependency of the effects in whatever direction (Also it would not make sense in any way). So if I get that right I should therefore consider the time-dependency to come from some kind of unobserved heterogeneity? Due to the nature of the data I can also not implement frailty or fixed effects to account for this. So how do I interpret a coefficient that increases/decreases as time progresses given that the theory does not indicate this?

I'm currently working with event history data studying how long after implementation countries abolish certain policies. Regarding the policies I also have an index on how far the countries went with their policies ranging from 0 to 100.

I wanted to control for this, in order be able to control for their point of departure. However the coefficient violates the proportionality assumption.

Can I stratify for the continuous variable of that index? I understand it so, that this would allow every country to have a different baseline hazard with respect to their point of departure. Playing around with the data this didn't produce an error.

Could anyone tell me if I can trust these results or if I have to categorize the variable first?

I want to perform a GEE analysis with repeated-measured factors and survival analysis (like Cox regression but with repeated measured variables) in SPSS. How do I proceed with this dataset? I had transformed into long form format but I can't work properly.

For instance, I have A treatment modality and I want to analyze overall and progression-free survival rates in 2 groups of patients : young and old ones. These two groups have two corresponding subgroups of patients according to disease-related factor. Is Cox-regression analysis appropriate in this case?

I have two groups including 100 patient. In first group there is one endpoint , in second group there are two endpoints. However SPSS did not give results of CI and HRs. Is there a solution to get HR and CI results.

I have noticed that when using the proc phreg in SAS and the coxph in R in the same data, the model should be different in order to get the same results. In proc phreg I model the time with the censor variable (where censor=1 means the patient withdrew from the study) in order to obtain a HR. In R I use the same model but this time censor=1 means the patient experienced the event, in order to obtain the same HR. Is there a difference in the definition of the censor variable or the models are different?

Hello dear professors and colleagues,

I have a data set where I want to applied three methods: Logistic regression , linear regression and survival models, of couse each method focalise on an information part of the data set, my questions are:

I'm doing survival analysis using cox regression model. to that end I have 2 different dataset, one for training and one for testing. once I apply cox regression on training dataset I'll have a set of coefficients each related to the corresponding feature and a C-index to report. when I want to test my model in the test dataset, should I use the same coefficients and extract the C-index or should I apply cox regression model again on the test dataset and extract the C-index??

Hi, Very new to survival analysis here. I am now trying to correlate the gene expression level with survival and prognosis for patients with lung cancer, and I want to run a cox regression analysis on it. However most of the example I've encountered so far are based on discrete covariate such as sex and I know we can analyze continuous covariate using the coxph function, but I can't see how the actual plot would look like for continuous variable? For instance, for discrete variables you would have the number of regression lines correspond to the number of discrete variables. eg. for gender you'd have two lines on the graph. But what about continuous covariate? Should we first turn the continuous covariate into discrete by assigning quantiles to them? Or else I don't know how to visualize the graph. What are the pros and cons for doing so?

Thanks!

Dear All,

Your advice is urgently needed!

I want to determine the predictors of death among women with breast cancer. My sample size is 437, and total with the event (failure)=127.

In the uni-variate Cox regression I found about 30 variables eligible (p-value<0.25) to multi-variable Cox regression.

Can I include all these many variables to the multi-variable Cox regression?

Regards,

Alem,

Cox Model has the proportional hazard and the log-linearity assumptions that a data must satisfy. What if the data fails to satisfy the assumptions?

Hi everyone, I have a question regarding the interpretation of the hazard ratio in a Cox regression analysis:

I'm analyzing in an oncological 2 factors with the influence on survival. Each of the two factors shows a significantly worse survival in the case of expression (1) than for patients where the factor is not detectable (0).

Now something happens in the cox regression that I do not understand. If I add a factor of both of them next to the usual variables (sex, age, tumor stage, lymph nodes) it is shown as an independent factor with an Exp(B) > 1 (p < 0.05). But if I insert factor A and B together into the cox model, Exp(B) is > 1 for factor A and < 1 for factor B. How can this happen? Is there a good explanation?

Thanks for the help.

#spss #cox regression #statistics #survival analysis

Hello! I’m conducting a meta-analysis of non-randomized comparative studies looking at length of hospital stay.

Some studies report hazard ratios (obtained from cox regression) while others report mean differences (obtained from linear regression).

Is there anyway these data can be pooled? e.g. by transforming these data into a common estimate?

Thanks a lot!

I have a cohort of 1948 cancer patients with overall survival data and germline genotyping that i have sub-grouped into different oncogenic molecular pathways. I wish to calculate the power to detect a SNP association at a significance threshold of 5e-8 for a GWAS using cox-proportional hazard regressions. How can i calculate this? what information do i need? and are there any simple to use packages available?

Be it logistic or survival analysis/cox regression, there is utility in determining cutoff points to categorise a continuos risk factor into various risk strata.

