Skip to main content
16 votes

How to explain Hazard Ratio in layperson's terms

A simple explanation of the hazard ratio is: At any given instant, people in group A are XXX as likely to have the event as people in group B. "Over the whole of the study period" is not ...
Peter Flom's user avatar
  • 124k
12 votes

How to explain Hazard Ratio in layperson's terms

We need to start with a premise. Suppose that either the HR is constant (so that what’s below applies to any t) or that we are speaking at the hazard ratio at a specific time t if the HR is not ...
Frank Harrell's user avatar
9 votes
Accepted

Diagnosing an unexpected pattern in a survival curve plot

The Kaplan Meier estimate is based on how many of the subjects that are still at risk have an event. If at the end only 2 are at risk, then one having an event means the survival curve goes down to 50%...
Björn's user avatar
  • 33.2k
7 votes
Accepted

Joint Models for Longitudinal and Time-to-Event Data vs Survival analysis with a time dependent covariate

Even though a longitudinal marker may be endogenous, I think that modeling it as a time-dependent covariate in a Cox model may be meaningful. Though without a causal interpretation, estimating how a ...
Frank Harrell's user avatar
6 votes

Cox model - unsure of time unit of analysis

Cox survival regressions don't directly model time at all. They just use the rank-ordering in time of event times, so you can use any time scale that makes sense. You can get predictions of survival ...
EdM's user avatar
  • 94.4k
5 votes

Why are my Hazard Ratio coefficients so large or small in Coxph regression?

Please also see my comment. But, assuming your data are correct: Your interpretation is incorrect. Cox PH is not about the likelihood of leaving, it is about the time to leaving. The hazard ratio is ...
Peter Flom's user avatar
  • 124k
5 votes

Cox model - unsure of time unit of analysis

I suppose that that tests are performed approximately every six months (rather than exactly every 180 days), so using months or even 6-month units as a measure of time would be appropriate. I don't ...
Roger V.'s user avatar
  • 4,421
3 votes

Joint Models for Longitudinal and Time-to-Event Data vs Survival analysis with a time dependent covariate

I'm not an expert (actually, I just started reading recently about it for my own research), but from my point of view, joint modelling can be used to model multiple longitudinal outcomes or a ...
Mathemagician777's user avatar
3 votes

Repeated observations for survival analysis

It's fairly common in this sort of scenario in medical data to analyse only the first event for each person, since they are more comparable (and everyone is guaranteed to be alive at the start of the ...
Thomas Lumley's user avatar
3 votes

How to handle time to event data when there is a competing risk with very low incidence rate

There's no problem in using a competing risks model here. If you only have a small number of death events then the coefficient estimates with respect to death will be relatively imprecise, but that ...
EdM's user avatar
  • 94.4k
3 votes

Log rank test for a time-dependent variable?

The abstract of the paper you cite notes: The Cox regression model is easily extended to the case of time-varying covariates; however, there is no clear approach for similarly extending the standard ...
EdM's user avatar
  • 94.4k
2 votes
Accepted

Subgroup analyses after propensity score matching

For moderation analysis, you need to ensure you have achieved balance within each subgroup of the moderator. The most straightforward way to do so is to match separately within each level of the ...
Noah's user avatar
  • 34.4k
2 votes

Subgroup analyses after propensity score matching

If you do the analysis separately for men and women then you cannot get statistics about the interaction -- no p values or confidence intervals or standard errors. On the other hand, the output may be ...
Peter Flom's user avatar
  • 124k
2 votes

How to compare 2 multiply imputed nested Cox proportional hazards models?

Newer likelihood ratio $\chi^2$ tests in the context of multiple imputation has been added to the R rms package. This method is based on stacking all the imputed ...
Frank Harrell's user avatar
2 votes
Accepted

Sample size and number of covariates in Cox regression

This has been solved by Richard Riley et al: https://pubmed.ncbi.nlm.nih.gov/30357870 and there is an R package to help with the calculations.
Frank Harrell's user avatar
2 votes
Accepted

Time-to-event analysis with left-truncation and right-censoring depending on the exposure

Both scenarios are related to left truncation: individuals in a study who provide no information about events that occur prior to some elapsed time since time = 0 ...
EdM's user avatar
  • 94.4k
2 votes

How to handle zero-inflated time in Cox proportional hazards model with categorical covariates?

The usual assumption in survival analysis is that there is 100% survival at time = 0. As Peter Flom notes, however, the times themselves aren't used for a Cox model....
EdM's user avatar
  • 94.4k
1 vote

Medication use as a time dependent covariate in cox regression?

Yes, that's definitely a reasonable way to do this, and a fairly common one. Plotting depends on your sample size, but it's useful to do some graphs showing individual trajectories on and off ...
Thomas Lumley's user avatar
1 vote
Accepted

Type of censoring in discrete time survival

Yes, you can consider a discrete-time model here. Interval censoring means that you know an event happened within some time interval, but you don’t know exactly when. That’s how you treat event times ...
EdM's user avatar
  • 94.4k
1 vote
Accepted

Changes in Schoenfeld Residuals Dependent on Follow-Up Time

First, before you worry about violating proportional hazards (PH), worry about overfitting. Having 113 or 219 events might seem like a lot, but your model needs to fit 26 regression coefficients (...
EdM's user avatar
  • 94.4k
1 vote
Accepted

Sample size for Survival Analysis (Cox regression) with discrete time

As Demetri Pananos implies in a comment, a discrete-time survival model (which seems appropriate in your case) is based on binomial regressions. You set up a separate data row for each individual ...
EdM's user avatar
  • 94.4k
1 vote

Sample size and number of covariates in Cox regression

As far as I can tell, the answer is "no". In SAS, there is the COXREG option in the POWER procedure, but it only handles a single covariate. In R, there is the powerSurvEpi package, but this ...
Peter Flom's user avatar
  • 124k
1 vote

How to compare 2 multiply imputed nested Cox proportional hazards models?

From https://bookdown.org/rwnahhas/RMPH/mi-cox.html : "Neither mi.anova() nor D3() works for Cox regression models, but D1() does." From https://github.com/amices/mice/issues/246#...
Jack Sprat's user avatar
1 vote

How to interpret interaction effects in a cox regression?

The interaction between injury and diet is telling you that the relationship between injury and the hazard ratio is different for different diets. Similarly, the relationship between diet and the HR ...
Peter Flom's user avatar
  • 124k

Only top scored, non community-wiki answers of a minimum length are eligible