search

Thie1e / cutpointr

Optimal cutpoints in R: determining and validating optimal cutpoints in binary classification
https://cran.r-project.org/package=cutpointr
86 stars 13 forks source link

Creating a composite biomarker score using regression coefficients #58

Closed LauraL231 closed 1 year ago

LauraL231 commented 1 year ago

Hello,

I wanted to create a composite biomarker score of 4 markers that I can then enter into further predictive analysis. I was planning on doing this by extracting the β coefficients from the multivariable logistic regression with all (standardized) biomarkers and then multiply those with the (standardized) biomarker levels to create a composite. I just wondered whether this method was ok and how I can go about this using R?

Thanks, Laura

Thie1e commented 1 year ago

Hi Laura,

I guess you could surely build a "standard" logistic regression model the way you described it using glm, for example.

But it seems to me this question does not directly relate to cutpointr. Generally, if you calculate cutpoints for your markers, just make sure not to feed dichotomized marker values into your regression model.

You can estimate a cutpoint on the resulting probabilities from the logistic regression model, if you are interested in diagnostic accuracy.

I would suggest to post a question to stats.stackexchange.com if you have a purely methodological / statistical question or to StackOverflow if you can't get the R code to work (questions directly related to cutpointr can of course still go here).

Best, Christian

LauraL231 commented 1 year ago

Thanks very much for the information, that's really helpful to know. Thanks, Laura