However there is a plethora or methods to go about defining the 'optimal' value depending on the nature of the model outcome. Generally they fall into methods that look at the model's sensitivity and specificity or methods that look to maximise the p value of significant difference between the survival curves of the resulting stratas.

Currently, i'm modeling the risk of body mass index on diagnosis of certain medical conditions through survival analysis to identify at risk individuals since the current bmi risk groupings prove inadequate to stratefy risk groups. For my purposes, i'm thinking of going with a method to maximise sensitivity with maintaining a certain threshold of specificity ,instead if going eith the standard procedure to maximise the youden index , since the cost of misclassifying in the model isn't a great deal.

The issue im facing is how to justify the specificity threshold chosen? Or is there a better method of determining the best cut off to use for catorising bmi risk groups. Also , how to determine the number of categories?

The implementation of this in R is also shrouded in difficulty and the packages i found are unclear in the innerworkings of the functions.

Anyone able to shed some light on this grey situation?

When reporting hazard ratios for Cox regression analysis, is it common to report the hazard ratio for the interaction term itself?

For example, I have a model with 3 terms:

a

b

a*b

Using hazard ratio statements in SAS 9.4, I get a hazard ratio for 1) a at the mean of b, and 2) b at the mean of a. My understanding is that these hazard ratios are hazard ratios for the main effect variable (variable a or b) while holding the interacting variable constant. Is my understanding correct?

Is it possible to get a hazard ratio for the interaction term?

Thank you

I can find some articles describing R codes for drawing Nomogram after logistic regression or cox regression. But if in my logistic regression model, penalized maximum likelihood has to be used to resolve the failure of maximum likelihood estimate to converge, is it still possible to draw a Nomogram?

If the answer is yes, how to write R codes for this condition?

Thanks for any reply.

Joan

I have generated various models with an identical core of fixed covariates with each model differing only in the study covariate (a categorical variable that I change in each model by altering the cutoff values).

i want to indirectly compare the study covariates by comparing the various models.

How do i do that? Is C-Statistic an option?

Hello dears,

In regression modelling process, somtimes we deal to make a categorization of a continuos variable ( DVs or IDVs), What are really the potential problems inherent of such transformation, on:

Thank you so much for any response and clarification

Hello, everyone

I am doing cox proportional regression with a cohort of more 10000 participants. For the purpose of my study, I need to select a subgroup of 300 people in the cohort as reference group, and select another three subgroups (500-1000 participants each) to form a hierarchical four groups together with the reference group. After that, I will do a cox regression with this new viable (four groups) as independent variable.

My question is:

Is this right? If not, how can I do to achieve the research goal?

Thank you so much,

Xiaofan

I was reading about using the multivariate cox proportional hazards model at this website: http://www.sthda.com/english/wiki/cox-proportional-hazards-model, which uses the Survival package for cox regression. The summary of a cox regression object outputs a bunch of information about the model, including a concordance index.

Is all of the data used to train the cox regression model? If so, is the concordance index found on that same training data? Does this cause overfitting?

Hi! I am currently working on a statistical analysis on a prospective cohort population. We are looking at a specific protein concentration at baseline and relate this to a certain outcome. The outcome group have been matched with twice as many control individuals from the same cohort. The statistical analysis we are using is a cox regression. We've received HRs and p-values that show a significant correlation between this protein and outcome but here comes my question: we've gotten lower HRs and better (lower) p-values after adjustment for risk factors connected to this exact outcome which to us is surprising. With unadjusted model the HR is 0.693 (0.555-0.865, p=0.001) and with adjusted model 0.55 (0.405-0.747, p=0.0001). Does anyone have a clue how this came to be? We've looked at all risk factors in both case and control groups and they are as you would except. Higher smoking prevalence in case group for example (we're looking at CVD as outcome). From what I understand, the HRs should rise a bit and p-values become less strong when you adjust for risk factors (i.e only the protein concentration is taken into consideration). I'm using SPSS and here's my syntax (with protein, variables and status/time variables changed to just y, x and z respectivley:

COXREG z /STATUS=z /METHOD=ENTER y /PRINT=CI(95) /CRITERIA=PIN(.05) POUT(.10) ITERATE(20). COXREG z /STATUS=z /CONTRAST (x)=Indicator /CONTRAST (x)=Indicator /CONTRAST (x)=Indicator /CONTRAST (x)=Indicator /CONTRAST (x)=Indicator /METHOD=ENTER y x x x x x x x x x x x x /PRINT=CI(95) /CRITERIA=PIN(.05) POUT(.10) ITERATE(20).

Grateful if somebody could help me wtih this one. Am I doing something wrong in SPSS? Am I doing things right but need to interperate it some other way? Have anyone heard about this phenomenae, i.e HRs/P-values going "the wrong way" after adjustment for risk factors ? Best regards, Filip

Cox regression analysis

HI!

My author asks me to regulate p-value!

But I don't think so.Multivariable Cox Regression already has some factors to limit.They are not multiple hypothesis tests.

It's first time for me to use this web.Through reading some question and answer，people are so sweet in there!!!!